Blame - jni/libpcre/doc/html/pcreapi.html - jami-client-android

blob: 3cbb6be0f6ef5f22841811b5913e363a32380d35 [file] [log] [blame]

Tristan Matthews	0461646	2013-11-14 16:09:34 -0500	[diff] [blame^]	1	<html>
				2	<head>
				3	<title>pcreapi specification</title>
				4	</head>
				5	<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
				6	<h1>pcreapi man page</h1>
				7	<p>
				8	Return to the <a href="index.html">PCRE index page</a>.
				9	</p>
				10	<p>
				11	This page is part of the PCRE HTML documentation. It was generated automatically
				12	from the original man page. If there is any nonsense in it, please consult the
				13	man page, in case the conversion went wrong.
				14	<br>
				15	<ul>
				16	<li><a name="TOC1" href="#SEC1">PCRE NATIVE API BASIC FUNCTIONS</a>
				17	<li><a name="TOC2" href="#SEC2">PCRE NATIVE API AUXILIARY FUNCTIONS</a>
				18	<li><a name="TOC3" href="#SEC3">PCRE NATIVE API INDIRECTED FUNCTIONS</a>
				19	<li><a name="TOC4" href="#SEC4">PCRE API OVERVIEW</a>
				20	<li><a name="TOC5" href="#SEC5">NEWLINES</a>
				21	<li><a name="TOC6" href="#SEC6">MULTITHREADING</a>
				22	<li><a name="TOC7" href="#SEC7">SAVING PRECOMPILED PATTERNS FOR LATER USE</a>
				23	<li><a name="TOC8" href="#SEC8">CHECKING BUILD-TIME OPTIONS</a>
				24	<li><a name="TOC9" href="#SEC9">COMPILING A PATTERN</a>
				25	<li><a name="TOC10" href="#SEC10">COMPILATION ERROR CODES</a>
				26	<li><a name="TOC11" href="#SEC11">STUDYING A PATTERN</a>
				27	<li><a name="TOC12" href="#SEC12">LOCALE SUPPORT</a>
				28	<li><a name="TOC13" href="#SEC13">INFORMATION ABOUT A PATTERN</a>
				29	<li><a name="TOC14" href="#SEC14">OBSOLETE INFO FUNCTION</a>
				30	<li><a name="TOC15" href="#SEC15">REFERENCE COUNTS</a>
				31	<li><a name="TOC16" href="#SEC16">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
				32	<li><a name="TOC17" href="#SEC17">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
				33	<li><a name="TOC18" href="#SEC18">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
				34	<li><a name="TOC19" href="#SEC19">DUPLICATE SUBPATTERN NAMES</a>
				35	<li><a name="TOC20" href="#SEC20">FINDING ALL POSSIBLE MATCHES</a>
				36	<li><a name="TOC21" href="#SEC21">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
				37	<li><a name="TOC22" href="#SEC22">SEE ALSO</a>
				38	<li><a name="TOC23" href="#SEC23">AUTHOR</a>
				39	<li><a name="TOC24" href="#SEC24">REVISION</a>
				40	</ul>
				41	<br><a name="SEC1" href="#TOC1">PCRE NATIVE API BASIC FUNCTIONS</a><br>
				42	<P>
				43	<b>#include <pcre.h></b>
				44	</P>
				45	<P>
				46	<b>pcre pcre_compile(const char <i>pattern</i>, int <i>options</i>,</b>
				47	<b>const char *<i>errptr</i>, int <i>erroffset</i>,</b>
				48	<b>const unsigned char *<i>tableptr</i>);</b>
				49	</P>
				50	<P>
				51	<b>pcre pcre_compile2(const char <i>pattern</i>, int <i>options</i>,</b>
				52	<b>int *<i>errorcodeptr</i>,</b>
				53	<b>const char *<i>errptr</i>, int <i>erroffset</i>,</b>
				54	<b>const unsigned char *<i>tableptr</i>);</b>
				55	</P>
				56	<P>
				57	<b>pcre_extra pcre_study(const pcre <i>code</i>, int <i>options</i>,</b>
				58	<b>const char **<i>errptr</i>);</b>
				59	</P>
				60	<P>
				61	<b>void pcre_free_study(pcre_extra *<i>extra</i>);</b>
				62	</P>
				63	<P>
				64	<b>int pcre_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	</P>
				68	<br><a name="SEC2" href="#TOC1">PCRE NATIVE API AUXILIARY FUNCTIONS</a><br>
				69	<P>
				70	<b>pcre_jit_stack *pcre_jit_stack_alloc(int <i>startsize</i>, int <i>maxsize</i>);</b>
				71	</P>
				72	<P>
				73	<b>void pcre_jit_stack_free(pcre_jit_stack *<i>stack</i>);</b>
				74	</P>
				75	<P>
				76	<b>void pcre_assign_jit_stack(pcre_extra *<i>extra</i>,</b>
				77	<b>pcre_jit_callback <i>callback</i>, void *<i>data</i>);</b>
				78	</P>
				79	<P>
				80	<b>int pcre_dfa_exec(const pcre <i>code</i>, const pcre_extra <i>extra</i>,</b>
				81	<b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
				82	<b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
				83	<b>int *<i>workspace</i>, int <i>wscount</i>);</b>
				84	</P>
				85	<P>
				86	<b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
				87	<b>const char <i>subject</i>, int <i>ovector</i>,</b>
				88	<b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
				89	<b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
				90	</P>
				91	<P>
				92	<b>int pcre_copy_substring(const char <i>subject</i>, int <i>ovector</i>,</b>
				93	<b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
				94	<b>int <i>buffersize</i>);</b>
				95	</P>
				96	<P>
				97	<b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
				98	<b>const char <i>subject</i>, int <i>ovector</i>,</b>
				99	<b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
				100	<b>const char **<i>stringptr</i>);</b>
				101	</P>
				102	<P>
				103	<b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
				104	<b>const char *<i>name</i>);</b>
				105	</P>
				106	<P>
				107	<b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
				108	<b>const char <i>name</i>, char <i>first</i>, char *<i>last</i>);</b>
				109	</P>
				110	<P>
				111	<b>int pcre_get_substring(const char <i>subject</i>, int <i>ovector</i>,</b>
				112	<b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
				113	<b>const char **<i>stringptr</i>);</b>
				114	</P>
				115	<P>
				116	<b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
				117	<b>int <i>ovector</i>, int <i>stringcount</i>, const char **<i>listptr</i>);</b>
				118	</P>
				119	<P>
				120	<b>void pcre_free_substring(const char *<i>stringptr</i>);</b>
				121	</P>
				122	<P>
				123	<b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b>
				124	</P>
				125	<P>
				126	<b>const unsigned char *pcre_maketables(void);</b>
				127	</P>
				128	<P>
				129	<b>int pcre_fullinfo(const pcre <i>code</i>, const pcre_extra <i>extra</i>,</b>
				130	<b>int <i>what</i>, void *<i>where</i>);</b>
				131	</P>
				132	<P>
				133	<b>int pcre_info(const pcre <i>code</i>, int <i>optptr</i>, int</b>
				134	<b>*<i>firstcharptr</i>);</b>
				135	</P>
				136	<P>
				137	<b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
				138	</P>
				139	<P>
				140	<b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
				141	</P>
				142	<P>
				143	<b>char *pcre_version(void);</b>
				144	</P>
				145	<br><a name="SEC3" href="#TOC1">PCRE NATIVE API INDIRECTED FUNCTIONS</a><br>
				146	<P>
				147	<b>void (pcre_malloc)(size_t);</b>
				148	</P>
				149	<P>
				150	<b>void (pcre_free)(void );</b>
				151	</P>
				152	<P>
				153	<b>void (pcre_stack_malloc)(size_t);</b>
				154	</P>
				155	<P>
				156	<b>void (pcre_stack_free)(void );</b>
				157	</P>
				158	<P>
				159	<b>int (pcre_callout)(pcre_callout_block );</b>
				160	</P>
				161	<br><a name="SEC4" href="#TOC1">PCRE API OVERVIEW</a><br>
				162	<P>
				163	PCRE has its own native API, which is described in this document. There are
				164	also some wrapper functions that correspond to the POSIX regular expression
				165	API, but they do not give access to all the functionality. They are described
				166	in the
				167	<a href="pcreposix.html"><b>pcreposix</b></a>
				168	documentation. Both of these APIs define a set of C function calls. A C++
				169	wrapper is also distributed with PCRE. It is documented in the
				170	<a href="pcrecpp.html"><b>pcrecpp</b></a>
				171	page.
				172	</P>
				173	<P>
				174	The native API C function prototypes are defined in the header file
				175	<b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
				176	It can normally be accessed by adding <b>-lpcre</b> to the command for linking
				177	an application that uses PCRE. The header file defines the macros PCRE_MAJOR
				178	and PCRE_MINOR to contain the major and minor release numbers for the library.
				179	Applications can use these to include support for different releases of PCRE.
				180	</P>
				181	<P>
				182	In a Windows environment, if you want to statically link an application program
				183	against a non-dll <b>pcre.a</b> file, you must define PCRE_STATIC before
				184	including <b>pcre.h</b> or <b>pcrecpp.h</b>, because otherwise the
				185	<b>pcre_malloc()</b> and <b>pcre_free()</b> exported functions will be declared
				186	<b>__declspec(dllimport)</b>, with unwanted results.
				187	</P>
				188	<P>
				189	The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
				190	and <b>pcre_exec()</b> are used for compiling and matching regular expressions
				191	in a Perl-compatible manner. A sample program that demonstrates the simplest
				192	way of using them is provided in the file called <i>pcredemo.c</i> in the PCRE
				193	source distribution. A listing of this program is given in the
				194	<a href="pcredemo.html"><b>pcredemo</b></a>
				195	documentation, and the
				196	<a href="pcresample.html"><b>pcresample</b></a>
				197	documentation describes how to compile and run it.
				198	</P>
				199	<P>
				200	Just-in-time compiler support is an optional feature of PCRE that can be built
				201	in appropriate hardware environments. It greatly speeds up the matching
				202	performance of many patterns. Simple programs can easily request that it be
				203	used if available, by setting an option that is ignored when it is not
				204	relevant. More complicated programs might need to make use of the functions
				205	<b>pcre_jit_stack_alloc()</b>, <b>pcre_jit_stack_free()</b>, and
				206	<b>pcre_assign_jit_stack()</b> in order to control the JIT code's memory usage.
				207	These functions are discussed in the
				208	<a href="pcrejit.html"><b>pcrejit</b></a>
				209	documentation.
				210	</P>
				211	<P>
				212	A second matching function, <b>pcre_dfa_exec()</b>, which is not
				213	Perl-compatible, is also provided. This uses a different algorithm for the
				214	matching. The alternative algorithm finds all possible matches (at a given
				215	point in the subject), and scans the subject just once (unless there are
				216	lookbehind assertions). However, this algorithm does not return captured
				217	substrings. A description of the two matching algorithms and their advantages
				218	and disadvantages is given in the
				219	<a href="pcrematching.html"><b>pcrematching</b></a>
				220	documentation.
				221	</P>
				222	<P>
				223	In addition to the main compiling and matching functions, there are convenience
				224	functions for extracting captured substrings from a subject string that is
				225	matched by <b>pcre_exec()</b>. They are:
				226	<pre>
				227	<b>pcre_copy_substring()</b>
				228	<b>pcre_copy_named_substring()</b>
				229	<b>pcre_get_substring()</b>
				230	<b>pcre_get_named_substring()</b>
				231	<b>pcre_get_substring_list()</b>
				232	<b>pcre_get_stringnumber()</b>
				233	<b>pcre_get_stringtable_entries()</b>
				234	</pre>
				235	<b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
				236	provided, to free the memory used for extracted strings.
				237	</P>
				238	<P>
				239	The function <b>pcre_maketables()</b> is used to build a set of character tables
				240	in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
				241	or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
				242	specialist use. Most commonly, no special tables are passed, in which case
				243	internal tables that are generated when PCRE is built are used.
				244	</P>
				245	<P>
				246	The function <b>pcre_fullinfo()</b> is used to find out information about a
				247	compiled pattern; <b>pcre_info()</b> is an obsolete version that returns only
				248	some of the available information, but is retained for backwards compatibility.
				249	The function <b>pcre_version()</b> returns a pointer to a string containing the
				250	version of PCRE and its date of release.
				251	</P>
				252	<P>
				253	The function <b>pcre_refcount()</b> maintains a reference count in a data block
				254	containing a compiled pattern. This is provided for the benefit of
				255	object-oriented applications.
				256	</P>
				257	<P>
				258	The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
				259	the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
				260	respectively. PCRE calls the memory management functions via these variables,
				261	so a calling program can replace them if it wishes to intercept the calls. This
				262	should be done before calling any PCRE functions.
				263	</P>
				264	<P>
				265	The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
				266	indirections to memory management functions. These special functions are used
				267	only when PCRE is compiled to use the heap for remembering data, instead of
				268	recursive function calls, when running the <b>pcre_exec()</b> function. See the
				269	<a href="pcrebuild.html"><b>pcrebuild</b></a>
				270	documentation for details of how to do this. It is a non-standard way of
				271	building PCRE, for use in environments that have limited stacks. Because of the
				272	greater use of memory management, it runs more slowly. Separate functions are
				273	provided so that special-purpose external code can be used for this case. When
				274	used, these functions are always called in a stack-like manner (last obtained,
				275	first freed), and always for memory blocks of the same size. There is a
				276	discussion about PCRE's stack usage in the
				277	<a href="pcrestack.html"><b>pcrestack</b></a>
				278	documentation.
				279	</P>
				280	<P>
				281	The global variable <b>pcre_callout</b> initially contains NULL. It can be set
				282	by the caller to a "callout" function, which PCRE will then call at specified
				283	points during a matching operation. Details are given in the
				284	<a href="pcrecallout.html"><b>pcrecallout</b></a>
				285	documentation.
				286	<a name="newlines"></a></P>
				287	<br><a name="SEC5" href="#TOC1">NEWLINES</a><br>
				288	<P>
				289	PCRE supports five different conventions for indicating line breaks in
				290	strings: a single CR (carriage return) character, a single LF (linefeed)
				291	character, the two-character sequence CRLF, any of the three preceding, or any
				292	Unicode newline sequence. The Unicode newline sequences are the three just
				293	mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
				294	U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
				295	(paragraph separator, U+2029).
				296	</P>
				297	<P>
				298	Each of the first three conventions is used by at least one operating system as
				299	its standard newline sequence. When PCRE is built, a default can be specified.
				300	The default default is LF, which is the Unix standard. When PCRE is run, the
				301	default can be overridden, either when a pattern is compiled, or when it is
				302	matched.
				303	</P>
				304	<P>
				305	At compile time, the newline convention can be specified by the <i>options</i>
				306	argument of <b>pcre_compile()</b>, or it can be specified by special text at the
				307	start of the pattern itself; this overrides any other settings. See the
				308	<a href="pcrepattern.html"><b>pcrepattern</b></a>
				309	page for details of the special character sequences.
				310	</P>
				311	<P>
				312	In the PCRE documentation the word "newline" is used to mean "the character or
				313	pair of characters that indicate a line break". The choice of newline
				314	convention affects the handling of the dot, circumflex, and dollar
				315	metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
				316	recognized line ending sequence, the match position advancement for a
				317	non-anchored pattern. There is more detail about this in the
				318	<a href="#execoptions">section on <b>pcre_exec()</b> options</a>
				319	below.
				320	</P>
				321	<P>
				322	The choice of newline convention does not affect the interpretation of
				323	the \n or \r escape sequences, nor does it affect what \R matches, which is
				324	controlled in a similar way, but by separate options.
				325	</P>
				326	<br><a name="SEC6" href="#TOC1">MULTITHREADING</a><br>
				327	<P>
				328	The PCRE functions can be used in multi-threading applications, with the
				329	proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
				330	<b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
				331	callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
				332	</P>
				333	<P>
				334	The compiled form of a regular expression is not altered during matching, so
				335	the same compiled pattern can safely be used by several threads at once.
				336	</P>
				337	<P>
				338	If the just-in-time optimization feature is being used, it needs separate
				339	memory stack areas for each thread. See the
				340	<a href="pcrejit.html"><b>pcrejit</b></a>
				341	documentation for more details.
				342	</P>
				343	<br><a name="SEC7" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
				344	<P>
				345	The compiled form of a regular expression can be saved and re-used at a later
				346	time, possibly by a different program, and even on a host other than the one on
				347	which it was compiled. Details are given in the
				348	<a href="pcreprecompile.html"><b>pcreprecompile</b></a>
				349	documentation. However, compiling a regular expression with one version of PCRE
				350	for use with a different version is not guaranteed to work and may cause
				351	crashes.
				352	</P>
				353	<br><a name="SEC8" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
				354	<P>
				355	<b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
				356	</P>
				357	<P>
				358	The function <b>pcre_config()</b> makes it possible for a PCRE client to
				359	discover which optional features have been compiled into the PCRE library. The
				360	<a href="pcrebuild.html"><b>pcrebuild</b></a>
				361	documentation has more details about these optional features.
				362	</P>
				363	<P>
				364	The first argument for <b>pcre_config()</b> is an integer, specifying which
				365	information is required; the second argument is a pointer to a variable into
				366	which the information is placed. The following information is available:
				367	<pre>
				368	PCRE_CONFIG_UTF8
				369	</pre>
				370	The output is an integer that is set to one if UTF-8 support is available;
				371	otherwise it is set to zero.
				372	<pre>
				373	PCRE_CONFIG_UNICODE_PROPERTIES
				374	</pre>
				375	The output is an integer that is set to one if support for Unicode character
				376	properties is available; otherwise it is set to zero.
				377	<pre>
				378	PCRE_CONFIG_JIT
				379	</pre>
				380	The output is an integer that is set to one if support for just-in-time
				381	compiling is available; otherwise it is set to zero.
				382	<pre>
				383	PCRE_CONFIG_NEWLINE
				384	</pre>
				385	The output is an integer whose value specifies the default character sequence
				386	that is recognized as meaning "newline". The four values that are supported
				387	are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
				388	Though they are derived from ASCII, the same values are returned in EBCDIC
				389	environments. The default should normally correspond to the standard sequence
				390	for your operating system.
				391	<pre>
				392	PCRE_CONFIG_BSR
				393	</pre>
				394	The output is an integer whose value indicates what character sequences the \R
				395	escape sequence matches by default. A value of 0 means that \R matches any
				396	Unicode line ending sequence; a value of 1 means that \R matches only CR, LF,
				397	or CRLF. The default can be overridden when a pattern is compiled or matched.
				398	<pre>
				399	PCRE_CONFIG_LINK_SIZE
				400	</pre>
				401	The output is an integer that contains the number of bytes used for internal
				402	linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
				403	allow larger regular expressions to be compiled, at the expense of slower
				404	matching. The default value of 2 is sufficient for all but the most massive
				405	patterns, since it allows the compiled pattern to be up to 64K in size.
				406	<pre>
				407	PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
				408	</pre>
				409	The output is an integer that contains the threshold above which the POSIX
				410	interface uses <b>malloc()</b> for output vectors. Further details are given in
				411	the
				412	<a href="pcreposix.html"><b>pcreposix</b></a>
				413	documentation.
				414	<pre>
				415	PCRE_CONFIG_MATCH_LIMIT
				416	</pre>
				417	The output is a long integer that gives the default limit for the number of
				418	internal matching function calls in a <b>pcre_exec()</b> execution. Further
				419	details are given with <b>pcre_exec()</b> below.
				420	<pre>
				421	PCRE_CONFIG_MATCH_LIMIT_RECURSION
				422	</pre>
				423	The output is a long integer that gives the default limit for the depth of
				424	recursion when calling the internal matching function in a <b>pcre_exec()</b>
				425	execution. Further details are given with <b>pcre_exec()</b> below.
				426	<pre>
				427	PCRE_CONFIG_STACKRECURSE
				428	</pre>
				429	The output is an integer that is set to one if internal recursion when running
				430	<b>pcre_exec()</b> is implemented by recursive function calls that use the stack
				431	to remember their state. This is the usual way that PCRE is compiled. The
				432	output is zero if PCRE was compiled to use blocks of data on the heap instead
				433	of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
				434	<b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
				435	avoiding the use of the stack.
				436	</P>
				437	<br><a name="SEC9" href="#TOC1">COMPILING A PATTERN</a><br>
				438	<P>
				439	<b>pcre pcre_compile(const char <i>pattern</i>, int <i>options</i>,</b>
				440	<b>const char *<i>errptr</i>, int <i>erroffset</i>,</b>
				441	<b>const unsigned char *<i>tableptr</i>);</b>
				442	<b>pcre pcre_compile2(const char <i>pattern</i>, int <i>options</i>,</b>
				443	<b>int *<i>errorcodeptr</i>,</b>
				444	<b>const char *<i>errptr</i>, int <i>erroffset</i>,</b>
				445	<b>const unsigned char *<i>tableptr</i>);</b>
				446	</P>
				447	<P>
				448	Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
				449	called to compile a pattern into an internal form. The only difference between
				450	the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
				451	<i>errorcodeptr</i>, via which a numerical error code can be returned. To avoid
				452	too much repetition, we refer just to <b>pcre_compile()</b> below, but the
				453	information applies equally to <b>pcre_compile2()</b>.
				454	</P>
				455	<P>
				456	The pattern is a C string terminated by a binary zero, and is passed in the
				457	<i>pattern</i> argument. A pointer to a single block of memory that is obtained
				458	via <b>pcre_malloc</b> is returned. This contains the compiled code and related
				459	data. The <b>pcre</b> type is defined for the returned block; this is a typedef
				460	for a structure whose contents are not externally defined. It is up to the
				461	caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
				462	</P>
				463	<P>
				464	Although the compiled code of a PCRE regex is relocatable, that is, it does not
				465	depend on memory location, the complete <b>pcre</b> data block is not
				466	fully relocatable, because it may contain a copy of the <i>tableptr</i>
				467	argument, which is an address (see below).
				468	</P>
				469	<P>
				470	The <i>options</i> argument contains various bit settings that affect the
				471	compilation. It should be zero if no options are required. The available
				472	options are described below. Some of them (in particular, those that are
				473	compatible with Perl, but some others as well) can also be set and unset from
				474	within the pattern (see the detailed description in the
				475	<a href="pcrepattern.html"><b>pcrepattern</b></a>
				476	documentation). For those options that can be different in different parts of
				477	the pattern, the contents of the <i>options</i> argument specifies their
				478	settings at the start of compilation and execution. The PCRE_ANCHORED,
				479	PCRE_BSR_<i>xxx</i>, PCRE_NEWLINE_<i>xxx</i>, PCRE_NO_UTF8_CHECK, and
				480	PCRE_NO_START_OPT options can be set at the time of matching as well as at
				481	compile time.
				482	</P>
				483	<P>
				484	If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
				485	Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
				486	NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
				487	error message. This is a static string that is part of the library. You must
				488	not try to free it. Normally, the offset from the start of the pattern to the
				489	byte that was being processed when the error was discovered is placed in the
				490	variable pointed to by <i>erroffset</i>, which must not be NULL (if it is, an
				491	immediate error is given). However, for an invalid UTF-8 string, the offset is
				492	that of the first byte of the failing character. Also, some errors are not
				493	detected until checks are carried out when the whole pattern has been scanned;
				494	in these cases the offset passed back is the length of the pattern.
				495	</P>
				496	<P>
				497	Note that the offset is in bytes, not characters, even in UTF-8 mode. It may
				498	sometimes point into the middle of a UTF-8 character.
				499	</P>
				500	<P>
				501	If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
				502	<i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
				503	returned via this argument in the event of an error. This is in addition to the
				504	textual error message. Error codes and messages are listed below.
				505	</P>
				506	<P>
				507	If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
				508	character tables that are built when PCRE is compiled, using the default C
				509	locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
				510	call to <b>pcre_maketables()</b>. This value is stored with the compiled
				511	pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
				512	passed to it. For more discussion, see the section on locale support below.
				513	</P>
				514	<P>
				515	This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
				516	<pre>
				517	pcre *re;
				518	const char *error;
				519	int erroffset;
				520	re = pcre_compile(
				521	"^A.Z", / the pattern */
				522	0, /* default options */
				523	&error, /* for error message */
				524	&erroffset, /* for error offset */
				525	NULL); /* use default character tables */
				526	</pre>
				527	The following names for option bits are defined in the <b>pcre.h</b> header
				528	file:
				529	<pre>
				530	PCRE_ANCHORED
				531	</pre>
				532	If this bit is set, the pattern is forced to be "anchored", that is, it is
				533	constrained to match only at the first matching point in the string that is
				534	being searched (the "subject string"). This effect can also be achieved by
				535	appropriate constructs in the pattern itself, which is the only way to do it in
				536	Perl.
				537	<pre>
				538	PCRE_AUTO_CALLOUT
				539	</pre>
				540	If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
				541	all with number 255, before each pattern item. For discussion of the callout
				542	facility, see the
				543	<a href="pcrecallout.html"><b>pcrecallout</b></a>
				544	documentation.
				545	<pre>
				546	PCRE_BSR_ANYCRLF
				547	PCRE_BSR_UNICODE
				548	</pre>
				549	These options (which are mutually exclusive) control what the \R escape
				550	sequence matches. The choice is either to match only CR, LF, or CRLF, or to
				551	match any Unicode newline sequence. The default is specified when PCRE is
				552	built. It can be overridden from within the pattern, or by setting an option
				553	when a compiled pattern is matched.
				554	<pre>
				555	PCRE_CASELESS
				556	</pre>
				557	If this bit is set, letters in the pattern match both upper and lower case
				558	letters. It is equivalent to Perl's /i option, and it can be changed within a
				559	pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
				560	concept of case for characters whose values are less than 128, so caseless
				561	matching is always possible. For characters with higher values, the concept of
				562	case is supported if PCRE is compiled with Unicode property support, but not
				563	otherwise. If you want to use caseless matching for characters 128 and above,
				564	you must ensure that PCRE is compiled with Unicode property support as well as
				565	with UTF-8 support.
				566	<pre>
				567	PCRE_DOLLAR_ENDONLY
				568	</pre>
				569	If this bit is set, a dollar metacharacter in the pattern matches only at the
				570	end of the subject string. Without this option, a dollar also matches
				571	immediately before a newline at the end of the string (but not before any other
				572	newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
				573	There is no equivalent to this option in Perl, and no way to set it within a
				574	pattern.
				575	<pre>
				576	PCRE_DOTALL
				577	</pre>
				578	If this bit is set, a dot metacharacter in the pattern matches a character of
				579	any value, including one that indicates a newline. However, it only ever
				580	matches one character, even if newlines are coded as CRLF. Without this option,
				581	a dot does not match when the current position is at a newline. This option is
				582	equivalent to Perl's /s option, and it can be changed within a pattern by a
				583	(?s) option setting. A negative class such as [^a] always matches newline
				584	characters, independent of the setting of this option.
				585	<pre>
				586	PCRE_DUPNAMES
				587	</pre>
				588	If this bit is set, names used to identify capturing subpatterns need not be
				589	unique. This can be helpful for certain types of pattern when it is known that
				590	only one instance of the named subpattern can ever be matched. There are more
				591	details of named subpatterns below; see also the
				592	<a href="pcrepattern.html"><b>pcrepattern</b></a>
				593	documentation.
				594	<pre>
				595	PCRE_EXTENDED
				596	</pre>
				597	If this bit is set, whitespace data characters in the pattern are totally
				598	ignored except when escaped or inside a character class. Whitespace does not
				599	include the VT character (code 11). In addition, characters between an
				600	unescaped # outside a character class and the next newline, inclusive, are also
				601	ignored. This is equivalent to Perl's /x option, and it can be changed within a
				602	pattern by a (?x) option setting.
				603	</P>
				604	<P>
				605	Which characters are interpreted as newlines is controlled by the options
				606	passed to <b>pcre_compile()</b> or by a special sequence at the start of the
				607	pattern, as described in the section entitled
				608	<a href="pcrepattern.html#newlines">"Newline conventions"</a>
				609	in the <b>pcrepattern</b> documentation. Note that the end of this type of
				610	comment is a literal newline sequence in the pattern; escape sequences that
				611	happen to represent a newline do not count.
				612	</P>
				613	<P>
				614	This option makes it possible to include comments inside complicated patterns.
				615	Note, however, that this applies only to data characters. Whitespace characters
				616	may never appear within special character sequences in a pattern, for example
				617	within the sequence (?( that introduces a conditional subpattern.
				618	<pre>
				619	PCRE_EXTRA
				620	</pre>
				621	This option was invented in order to turn on additional functionality of PCRE
				622	that is incompatible with Perl, but it is currently of very little use. When
				623	set, any backslash in a pattern that is followed by a letter that has no
				624	special meaning causes an error, thus reserving these combinations for future
				625	expansion. By default, as in Perl, a backslash followed by a letter with no
				626	special meaning is treated as a literal. (Perl can, however, be persuaded to
				627	give an error for this, by running it with the -w option.) There are at present
				628	no other features controlled by this option. It can also be set by a (?X)
				629	option setting within a pattern.
				630	<pre>
				631	PCRE_FIRSTLINE
				632	</pre>
				633	If this option is set, an unanchored pattern is required to match before or at
				634	the first newline in the subject string, though the matched text may continue
				635	over the newline.
				636	<pre>
				637	PCRE_JAVASCRIPT_COMPAT
				638	</pre>
				639	If this option is set, PCRE's behaviour is changed in some ways so that it is
				640	compatible with JavaScript rather than Perl. The changes are as follows:
				641	</P>
				642	<P>
				643	(1) A lone closing square bracket in a pattern causes a compile-time error,
				644	because this is illegal in JavaScript (by default it is treated as a data
				645	character). Thus, the pattern AB]CD becomes illegal when this option is set.
				646	</P>
				647	<P>
				648	(2) At run time, a back reference to an unset subpattern group matches an empty
				649	string (by default this causes the current matching alternative to fail). A
				650	pattern such as (\1)(a) succeeds when this option is set (assuming it can find
				651	an "a" in the subject), whereas it fails by default, for Perl compatibility.
				652	</P>
				653	<P>
				654	(3) \U matches an upper case "U" character; by default \U causes a compile
				655	time error (Perl uses \U to upper case subsequent characters).
				656	</P>
				657	<P>
				658	(4) \u matches a lower case "u" character unless it is followed by four
				659	hexadecimal digits, in which case the hexadecimal number defines the code point
				660	to match. By default, \u causes a compile time error (Perl uses it to upper
				661	case the following character).
				662	</P>
				663	<P>
				664	(5) \x matches a lower case "x" character unless it is followed by two
				665	hexadecimal digits, in which case the hexadecimal number defines the code point
				666	to match. By default, as in Perl, a hexadecimal number is always expected after
				667	\x, but it may have zero, one, or two digits (so, for example, \xz matches a
				668	binary zero character followed by z).
				669	<pre>
				670	PCRE_MULTILINE
				671	</pre>
				672	By default, PCRE treats the subject string as consisting of a single line of
				673	characters (even if it actually contains newlines). The "start of line"
				674	metacharacter (^) matches only at the start of the string, while the "end of
				675	line" metacharacter ($) matches only at the end of the string, or before a
				676	terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
				677	Perl.
				678	</P>
				679	<P>
				680	When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
				681	match immediately following or immediately before internal newlines in the
				682	subject string, respectively, as well as at the very start and end. This is
				683	equivalent to Perl's /m option, and it can be changed within a pattern by a
				684	(?m) option setting. If there are no newlines in a subject string, or no
				685	occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
				686	<pre>
				687	PCRE_NEWLINE_CR
				688	PCRE_NEWLINE_LF
				689	PCRE_NEWLINE_CRLF
				690	PCRE_NEWLINE_ANYCRLF
				691	PCRE_NEWLINE_ANY
				692	</pre>
				693	These options override the default newline definition that was chosen when PCRE
				694	was built. Setting the first or the second specifies that a newline is
				695	indicated by a single character (CR or LF, respectively). Setting
				696	PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
				697	CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
				698	preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
				699	that any Unicode newline sequence should be recognized. The Unicode newline
				700	sequences are the three just mentioned, plus the single characters VT (vertical
				701	tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
				702	separator, U+2028), and PS (paragraph separator, U+2029). The last two are
				703	recognized only in UTF-8 mode.
				704	</P>
				705	<P>
				706	The newline setting in the options word uses three bits that are treated
				707	as a number, giving eight possibilities. Currently only six are used (default
				708	plus the five values above). This means that if you set more than one newline
				709	option, the combination may or may not be sensible. For example,
				710	PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
				711	other combinations may yield unused numbers and cause an error.
				712	</P>
				713	<P>
				714	The only time that a line break in a pattern is specially recognized when
				715	compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
				716	and so are ignored in this mode. Also, an unescaped # outside a character class
				717	indicates a comment that lasts until after the next line break sequence. In
				718	other circumstances, line break sequences in patterns are treated as literal
				719	data.
				720	</P>
				721	<P>
				722	The newline option that is set at compile time becomes the default that is used
				723	for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
				724	<pre>
				725	PCRE_NO_AUTO_CAPTURE
				726	</pre>
				727	If this option is set, it disables the use of numbered capturing parentheses in
				728	the pattern. Any opening parenthesis that is not followed by ? behaves as if it
				729	were followed by ?: but named parentheses can still be used for capturing (and
				730	they acquire numbers in the usual way). There is no equivalent of this option
				731	in Perl.
				732	<pre>
				733	NO_START_OPTIMIZE
				734	</pre>
				735	This is an option that acts at matching time; that is, it is really an option
				736	for <b>pcre_exec()</b> or <b>pcre_dfa_exec()</b>. If it is set at compile time,
				737	it is remembered with the compiled pattern and assumed at matching time. For
				738	details see the discussion of PCRE_NO_START_OPTIMIZE
				739	<a href="#execoptions">below.</a>
				740	<pre>
				741	PCRE_UCP
				742	</pre>
				743	This option changes the way PCRE processes \B, \b, \D, \d, \S, \s, \W,
				744	\w, and some of the POSIX character classes. By default, only ASCII characters
				745	are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
				746	classify characters. More details are given in the section on
				747	<a href="pcre.html#genericchartypes">generic character types</a>
				748	in the
				749	<a href="pcrepattern.html"><b>pcrepattern</b></a>
				750	page. If you set PCRE_UCP, matching one of the items it affects takes much
				751	longer. The option is available only if PCRE has been compiled with Unicode
				752	property support.
				753	<pre>
				754	PCRE_UNGREEDY
				755	</pre>
				756	This option inverts the "greediness" of the quantifiers so that they are not
				757	greedy by default, but become greedy if followed by "?". It is not compatible
				758	with Perl. It can also be set by a (?U) option setting within the pattern.
				759	<pre>
				760	PCRE_UTF8
				761	</pre>
				762	This option causes PCRE to regard both the pattern and the subject as strings
				763	of UTF-8 characters instead of single-byte character strings. However, it is
				764	available only when PCRE is built to include UTF-8 support. If not, the use
				765	of this option provokes an error. Details of how this option changes the
				766	behaviour of PCRE are given in the
				767	<a href="pcreunicode.html"><b>pcreunicode</b></a>
				768	page.
				769	<pre>
				770	PCRE_NO_UTF8_CHECK
				771	</pre>
				772	When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
				773	automatically checked. There is a discussion about the
				774	<a href="pcre.html#utf8strings">validity of UTF-8 strings</a>
				775	in the main
				776	<a href="pcre.html"><b>pcre</b></a>
				777	page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_compile()</b>
				778	returns an error. If you already know that your pattern is valid, and you want
				779	to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
				780	option. When it is set, the effect of passing an invalid UTF-8 string as a
				781	pattern is undefined. It may cause your program to crash. Note that this option
				782	can also be passed to <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, to suppress
				783	the UTF-8 validity checking of subject strings.
				784	</P>
				785	<br><a name="SEC10" href="#TOC1">COMPILATION ERROR CODES</a><br>
				786	<P>
				787	The following table lists the error codes than may be returned by
				788	<b>pcre_compile2()</b>, along with the error messages that may be returned by
				789	both compiling functions. As PCRE has developed, some error codes have fallen
				790	out of use. To avoid confusion, they have not been re-used.
				791	<pre>
				792	0 no error
				793	1 \ at end of pattern
				794	2 \c at end of pattern
				795	3 unrecognized character follows \
				796	4 numbers out of order in {} quantifier
				797	5 number too big in {} quantifier
				798	6 missing terminating ] for character class
				799	7 invalid escape sequence in character class
				800	8 range out of order in character class
				801	9 nothing to repeat
				802	10 [this code is not in use]
				803	11 internal error: unexpected repeat
				804	12 unrecognized character after (? or (?-
				805	13 POSIX named classes are supported only within a class
				806	14 missing )
				807	15 reference to non-existent subpattern
				808	16 erroffset passed as NULL
				809	17 unknown option bit(s) set
				810	18 missing ) after comment
				811	19 [this code is not in use]
				812	20 regular expression is too large
				813	21 failed to get memory
				814	22 unmatched parentheses
				815	23 internal error: code overflow
				816	24 unrecognized character after (?<
				817	25 lookbehind assertion is not fixed length
				818	26 malformed number or name after (?(
				819	27 conditional group contains more than two branches
				820	28 assertion expected after (?(
				821	29 (?R or (?[+-]digits must be followed by )
				822	30 unknown POSIX class name
				823	31 POSIX collating elements are not supported
				824	32 this version of PCRE is not compiled with PCRE_UTF8 support
				825	33 [this code is not in use]
				826	34 character value in \x{...} sequence is too large
				827	35 invalid condition (?(0)
				828	36 \C not allowed in lookbehind assertion
				829	37 PCRE does not support \L, \l, \N{name}, \U, or \u
				830	38 number after (?C is > 255
				831	39 closing ) for (?C expected
				832	40 recursive call could loop indefinitely
				833	41 unrecognized character after (?P
				834	42 syntax error in subpattern name (missing terminator)
				835	43 two named subpatterns have the same name
				836	44 invalid UTF-8 string
				837	45 support for \P, \p, and \X has not been compiled
				838	46 malformed \P or \p sequence
				839	47 unknown property name after \P or \p
				840	48 subpattern name is too long (maximum 32 characters)
				841	49 too many named subpatterns (maximum 10000)
				842	50 [this code is not in use]
				843	51 octal value is greater than \377 (not in UTF-8 mode)
				844	52 internal error: overran compiling workspace
				845	53 internal error: previously-checked referenced subpattern
				846	not found
				847	54 DEFINE group contains more than one branch
				848	55 repeating a DEFINE group is not allowed
				849	56 inconsistent NEWLINE options
				850	57 \g is not followed by a braced, angle-bracketed, or quoted
				851	name/number or by a plain number
				852	58 a numbered reference must not be zero
				853	59 an argument is not allowed for (ACCEPT), (FAIL), or (*COMMIT)
				854	60 (*VERB) not recognized
				855	61 number is too big
				856	62 subpattern name expected
				857	63 digit expected after (?+
				858	64 ] is an invalid data character in JavaScript compatibility mode
				859	65 different names for subpatterns of the same number are
				860	not allowed
				861	66 (*MARK) must have an argument
				862	67 this version of PCRE is not compiled with PCRE_UCP support
				863	68 \c must be followed by an ASCII character
				864	69 \k is not followed by a braced, angle-bracketed, or quoted name
				865	</pre>
				866	The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
				867	be used if the limits were changed when PCRE was built.
				868	<a name="studyingapattern"></a></P>
				869	<br><a name="SEC11" href="#TOC1">STUDYING A PATTERN</a><br>
				870	<P>
				871	<b>pcre_extra pcre_study(const pcre <i>code</i>, int <i>options</i></b>
				872	<b>const char **<i>errptr</i>);</b>
				873	</P>
				874	<P>
				875	If a compiled pattern is going to be used several times, it is worth spending
				876	more time analyzing it in order to speed up the time taken for matching. The
				877	function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
				878	argument. If studying the pattern produces additional information that will
				879	help speed up matching, <b>pcre_study()</b> returns a pointer to a
				880	<b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
				881	results of the study.
				882	</P>
				883	<P>
				884	The returned value from <b>pcre_study()</b> can be passed directly to
				885	<b>pcre_exec()</b> or <b>pcre_dfa_exec()</b>. However, a <b>pcre_extra</b> block
				886	also contains other fields that can be set by the caller before the block is
				887	passed; these are described
				888	<a href="#extradata">below</a>
				889	in the section on matching a pattern.
				890	</P>
				891	<P>
				892	If studying the pattern does not produce any useful information,
				893	<b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
				894	wants to pass any of the other fields to <b>pcre_exec()</b> or
				895	<b>pcre_dfa_exec()</b>, it must set up its own <b>pcre_extra</b> block.
				896	</P>
				897	<P>
				898	The second argument of <b>pcre_study()</b> contains option bits. There is only
				899	one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
				900	compiler is available, the pattern is further compiled into machine code that
				901	executes much faster than the <b>pcre_exec()</b> matching function. If
				902	the just-in-time compiler is not available, this option is ignored. All other
				903	bits in the <i>options</i> argument must be zero.
				904	</P>
				905	<P>
				906	JIT compilation is a heavyweight optimization. It can take some time for
				907	patterns to be analyzed, and for one-off matches and simple patterns the
				908	benefit of faster execution might be offset by a much slower study time.
				909	Not all patterns can be optimized by the JIT compiler. For those that cannot be
				910	handled, matching automatically falls back to the <b>pcre_exec()</b>
				911	interpreter. For more details, see the
				912	<a href="pcrejit.html"><b>pcrejit</b></a>
				913	documentation.
				914	</P>
				915	<P>
				916	The third argument for <b>pcre_study()</b> is a pointer for an error message. If
				917	studying succeeds (even if no data is returned), the variable it points to is
				918	set to NULL. Otherwise it is set to point to a textual error message. This is a
				919	static string that is part of the library. You must not try to free it. You
				920	should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
				921	sure that it has run successfully.
				922	</P>
				923	<P>
				924	When you are finished with a pattern, you can free the memory used for the
				925	study data by calling <b>pcre_free_study()</b>. This function was added to the
				926	API for release 8.20. For earlier versions, the memory could be freed with
				927	<b>pcre_free()</b>, just like the pattern itself. This will still work in cases
				928	where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
				929	new function when convenient.
				930	</P>
				931	<P>
				932	This is a typical way in which <b>pcre_study</b>() is used (except that in a
				933	real application there should be tests for errors):
				934	<pre>
				935	int rc;
				936	pcre *re;
				937	pcre_extra *sd;
				938	re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
				939	sd = pcre_study(
				940	re, /* result of pcre_compile() */
				941	0, /* no options */
				942	&error); /* set to NULL or points to a message */
				943	rc = pcre_exec( /* see below for details of pcre_exec() options */
				944	re, sd, "subject", 7, 0, 0, ovector, 30);
				945	...
				946	pcre_free_study(sd);
				947	pcre_free(re);
				948	</pre>
				949	Studying a pattern does two things: first, a lower bound for the length of
				950	subject string that is needed to match the pattern is computed. This does not
				951	mean that there are any strings of that length that match, but it does
				952	guarantee that no shorter strings match. The value is used by
				953	<b>pcre_exec()</b> and <b>pcre_dfa_exec()</b> to avoid wasting time by trying to
				954	match strings that are shorter than the lower bound. You can find out the value
				955	in a calling program via the <b>pcre_fullinfo()</b> function.
				956	</P>
				957	<P>
				958	Studying a pattern is also useful for non-anchored patterns that do not have a
				959	single fixed starting character. A bitmap of possible starting bytes is
				960	created. This speeds up finding a position in the subject at which to start
				961	matching.
				962	</P>
				963	<P>
				964	These two optimizations apply to both <b>pcre_exec()</b> and
				965	<b>pcre_dfa_exec()</b>. However, they are not used by <b>pcre_exec()</b> if
				966	<b>pcre_study()</b> is called with the PCRE_STUDY_JIT_COMPILE option, and
				967	just-in-time compiling is successful. The optimizations can be disabled by
				968	setting the PCRE_NO_START_OPTIMIZE option when calling <b>pcre_exec()</b> or
				969	<b>pcre_dfa_exec()</b>. You might want to do this if your pattern contains
				970	callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
				971	to make use of these facilities in cases where matching fails. See the
				972	discussion of PCRE_NO_START_OPTIMIZE
				973	<a href="#execoptions">below.</a>
				974	<a name="localesupport"></a></P>
				975	<br><a name="SEC12" href="#TOC1">LOCALE SUPPORT</a><br>
				976	<P>
				977	PCRE handles caseless matching, and determines whether characters are letters,
				978	digits, or whatever, by reference to a set of tables, indexed by character
				979	value. When running in UTF-8 mode, this applies only to characters with codes
				980	less than 128. By default, higher-valued codes never match escapes such as \w
				981	or \d, but they can be tested with \p if PCRE is built with Unicode character
				982	property support. Alternatively, the PCRE_UCP option can be set at compile
				983	time; this causes \w and friends to use Unicode property support instead of
				984	built-in tables. The use of locales with Unicode is discouraged. If you are
				985	handling characters with codes greater than 128, you should either use UTF-8
				986	and Unicode, or use locales, but not try to mix the two.
				987	</P>
				988	<P>
				989	PCRE contains an internal set of tables that are used when the final argument
				990	of <b>pcre_compile()</b> is NULL. These are sufficient for many applications.
				991	Normally, the internal tables recognize only ASCII characters. However, when
				992	PCRE is built, it is possible to cause the internal tables to be rebuilt in the
				993	default "C" locale of the local system, which may cause them to be different.
				994	</P>
				995	<P>
				996	The internal tables can always be overridden by tables supplied by the
				997	application that calls PCRE. These may be created in a different locale from
				998	the default. As more and more applications change to using Unicode, the need
				999	for this locale support is expected to die away.
				1000	</P>
				1001	<P>
				1002	External tables are built by calling the <b>pcre_maketables()</b> function,
				1003	which has no arguments, in the relevant locale. The result can then be passed
				1004	to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
				1005	example, to build and use tables that are appropriate for the French locale
				1006	(where accented characters with values greater than 128 are treated as letters),
				1007	the following code could be used:
				1008	<pre>
				1009	setlocale(LC_CTYPE, "fr_FR");
				1010	tables = pcre_maketables();
				1011	re = pcre_compile(..., tables);
				1012	</pre>
				1013	The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
				1014	are using Windows, the name for the French locale is "french".
				1015	</P>
				1016	<P>
				1017	When <b>pcre_maketables()</b> runs, the tables are built in memory that is
				1018	obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
				1019	that the memory containing the tables remains available for as long as it is
				1020	needed.
				1021	</P>
				1022	<P>
				1023	The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
				1024	pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
				1025	and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
				1026	pattern, compilation, studying and matching all happen in the same locale, but
				1027	different patterns can be compiled in different locales.
				1028	</P>
				1029	<P>
				1030	It is possible to pass a table pointer or NULL (indicating the use of the
				1031	internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
				1032	this facility could be used to match a pattern in a different locale from the
				1033	one in which it was compiled. Passing table pointers at run time is discussed
				1034	below in the section on matching a pattern.
				1035	<a name="infoaboutpattern"></a></P>
				1036	<br><a name="SEC13" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
				1037	<P>
				1038	<b>int pcre_fullinfo(const pcre <i>code</i>, const pcre_extra <i>extra</i>,</b>
				1039	<b>int <i>what</i>, void *<i>where</i>);</b>
				1040	</P>
				1041	<P>
				1042	The <b>pcre_fullinfo()</b> function returns information about a compiled
				1043	pattern. It replaces the obsolete <b>pcre_info()</b> function, which is
				1044	nevertheless retained for backwards compability (and is documented below).
				1045	</P>
				1046	<P>
				1047	The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
				1048	pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
				1049	the pattern was not studied. The third argument specifies which piece of
				1050	information is required, and the fourth argument is a pointer to a variable
				1051	to receive the data. The yield of the function is zero for success, or one of
				1052	the following negative numbers:
				1053	<pre>
				1054	PCRE_ERROR_NULL the argument <i>code</i> was NULL
				1055	the argument <i>where</i> was NULL
				1056	PCRE_ERROR_BADMAGIC the "magic number" was not found
				1057	PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
				1058	</pre>
				1059	The "magic number" is placed at the start of each compiled pattern as an simple
				1060	check against passing an arbitrary memory pointer. Here is a typical call of
				1061	<b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern:
				1062	<pre>
				1063	int rc;
				1064	size_t length;
				1065	rc = pcre_fullinfo(
				1066	re, /* result of pcre_compile() */
				1067	sd, /* result of pcre_study(), or NULL */
				1068	PCRE_INFO_SIZE, /* what is required */
				1069	&length); /* where to put the data */
				1070	</pre>
				1071	The possible values for the third argument are defined in <b>pcre.h</b>, and are
				1072	as follows:
				1073	<pre>
				1074	PCRE_INFO_BACKREFMAX
				1075	</pre>
				1076	Return the number of the highest back reference in the pattern. The fourth
				1077	argument should point to an <b>int</b> variable. Zero is returned if there are
				1078	no back references.
				1079	<pre>
				1080	PCRE_INFO_CAPTURECOUNT
				1081	</pre>
				1082	Return the number of capturing subpatterns in the pattern. The fourth argument
				1083	should point to an <b>int</b> variable.
				1084	<pre>
				1085	PCRE_INFO_DEFAULT_TABLES
				1086	</pre>
				1087	Return a pointer to the internal default character tables within PCRE. The
				1088	fourth argument should point to an <b>unsigned char *</b> variable. This
				1089	information call is provided for internal use by the <b>pcre_study()</b>
				1090	function. External callers can cause PCRE to use its internal tables by passing
				1091	a NULL table pointer.
				1092	<pre>
				1093	PCRE_INFO_FIRSTBYTE
				1094	</pre>
				1095	Return information about the first byte of any matched string, for a
				1096	non-anchored pattern. The fourth argument should point to an <b>int</b>
				1097	variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
				1098	still recognized for backwards compatibility.)
				1099	</P>
				1100	<P>
				1101	If there is a fixed first byte, for example, from a pattern such as
				1102	(cat\|cow\|coyote), its value is returned. Otherwise, if either
				1103	<br>
				1104	<br>
				1105	(a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
				1106	starts with "^", or
				1107	<br>
				1108	<br>
				1109	(b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
				1110	(if it were set, the pattern would be anchored),
				1111	<br>
				1112	<br>
				1113	-1 is returned, indicating that the pattern matches only at the start of a
				1114	subject string or after any newline within the string. Otherwise -2 is
				1115	returned. For anchored patterns, -2 is returned.
				1116	<pre>
				1117	PCRE_INFO_FIRSTTABLE
				1118	</pre>
				1119	If the pattern was studied, and this resulted in the construction of a 256-bit
				1120	table indicating a fixed set of bytes for the first byte in any matching
				1121	string, a pointer to the table is returned. Otherwise NULL is returned. The
				1122	fourth argument should point to an <b>unsigned char *</b> variable.
				1123	<pre>
				1124	PCRE_INFO_HASCRORLF
				1125	</pre>
				1126	Return 1 if the pattern contains any explicit matches for CR or LF characters,
				1127	otherwise 0. The fourth argument should point to an <b>int</b> variable. An
				1128	explicit match is either a literal CR or LF character, or \r or \n.
				1129	<pre>
				1130	PCRE_INFO_JCHANGED
				1131	</pre>
				1132	Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
				1133	0. The fourth argument should point to an <b>int</b> variable. (?J) and
				1134	(?-J) set and unset the local PCRE_DUPNAMES option, respectively.
				1135	<pre>
				1136	PCRE_INFO_JIT
				1137	</pre>
				1138	Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
				1139	just-in-time compiling was successful. The fourth argument should point to an
				1140	<b>int</b> variable. A return value of 0 means that JIT support is not available
				1141	in this version of PCRE, or that the pattern was not studied with the
				1142	PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
				1143	particular pattern. See the
				1144	<a href="pcrejit.html"><b>pcrejit</b></a>
				1145	documentation for details of what can and cannot be handled.
				1146	<pre>
				1147	PCRE_INFO_JITSIZE
				1148	</pre>
				1149	If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
				1150	return the size of the JIT compiled code, otherwise return zero. The fourth
				1151	argument should point to a <b>size_t</b> variable.
				1152	<pre>
				1153	PCRE_INFO_LASTLITERAL
				1154	</pre>
				1155	Return the value of the rightmost literal byte that must exist in any matched
				1156	string, other than at its start, if such a byte has been recorded. The fourth
				1157	argument should point to an <b>int</b> variable. If there is no such byte, -1 is
				1158	returned. For anchored patterns, a last literal byte is recorded only if it
				1159	follows something of variable length. For example, for the pattern
				1160	/^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
				1161	is -1.
				1162	<pre>
				1163	PCRE_INFO_MINLENGTH
				1164	</pre>
				1165	If the pattern was studied and a minimum length for matching subject strings
				1166	was computed, its value is returned. Otherwise the returned value is -1. The
				1167	value is a number of characters, not bytes (this may be relevant in UTF-8
				1168	mode). The fourth argument should point to an <b>int</b> variable. A
				1169	non-negative value is a lower bound to the length of any matching string. There
				1170	may not be any strings of that length that do actually match, but every string
				1171	that does match is at least that long.
				1172	<pre>
				1173	PCRE_INFO_NAMECOUNT
				1174	PCRE_INFO_NAMEENTRYSIZE
				1175	PCRE_INFO_NAMETABLE
				1176	</pre>
				1177	PCRE supports the use of named as well as numbered capturing parentheses. The
				1178	names are just an additional way of identifying the parentheses, which still
				1179	acquire numbers. Several convenience functions such as
				1180	<b>pcre_get_named_substring()</b> are provided for extracting captured
				1181	substrings by name. It is also possible to extract the data directly, by first
				1182	converting the name to a number in order to access the correct pointers in the
				1183	output vector (described with <b>pcre_exec()</b> below). To do the conversion,
				1184	you need to use the name-to-number map, which is described by these three
				1185	values.
				1186	</P>
				1187	<P>
				1188	The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
				1189	the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
				1190	entry; both of these return an <b>int</b> value. The entry size depends on the
				1191	length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
				1192	entry of the table (a pointer to <b>char</b>). The first two bytes of each entry
				1193	are the number of the capturing parenthesis, most significant byte first. The
				1194	rest of the entry is the corresponding name, zero terminated.
				1195	</P>
				1196	<P>
				1197	The names are in alphabetical order. Duplicate names may appear if (?\| is used
				1198	to create multiple groups with the same number, as described in the
				1199	<a href="pcrepattern.html#dupsubpatternnumber">section on duplicate subpattern numbers</a>
				1200	in the
				1201	<a href="pcrepattern.html"><b>pcrepattern</b></a>
				1202	page. Duplicate names for subpatterns with different numbers are permitted only
				1203	if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
				1204	table in the order in which they were found in the pattern. In the absence of
				1205	(?\| this is the order of increasing number; when (?\| is used this is not
				1206	necessarily the case because later subpatterns may have lower numbers.
				1207	</P>
				1208	<P>
				1209	As a simple example of the name/number table, consider the following pattern
				1210	(assume PCRE_EXTENDED is set, so white space - including newlines - is
				1211	ignored):
				1212	<pre>
				1213	(?<date> (?<year>(\d\d)?\d\d) - (?<month>\d\d) - (?<day>\d\d) )
				1214	</pre>
				1215	There are four named subpatterns, so the table has four entries, and each entry
				1216	in the table is eight bytes long. The table is as follows, with non-printing
				1217	bytes shows in hexadecimal, and undefined bytes shown as ??:
				1218	<pre>
				1219	00 01 d a t e 00 ??
				1220	00 05 d a y 00 ?? ??
				1221	00 04 m o n t h 00
				1222	00 02 y e a r 00 ??
				1223	</pre>
				1224	When writing code to extract data from named subpatterns using the
				1225	name-to-number map, remember that the length of the entries is likely to be
				1226	different for each compiled pattern.
				1227	<pre>
				1228	PCRE_INFO_OKPARTIAL
				1229	</pre>
				1230	Return 1 if the pattern can be used for partial matching with
				1231	<b>pcre_exec()</b>, otherwise 0. The fourth argument should point to an
				1232	<b>int</b> variable. From release 8.00, this always returns 1, because the
				1233	restrictions that previously applied to partial matching have been lifted. The
				1234	<a href="pcrepartial.html"><b>pcrepartial</b></a>
				1235	documentation gives details of partial matching.
				1236	<pre>
				1237	PCRE_INFO_OPTIONS
				1238	</pre>
				1239	Return a copy of the options with which the pattern was compiled. The fourth
				1240	argument should point to an <b>unsigned long int</b> variable. These option bits
				1241	are those specified in the call to <b>pcre_compile()</b>, modified by any
				1242	top-level option settings at the start of the pattern itself. In other words,
				1243	they are the options that will be in force when matching starts. For example,
				1244	if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
				1245	result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
				1246	</P>
				1247	<P>
				1248	A pattern is automatically anchored by PCRE if all of its top-level
				1249	alternatives begin with one of the following:
				1250	<pre>
				1251	^ unless PCRE_MULTILINE is set
				1252	\A always
				1253	\G always
				1254	.* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
				1255	</pre>
				1256	For such patterns, the PCRE_ANCHORED bit is set in the options returned by
				1257	<b>pcre_fullinfo()</b>.
				1258	<pre>
				1259	PCRE_INFO_SIZE
				1260	</pre>
				1261	Return the size of the compiled pattern. The fourth argument should point to a
				1262	<b>size_t</b> variable. This value does not include the size of the <b>pcre</b>
				1263	structure that is returned by <b>pcre_compile()</b>. The value that is passed as
				1264	the argument to <b>pcre_malloc()</b> when <b>pcre_compile()</b> is getting memory
				1265	in which to place the compiled data is the value returned by this option plus
				1266	the size of the <b>pcre</b> structure. Studying a compiled pattern, with or
				1267	without JIT, does not alter the value returned by this option.
				1268	<pre>
				1269	PCRE_INFO_STUDYSIZE
				1270	</pre>
				1271	Return the size of the data block pointed to by the <i>study_data</i> field in a
				1272	<b>pcre_extra</b> block. If <b>pcre_extra</b> is NULL, or there is no study data,
				1273	zero is returned. The fourth argument should point to a <b>size_t</b> variable.
				1274	The <i>study_data</i> field is set by <b>pcre_study()</b> to record information
				1275	that will speed up matching (see the section entitled
				1276	<a href="#studyingapattern">"Studying a pattern"</a>
				1277	above). The format of the <i>study_data</i> block is private, but its length
				1278	is made available via this option so that it can be saved and restored (see the
				1279	<a href="pcreprecompile.html"><b>pcreprecompile</b></a>
				1280	documentation for details).
				1281	</P>
				1282	<br><a name="SEC14" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
				1283	<P>
				1284	<b>int pcre_info(const pcre <i>code</i>, int <i>optptr</i>, int</b>
				1285	<b>*<i>firstcharptr</i>);</b>
				1286	</P>
				1287	<P>
				1288	The <b>pcre_info()</b> function is now obsolete because its interface is too
				1289	restrictive to return all the available data about a compiled pattern. New
				1290	programs should use <b>pcre_fullinfo()</b> instead. The yield of
				1291	<b>pcre_info()</b> is the number of capturing subpatterns, or one of the
				1292	following negative numbers:
				1293	<pre>
				1294	PCRE_ERROR_NULL the argument <i>code</i> was NULL
				1295	PCRE_ERROR_BADMAGIC the "magic number" was not found
				1296	</pre>
				1297	If the <i>optptr</i> argument is not NULL, a copy of the options with which the
				1298	pattern was compiled is placed in the integer it points to (see
				1299	PCRE_INFO_OPTIONS above).
				1300	</P>
				1301	<P>
				1302	If the pattern is not anchored and the <i>firstcharptr</i> argument is not NULL,
				1303	it is used to pass back information about the first character of any matched
				1304	string (see PCRE_INFO_FIRSTBYTE above).
				1305	</P>
				1306	<br><a name="SEC15" href="#TOC1">REFERENCE COUNTS</a><br>
				1307	<P>
				1308	<b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
				1309	</P>
				1310	<P>
				1311	The <b>pcre_refcount()</b> function is used to maintain a reference count in the
				1312	data block that contains a compiled pattern. It is provided for the benefit of
				1313	applications that operate in an object-oriented manner, where different parts
				1314	of the application may be using the same compiled pattern, but you want to free
				1315	the block when they are all done.
				1316	</P>
				1317	<P>
				1318	When a pattern is compiled, the reference count field is initialized to zero.
				1319	It is changed only by calling this function, whose action is to add the
				1320	<i>adjust</i> value (which may be positive or negative) to it. The yield of the
				1321	function is the new value. However, the value of the count is constrained to
				1322	lie between 0 and 65535, inclusive. If the new value is outside these limits,
				1323	it is forced to the appropriate limit value.
				1324	</P>
				1325	<P>
				1326	Except when it is zero, the reference count is not correctly preserved if a
				1327	pattern is compiled on one host and then transferred to a host whose byte-order
				1328	is different. (This seems a highly unlikely scenario.)
				1329	</P>
				1330	<br><a name="SEC16" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
				1331	<P>
				1332	<b>int pcre_exec(const pcre <i>code</i>, const pcre_extra <i>extra</i>,</b>
				1333	<b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
				1334	<b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
				1335	</P>
				1336	<P>
				1337	The function <b>pcre_exec()</b> is called to match a subject string against a
				1338	compiled pattern, which is passed in the <i>code</i> argument. If the
				1339	pattern was studied, the result of the study should be passed in the
				1340	<i>extra</i> argument. You can call <b>pcre_exec()</b> with the same <i>code</i>
				1341	and <i>extra</i> arguments as many times as you like, in order to match
				1342	different subject strings with the same pattern.
				1343	</P>
				1344	<P>
				1345	This function is the main matching facility of the library, and it operates in
				1346	a Perl-like manner. For specialist use there is also an alternative matching
				1347	function, which is described
				1348	<a href="#dfamatch">below</a>
				1349	in the section about the <b>pcre_dfa_exec()</b> function.
				1350	</P>
				1351	<P>
				1352	In most applications, the pattern will have been compiled (and optionally
				1353	studied) in the same process that calls <b>pcre_exec()</b>. However, it is
				1354	possible to save compiled patterns and study data, and then use them later
				1355	in different processes, possibly even on different hosts. For a discussion
				1356	about this, see the
				1357	<a href="pcreprecompile.html"><b>pcreprecompile</b></a>
				1358	documentation.
				1359	</P>
				1360	<P>
				1361	Here is an example of a simple call to <b>pcre_exec()</b>:
				1362	<pre>
				1363	int rc;
				1364	int ovector[30];
				1365	rc = pcre_exec(
				1366	re, /* result of pcre_compile() */
				1367	NULL, /* we didn't study the pattern */
				1368	"some string", /* the subject string */
				1369	11, /* the length of the subject string */
				1370	0, /* start at offset 0 in the subject */
				1371	0, /* default options */
				1372	ovector, /* vector of integers for substring information */
				1373	30); /* number of elements (NOT size in bytes) */
				1374	<a name="extradata"></a></PRE>
				1375	</P>
				1376	<br><b>
				1377	Extra data for <b>pcre_exec()</b>
				1378	</b><br>
				1379	<P>
				1380	If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
				1381	data block. The <b>pcre_study()</b> function returns such a block (when it
				1382	doesn't return NULL), but you can also create one for yourself, and pass
				1383	additional information in it. The <b>pcre_extra</b> block contains the following
				1384	fields (not necessarily in this order):
				1385	<pre>
				1386	unsigned long int <i>flags</i>;
				1387	void *<i>study_data</i>;
				1388	void *<i>executable_jit</i>;
				1389	unsigned long int <i>match_limit</i>;
				1390	unsigned long int <i>match_limit_recursion</i>;
				1391	void *<i>callout_data</i>;
				1392	const unsigned char *<i>tables</i>;
				1393	unsigned char **<i>mark</i>;
				1394	</pre>
				1395	The <i>flags</i> field is a bitmap that specifies which of the other fields
				1396	are set. The flag bits are:
				1397	<pre>
				1398	PCRE_EXTRA_STUDY_DATA
				1399	PCRE_EXTRA_EXECUTABLE_JIT
				1400	PCRE_EXTRA_MATCH_LIMIT
				1401	PCRE_EXTRA_MATCH_LIMIT_RECURSION
				1402	PCRE_EXTRA_CALLOUT_DATA
				1403	PCRE_EXTRA_TABLES
				1404	PCRE_EXTRA_MARK
				1405	</pre>
				1406	Other flag bits should be set to zero. The <i>study_data</i> field and sometimes
				1407	the <i>executable_jit</i> field are set in the <b>pcre_extra</b> block that is
				1408	returned by <b>pcre_study()</b>, together with the appropriate flag bits. You
				1409	should not set these yourself, but you may add to the block by setting the
				1410	other fields and their corresponding flag bits.
				1411	</P>
				1412	<P>
				1413	The <i>match_limit</i> field provides a means of preventing PCRE from using up a
				1414	vast amount of resources when running patterns that are not going to match,
				1415	but which have a very large number of possibilities in their search trees. The
				1416	classic example is a pattern that uses nested unlimited repeats.
				1417	</P>
				1418	<P>
				1419	Internally, <b>pcre_exec()</b> uses a function called <b>match()</b>, which it
				1420	calls repeatedly (sometimes recursively). The limit set by <i>match_limit</i> is
				1421	imposed on the number of times this function is called during a match, which
				1422	has the effect of limiting the amount of backtracking that can take place. For
				1423	patterns that are not anchored, the count restarts from zero for each position
				1424	in the subject string.
				1425	</P>
				1426	<P>
				1427	When <b>pcre_exec()</b> is called with a pattern that was successfully studied
				1428	with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
				1429	is entirely different. However, there is still the possibility of runaway
				1430	matching that goes on for a very long time, and so the <i>match_limit</i> value
				1431	is also used in this case (but in a different way) to limit how long the
				1432	matching can continue.
				1433	</P>
				1434	<P>
				1435	The default value for the limit can be set when PCRE is built; the default
				1436	default is 10 million, which handles all but the most extreme cases. You can
				1437	override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
				1438	block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
				1439	the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
				1440	PCRE_ERROR_MATCHLIMIT.
				1441	</P>
				1442	<P>
				1443	The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
				1444	instead of limiting the total number of times that <b>match()</b> is called, it
				1445	limits the depth of recursion. The recursion depth is a smaller number than the
				1446	total number of calls, because not all calls to <b>match()</b> are recursive.
				1447	This limit is of use only if it is set smaller than <i>match_limit</i>.
				1448	</P>
				1449	<P>
				1450	Limiting the recursion depth limits the amount of machine stack that can be
				1451	used, or, when PCRE has been compiled to use memory on the heap instead of the
				1452	stack, the amount of heap memory that can be used. This limit is not relevant,
				1453	and is ignored, if the pattern was successfully studied with
				1454	PCRE_STUDY_JIT_COMPILE.
				1455	</P>
				1456	<P>
				1457	The default value for <i>match_limit_recursion</i> can be set when PCRE is
				1458	built; the default default is the same value as the default for
				1459	<i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
				1460	with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
				1461	PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
				1462	is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
				1463	</P>
				1464	<P>
				1465	The <i>callout_data</i> field is used in conjunction with the "callout" feature,
				1466	and is described in the
				1467	<a href="pcrecallout.html"><b>pcrecallout</b></a>
				1468	documentation.
				1469	</P>
				1470	<P>
				1471	The <i>tables</i> field is used to pass a character tables pointer to
				1472	<b>pcre_exec()</b>; this overrides the value that is stored with the compiled
				1473	pattern. A non-NULL value is stored with the compiled pattern only if custom
				1474	tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
				1475	If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
				1476	internal tables to be used. This facility is helpful when re-using patterns
				1477	that have been saved after compiling with an external set of tables, because
				1478	the external tables might be at a different address when <b>pcre_exec()</b> is
				1479	called. See the
				1480	<a href="pcreprecompile.html"><b>pcreprecompile</b></a>
				1481	documentation for a discussion of saving compiled patterns for later use.
				1482	</P>
				1483	<P>
				1484	If PCRE_EXTRA_MARK is set in the <i>flags</i> field, the <i>mark</i> field must
				1485	be set to point to a <b>char *</b> variable. If the pattern contains any
				1486	backtracking control verbs such as (*MARK:NAME), and the execution ends up with
				1487	a name to pass back, a pointer to the name string (zero terminated) is placed
				1488	in the variable pointed to by the <i>mark</i> field. The names are within the
				1489	compiled pattern; if you wish to retain such a name you must copy it before
				1490	freeing the memory of a compiled pattern. If there is no name to pass back, the
				1491	variable pointed to by the <i>mark</i> field set to NULL. For details of the
				1492	backtracking control verbs, see the section entitled
				1493	<a href="pcrepattern#backtrackcontrol">"Backtracking control"</a>
				1494	in the
				1495	<a href="pcrepattern.html"><b>pcrepattern</b></a>
				1496	documentation.
				1497	<a name="execoptions"></a></P>
				1498	<br><b>
				1499	Option bits for <b>pcre_exec()</b>
				1500	</b><br>
				1501	<P>
				1502	The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
				1503	zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
				1504	PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
				1505	PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
				1506	PCRE_PARTIAL_HARD.
				1507	</P>
				1508	<P>
				1509	If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
				1510	the only supported options for JIT execution are PCRE_NO_UTF8_CHECK,
				1511	PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, and PCRE_NOTEMPTY_ATSTART. Note in
				1512	particular that partial matching is not supported. If an unsupported option is
				1513	used, JIT execution is disabled and the normal interpretive code in
				1514	<b>pcre_exec()</b> is run.
				1515	<pre>
				1516	PCRE_ANCHORED
				1517	</pre>
				1518	The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
				1519	matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
				1520	to be anchored by virtue of its contents, it cannot be made unachored at
				1521	matching time.
				1522	<pre>
				1523	PCRE_BSR_ANYCRLF
				1524	PCRE_BSR_UNICODE
				1525	</pre>
				1526	These options (which are mutually exclusive) control what the \R escape
				1527	sequence matches. The choice is either to match only CR, LF, or CRLF, or to
				1528	match any Unicode newline sequence. These options override the choice that was
				1529	made or defaulted when the pattern was compiled.
				1530	<pre>
				1531	PCRE_NEWLINE_CR
				1532	PCRE_NEWLINE_LF
				1533	PCRE_NEWLINE_CRLF
				1534	PCRE_NEWLINE_ANYCRLF
				1535	PCRE_NEWLINE_ANY
				1536	</pre>
				1537	These options override the newline definition that was chosen or defaulted when
				1538	the pattern was compiled. For details, see the description of
				1539	<b>pcre_compile()</b> above. During matching, the newline choice affects the
				1540	behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
				1541	the way the match position is advanced after a match failure for an unanchored
				1542	pattern.
				1543	</P>
				1544	<P>
				1545	When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
				1546	match attempt for an unanchored pattern fails when the current position is at a
				1547	CRLF sequence, and the pattern contains no explicit matches for CR or LF
				1548	characters, the match position is advanced by two characters instead of one, in
				1549	other words, to after the CRLF.
				1550	</P>
				1551	<P>
				1552	The above rule is a compromise that makes the most common cases work as
				1553	expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
				1554	set), it does not match the string "\r\nA" because, after failing at the
				1555	start, it skips both the CR and the LF before retrying. However, the pattern
				1556	[\r\n]A does match that string, because it contains an explicit CR or LF
				1557	reference, and so advances only by one character after the first failure.
				1558	</P>
				1559	<P>
				1560	An explicit match for CR of LF is either a literal appearance of one of those
				1561	characters, or one of the \r or \n escape sequences. Implicit matches such as
				1562	[^X] do not count, nor does \s (which includes CR and LF in the characters
				1563	that it matches).
				1564	</P>
				1565	<P>
				1566	Notwithstanding the above, anomalous effects may still occur when CRLF is a
				1567	valid newline sequence and explicit \r or \n escapes appear in the pattern.
				1568	<pre>
				1569	PCRE_NOTBOL
				1570	</pre>
				1571	This option specifies that first character of the subject string is not the
				1572	beginning of a line, so the circumflex metacharacter should not match before
				1573	it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
				1574	never to match. This option affects only the behaviour of the circumflex
				1575	metacharacter. It does not affect \A.
				1576	<pre>
				1577	PCRE_NOTEOL
				1578	</pre>
				1579	This option specifies that the end of the subject string is not the end of a
				1580	line, so the dollar metacharacter should not match it nor (except in multiline
				1581	mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
				1582	compile time) causes dollar never to match. This option affects only the
				1583	behaviour of the dollar metacharacter. It does not affect \Z or \z.
				1584	<pre>
				1585	PCRE_NOTEMPTY
				1586	</pre>
				1587	An empty string is not considered to be a valid match if this option is set. If
				1588	there are alternatives in the pattern, they are tried. If all the alternatives
				1589	match the empty string, the entire match fails. For example, if the pattern
				1590	<pre>
				1591	a?b?
				1592	</pre>
				1593	is applied to a string not beginning with "a" or "b", it matches an empty
				1594	string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
				1595	valid, so PCRE searches further into the string for occurrences of "a" or "b".
				1596	<pre>
				1597	PCRE_NOTEMPTY_ATSTART
				1598	</pre>
				1599	This is like PCRE_NOTEMPTY, except that an empty string match that is not at
				1600	the start of the subject is permitted. If the pattern is anchored, such a match
				1601	can occur only if the pattern contains \K.
				1602	</P>
				1603	<P>
				1604	Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
				1605	does make a special case of a pattern match of the empty string within its
				1606	<b>split()</b> function, and when using the /g modifier. It is possible to
				1607	emulate Perl's behaviour after matching a null string by first trying the match
				1608	again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
				1609	if that fails, by advancing the starting offset (see below) and trying an
				1610	ordinary match again. There is some code that demonstrates how to do this in
				1611	the
				1612	<a href="pcredemo.html"><b>pcredemo</b></a>
				1613	sample program. In the most general case, you have to check to see if the
				1614	newline convention recognizes CRLF as a newline, and if so, and the current
				1615	character is CR followed by LF, advance the starting offset by two characters
				1616	instead of one.
				1617	<pre>
				1618	PCRE_NO_START_OPTIMIZE
				1619	</pre>
				1620	There are a number of optimizations that <b>pcre_exec()</b> uses at the start of
				1621	a match, in order to speed up the process. For example, if it is known that an
				1622	unanchored match must start with a specific character, it searches the subject
				1623	for that character, and fails immediately if it cannot find it, without
				1624	actually running the main matching function. This means that a special item
				1625	such as (*COMMIT) at the start of a pattern is not considered until after a
				1626	suitable starting point for the match has been found. When callouts or (*MARK)
				1627	items are in use, these "start-up" optimizations can cause them to be skipped
				1628	if the pattern is never actually used. The start-up optimizations are in effect
				1629	a pre-scan of the subject that takes place before the pattern is run.
				1630	</P>
				1631	<P>
				1632	The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
				1633	causing performance to suffer, but ensuring that in cases where the result is
				1634	"no match", the callouts do occur, and that items such as (COMMIT) and (MARK)
				1635	are considered at every possible starting position in the subject string. If
				1636	PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
				1637	time.
				1638	</P>
				1639	<P>
				1640	Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
				1641	Consider the pattern
				1642	<pre>
				1643	(*COMMIT)ABC
				1644	</pre>
				1645	When this is compiled, PCRE records the fact that a match must start with the
				1646	character "A". Suppose the subject string is "DEFABC". The start-up
				1647	optimization scans along the subject, finds "A" and runs the first match
				1648	attempt from there. The (*COMMIT) item means that the pattern must match the
				1649	current starting position, which in this case, it does. However, if the same
				1650	match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
				1651	subject string does not happen. The first match attempt is run starting from
				1652	"D" and when this fails, (*COMMIT) prevents any further matches being tried, so
				1653	the overall result is "no match". If the pattern is studied, more start-up
				1654	optimizations may be used. For example, a minimum length for the subject may be
				1655	recorded. Consider the pattern
				1656	<pre>
				1657	(*MARK:A)(X\|Y)
				1658	</pre>
				1659	The minimum length for a match is one character. If the subject is "ABC", there
				1660	will be attempts to match "ABC", "BC", "C", and then finally an empty string.
				1661	If the pattern is studied, the final attempt does not take place, because PCRE
				1662	knows that the subject is too short, and so the (*MARK) is never encountered.
				1663	In this case, studying the pattern does not affect the overall match result,
				1664	which is still "no match", but it does affect the auxiliary information that is
				1665	returned.
				1666	<pre>
				1667	PCRE_NO_UTF8_CHECK
				1668	</pre>
				1669	When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
				1670	string is automatically checked when <b>pcre_exec()</b> is subsequently called.
				1671	The value of <i>startoffset</i> is also checked to ensure that it points to the
				1672	start of a UTF-8 character. There is a discussion about the validity of UTF-8
				1673	strings in the
				1674	<a href="pcre.html#utf8strings">section on UTF-8 support</a>
				1675	in the main
				1676	<a href="pcre.html"><b>pcre</b></a>
				1677	page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_exec()</b> returns
				1678	the error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is
				1679	a truncated UTF-8 character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
				1680	both cases, information about the precise nature of the error may also be
				1681	returned (see the descriptions of these errors in the section entitled \fIError
				1682	return values from\fP <b>pcre_exec()</b>
				1683	<a href="#errorlist">below).</a>
				1684	If <i>startoffset</i> contains a value that does not point to the start of a
				1685	UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
				1686	returned.
				1687	</P>
				1688	<P>
				1689	If you already know that your subject is valid, and you want to skip these
				1690	checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
				1691	calling <b>pcre_exec()</b>. You might want to do this for the second and
				1692	subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
				1693	all the matches in a single subject string. However, you should be sure that
				1694	the value of <i>startoffset</i> points to the start of a UTF-8 character (or the
				1695	end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
				1696	invalid UTF-8 string as a subject or an invalid value of <i>startoffset</i> is
				1697	undefined. Your program may crash.
				1698	<pre>
				1699	PCRE_PARTIAL_HARD
				1700	PCRE_PARTIAL_SOFT
				1701	</pre>
				1702	These options turn on the partial matching feature. For backwards
				1703	compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
				1704	occurs if the end of the subject string is reached successfully, but there are
				1705	not enough subject characters to complete the match. If this happens when
				1706	PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
				1707	testing any remaining alternatives. Only if no complete match can be found is
				1708	PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
				1709	PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
				1710	but only if no complete match can be found.
				1711	</P>
				1712	<P>
				1713	If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
				1714	partial match is found, <b>pcre_exec()</b> immediately returns
				1715	PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
				1716	when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
				1717	important that an alternative complete match.
				1718	</P>
				1719	<P>
				1720	In both cases, the portion of the string that was inspected when the partial
				1721	match was found is set as the first matching string. There is a more detailed
				1722	discussion of partial and multi-segment matching, with examples, in the
				1723	<a href="pcrepartial.html"><b>pcrepartial</b></a>
				1724	documentation.
				1725	</P>
				1726	<br><b>
				1727	The string to be matched by <b>pcre_exec()</b>
				1728	</b><br>
				1729	<P>
				1730	The subject string is passed to <b>pcre_exec()</b> as a pointer in
				1731	<i>subject</i>, a length (in bytes) in <i>length</i>, and a starting byte offset
				1732	in <i>startoffset</i>. If this is negative or greater than the length of the
				1733	subject, <b>pcre_exec()</b> returns PCRE_ERROR_BADOFFSET. When the starting
				1734	offset is zero, the search for a match starts at the beginning of the subject,
				1735	and this is by far the most common case. In UTF-8 mode, the byte offset must
				1736	point to the start of a UTF-8 character (or the end of the subject). Unlike the
				1737	pattern string, the subject may contain binary zero bytes.
				1738	</P>
				1739	<P>
				1740	A non-zero starting offset is useful when searching for another match in the
				1741	same subject by calling <b>pcre_exec()</b> again after a previous success.
				1742	Setting <i>startoffset</i> differs from just passing over a shortened string and
				1743	setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
				1744	lookbehind. For example, consider the pattern
				1745	<pre>
				1746	\Biss\B
				1747	</pre>
				1748	which finds occurrences of "iss" in the middle of words. (\B matches only if
				1749	the current position in the subject is not a word boundary.) When applied to
				1750	the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
				1751	occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
				1752	subject, namely "issipi", it does not match, because \B is always false at the
				1753	start of the subject, which is deemed to be a word boundary. However, if
				1754	<b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
				1755	set to 4, it finds the second occurrence of "iss" because it is able to look
				1756	behind the starting point to discover that it is preceded by a letter.
				1757	</P>
				1758	<P>
				1759	Finding all the matches in a subject is tricky when the pattern can match an
				1760	empty string. It is possible to emulate Perl's /g behaviour by first trying the
				1761	match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
				1762	PCRE_ANCHORED options, and then if that fails, advancing the starting offset
				1763	and trying an ordinary match again. There is some code that demonstrates how to
				1764	do this in the
				1765	<a href="pcredemo.html"><b>pcredemo</b></a>
				1766	sample program. In the most general case, you have to check to see if the
				1767	newline convention recognizes CRLF as a newline, and if so, and the current
				1768	character is CR followed by LF, advance the starting offset by two characters
				1769	instead of one.
				1770	</P>
				1771	<P>
				1772	If a non-zero starting offset is passed when the pattern is anchored, one
				1773	attempt to match at the given offset is made. This can only succeed if the
				1774	pattern does not require the match to be at the start of the subject.
				1775	</P>
				1776	<br><b>
				1777	How <b>pcre_exec()</b> returns captured substrings
				1778	</b><br>
				1779	<P>
				1780	In general, a pattern matches a certain portion of the subject, and in
				1781	addition, further substrings from the subject may be picked out by parts of the
				1782	pattern. Following the usage in Jeffrey Friedl's book, this is called
				1783	"capturing" in what follows, and the phrase "capturing subpattern" is used for
				1784	a fragment of a pattern that picks out a substring. PCRE supports several other
				1785	kinds of parenthesized subpattern that do not cause substrings to be captured.
				1786	</P>
				1787	<P>
				1788	Captured substrings are returned to the caller via a vector of integers whose
				1789	address is passed in <i>ovector</i>. The number of elements in the vector is
				1790	passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>: this
				1791	argument is NOT the size of <i>ovector</i> in bytes.
				1792	</P>
				1793	<P>
				1794	The first two-thirds of the vector is used to pass back captured substrings,
				1795	each substring using a pair of integers. The remaining third of the vector is
				1796	used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
				1797	and is not available for passing back information. The number passed in
				1798	<i>ovecsize</i> should always be a multiple of three. If it is not, it is
				1799	rounded down.
				1800	</P>
				1801	<P>
				1802	When a match is successful, information about captured substrings is returned
				1803	in pairs of integers, starting at the beginning of <i>ovector</i>, and
				1804	continuing up to two-thirds of its length at the most. The first element of
				1805	each pair is set to the byte offset of the first character in a substring, and
				1806	the second is set to the byte offset of the first character after the end of a
				1807	substring. <b>Note</b>: these values are always byte offsets, even in UTF-8
				1808	mode. They are not character counts.
				1809	</P>
				1810	<P>
				1811	The first pair of integers, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the
				1812	portion of the subject string matched by the entire pattern. The next pair is
				1813	used for the first capturing subpattern, and so on. The value returned by
				1814	<b>pcre_exec()</b> is one more than the highest numbered pair that has been set.
				1815	For example, if two substrings have been captured, the returned value is 3. If
				1816	there are no capturing subpatterns, the return value from a successful match is
				1817	1, indicating that just the first pair of offsets has been set.
				1818	</P>
				1819	<P>
				1820	If a capturing subpattern is matched repeatedly, it is the last portion of the
				1821	string that it matched that is returned.
				1822	</P>
				1823	<P>
				1824	If the vector is too small to hold all the captured substring offsets, it is
				1825	used as far as possible (up to two-thirds of its length), and the function
				1826	returns a value of zero. If neither the actual string matched not any captured
				1827	substrings are of interest, <b>pcre_exec()</b> may be called with <i>ovector</i>
				1828	passed as NULL and <i>ovecsize</i> as zero. However, if the pattern contains
				1829	back references and the <i>ovector</i> is not big enough to remember the related
				1830	substrings, PCRE has to get additional memory for use during matching. Thus it
				1831	is usually advisable to supply an <i>ovector</i> of reasonable size.
				1832	</P>
				1833	<P>
				1834	There are some cases where zero is returned (indicating vector overflow) when
				1835	in fact the vector is exactly the right size for the final match. For example,
				1836	consider the pattern
				1837	<pre>
				1838	(a)(?:(b)c\|bd)
				1839	</pre>
				1840	If a vector of 6 elements (allowing for only 1 captured substring) is given
				1841	with subject string "abd", <b>pcre_exec()</b> will try to set the second
				1842	captured string, thereby recording a vector overflow, before failing to match
				1843	"c" and backing up to try the second alternative. The zero return, however,
				1844	does correctly indicate that the maximum number of slots (namely 2) have been
				1845	filled. In similar cases where there is temporary overflow, but the final
				1846	number of used slots is actually less than the maximum, a non-zero value is
				1847	returned.
				1848	</P>
				1849	<P>
				1850	The <b>pcre_fullinfo()</b> function can be used to find out how many capturing
				1851	subpatterns there are in a compiled pattern. The smallest size for
				1852	<i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
				1853	the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
				1854	</P>
				1855	<P>
				1856	It is possible for capturing subpattern number <i>n+1</i> to match some part of
				1857	the subject when subpattern <i>n</i> has not been used at all. For example, if
				1858	the string "abc" is matched against the pattern (a\|(z))(bc) the return from the
				1859	function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
				1860	happens, both values in the offset pairs corresponding to unused subpatterns
				1861	are set to -1.
				1862	</P>
				1863	<P>
				1864	Offset values that correspond to unused subpatterns at the end of the
				1865	expression are also set to -1. For example, if the string "abc" is matched
				1866	against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
				1867	return from the function is 2, because the highest used capturing subpattern
				1868	number is 1, and the offsets for for the second and third capturing subpatterns
				1869	(assuming the vector is large enough, of course) are set to -1.
				1870	</P>
				1871	<P>
				1872	<b>Note</b>: Elements in the first two-thirds of <i>ovector</i> that do not
				1873	correspond to capturing parentheses in the pattern are never changed. That is,
				1874	if a pattern contains <i>n</i> capturing parentheses, no more than
				1875	<i>ovector[0]</i> to <i>ovector[2n+1]</i> are set by <b>pcre_exec()</b>. The other
				1876	elements (in the first two-thirds) retain whatever values they previously had.
				1877	</P>
				1878	<P>
				1879	Some convenience functions are provided for extracting the captured substrings
				1880	as separate strings. These are described below.
				1881	<a name="errorlist"></a></P>
				1882	<br><b>
				1883	Error return values from <b>pcre_exec()</b>
				1884	</b><br>
				1885	<P>
				1886	If <b>pcre_exec()</b> fails, it returns a negative number. The following are
				1887	defined in the header file:
				1888	<pre>
				1889	PCRE_ERROR_NOMATCH (-1)
				1890	</pre>
				1891	The subject string did not match the pattern.
				1892	<pre>
				1893	PCRE_ERROR_NULL (-2)
				1894	</pre>
				1895	Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
				1896	NULL and <i>ovecsize</i> was not zero.
				1897	<pre>
				1898	PCRE_ERROR_BADOPTION (-3)
				1899	</pre>
				1900	An unrecognized bit was set in the <i>options</i> argument.
				1901	<pre>
				1902	PCRE_ERROR_BADMAGIC (-4)
				1903	</pre>
				1904	PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
				1905	the case when it is passed a junk pointer and to detect when a pattern that was
				1906	compiled in an environment of one endianness is run in an environment with the
				1907	other endianness. This is the error that PCRE gives when the magic number is
				1908	not present.
				1909	<pre>
				1910	PCRE_ERROR_UNKNOWN_OPCODE (-5)
				1911	</pre>
				1912	While running the pattern match, an unknown item was encountered in the
				1913	compiled pattern. This error could be caused by a bug in PCRE or by overwriting
				1914	of the compiled pattern.
				1915	<pre>
				1916	PCRE_ERROR_NOMEMORY (-6)
				1917	</pre>
				1918	If a pattern contains back references, but the <i>ovector</i> that is passed to
				1919	<b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
				1920	gets a block of memory at the start of matching to use for this purpose. If the
				1921	call via <b>pcre_malloc()</b> fails, this error is given. The memory is
				1922	automatically freed at the end of matching.
				1923	</P>
				1924	<P>
				1925	This error is also given if <b>pcre_stack_malloc()</b> fails in
				1926	<b>pcre_exec()</b>. This can happen only when PCRE has been compiled with
				1927	<b>--disable-stack-for-recursion</b>.
				1928	<pre>
				1929	PCRE_ERROR_NOSUBSTRING (-7)
				1930	</pre>
				1931	This error is used by the <b>pcre_copy_substring()</b>,
				1932	<b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
				1933	below). It is never returned by <b>pcre_exec()</b>.
				1934	<pre>
				1935	PCRE_ERROR_MATCHLIMIT (-8)
				1936	</pre>
				1937	The backtracking limit, as specified by the <i>match_limit</i> field in a
				1938	<b>pcre_extra</b> structure (or defaulted) was reached. See the description
				1939	above.
				1940	<pre>
				1941	PCRE_ERROR_CALLOUT (-9)
				1942	</pre>
				1943	This error is never generated by <b>pcre_exec()</b> itself. It is provided for
				1944	use by callout functions that want to yield a distinctive error code. See the
				1945	<a href="pcrecallout.html"><b>pcrecallout</b></a>
				1946	documentation for details.
				1947	<pre>
				1948	PCRE_ERROR_BADUTF8 (-10)
				1949	</pre>
				1950	A string that contains an invalid UTF-8 byte sequence was passed as a subject,
				1951	and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
				1952	(<i>ovecsize</i>) is at least 2, the byte offset to the start of the the invalid
				1953	UTF-8 character is placed in the first element, and a reason code is placed in
				1954	the second element. The reason codes are listed in the
				1955	<a href="#badutf8reasons">following section.</a>
				1956	For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
				1957	truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
				1958	PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
				1959	<pre>
				1960	PCRE_ERROR_BADUTF8_OFFSET (-11)
				1961	</pre>
				1962	The UTF-8 byte sequence that was passed as a subject was checked and found to
				1963	be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
				1964	<i>startoffset</i> did not point to the beginning of a UTF-8 character or the
				1965	end of the subject.
				1966	<pre>
				1967	PCRE_ERROR_PARTIAL (-12)
				1968	</pre>
				1969	The subject string did not match, but it did match partially. See the
				1970	<a href="pcrepartial.html"><b>pcrepartial</b></a>
				1971	documentation for details of partial matching.
				1972	<pre>
				1973	PCRE_ERROR_BADPARTIAL (-13)
				1974	</pre>
				1975	This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
				1976	option was used with a compiled pattern containing items that were not
				1977	supported for partial matching. From release 8.00 onwards, there are no
				1978	restrictions on partial matching.
				1979	<pre>
				1980	PCRE_ERROR_INTERNAL (-14)
				1981	</pre>
				1982	An unexpected internal error has occurred. This error could be caused by a bug
				1983	in PCRE or by overwriting of the compiled pattern.
				1984	<pre>
				1985	PCRE_ERROR_BADCOUNT (-15)
				1986	</pre>
				1987	This error is given if the value of the <i>ovecsize</i> argument is negative.
				1988	<pre>
				1989	PCRE_ERROR_RECURSIONLIMIT (-21)
				1990	</pre>
				1991	The internal recursion limit, as specified by the <i>match_limit_recursion</i>
				1992	field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
				1993	description above.
				1994	<pre>
				1995	PCRE_ERROR_BADNEWLINE (-23)
				1996	</pre>
				1997	An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
				1998	<pre>
				1999	PCRE_ERROR_BADOFFSET (-24)
				2000	</pre>
				2001	The value of <i>startoffset</i> was negative or greater than the length of the
				2002	subject, that is, the value in <i>length</i>.
				2003	<pre>
				2004	PCRE_ERROR_SHORTUTF8 (-25)
				2005	</pre>
				2006	This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
				2007	ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
				2008	Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
				2009	fact sufficient to detect this case, but this special error code for
				2010	PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
				2011	retained for backwards compatibility.
				2012	<pre>
				2013	PCRE_ERROR_RECURSELOOP (-26)
				2014	</pre>
				2015	This error is returned when <b>pcre_exec()</b> detects a recursion loop within
				2016	the pattern. Specifically, it means that either the whole pattern or a
				2017	subpattern has been called recursively for the second time at the same position
				2018	in the subject string. Some simple patterns that might do this are detected and
				2019	faulted at compile time, but more complicated cases, in particular mutual
				2020	recursions between two different subpatterns, cannot be detected until run
				2021	time.
				2022	<pre>
				2023	PCRE_ERROR_JIT_STACKLIMIT (-27)
				2024	</pre>
				2025	This error is returned when a pattern that was successfully studied using the
				2026	PCRE_STUDY_JIT_COMPILE option is being matched, but the memory available for
				2027	the just-in-time processing stack is not large enough. See the
				2028	<a href="pcrejit.html"><b>pcrejit</b></a>
				2029	documentation for more details.
				2030	</P>
				2031	<P>
				2032	Error numbers -16 to -20 and -22 are not used by <b>pcre_exec()</b>.
				2033	<a name="badutf8reasons"></a></P>
				2034	<br><b>
				2035	Reason codes for invalid UTF-8 strings
				2036	</b><br>
				2037	<P>
				2038	When <b>pcre_exec()</b> returns either PCRE_ERROR_BADUTF8 or
				2039	PCRE_ERROR_SHORTUTF8, and the size of the output vector (<i>ovecsize</i>) is at
				2040	least 2, the offset of the start of the invalid UTF-8 character is placed in
				2041	the first output vector element (<i>ovector[0]</i>) and a reason code is placed
				2042	in the second element (<i>ovector[1]</i>). The reason codes are given names in
				2043	the <b>pcre.h</b> header file:
				2044	<pre>
				2045	PCRE_UTF8_ERR1
				2046	PCRE_UTF8_ERR2
				2047	PCRE_UTF8_ERR3
				2048	PCRE_UTF8_ERR4
				2049	PCRE_UTF8_ERR5
				2050	</pre>
				2051	The string ends with a truncated UTF-8 character; the code specifies how many
				2052	bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
				2053	no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
				2054	allows for up to 6 bytes, and this is checked first; hence the possibility of
				2055	4 or 5 missing bytes.
				2056	<pre>
				2057	PCRE_UTF8_ERR6
				2058	PCRE_UTF8_ERR7
				2059	PCRE_UTF8_ERR8
				2060	PCRE_UTF8_ERR9
				2061	PCRE_UTF8_ERR10
				2062	</pre>
				2063	The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
				2064	character do not have the binary value 0b10 (that is, either the most
				2065	significant bit is 0, or the next bit is 1).
				2066	<pre>
				2067	PCRE_UTF8_ERR11
				2068	PCRE_UTF8_ERR12
				2069	</pre>
				2070	A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
				2071	these code points are excluded by RFC 3629.
				2072	<pre>
				2073	PCRE_UTF8_ERR13
				2074	</pre>
				2075	A 4-byte character has a value greater than 0x10fff; these code points are
				2076	excluded by RFC 3629.
				2077	<pre>
				2078	PCRE_UTF8_ERR14
				2079	</pre>
				2080	A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
				2081	code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
				2082	from UTF-8.
				2083	<pre>
				2084	PCRE_UTF8_ERR15
				2085	PCRE_UTF8_ERR16
				2086	PCRE_UTF8_ERR17
				2087	PCRE_UTF8_ERR18
				2088	PCRE_UTF8_ERR19
				2089	</pre>
				2090	A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
				2091	value that can be represented by fewer bytes, which is invalid. For example,
				2092	the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
				2093	one byte.
				2094	<pre>
				2095	PCRE_UTF8_ERR20
				2096	</pre>
				2097	The two most significant bits of the first byte of a character have the binary
				2098	value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
				2099	byte can only validly occur as the second or subsequent byte of a multi-byte
				2100	character.
				2101	<pre>
				2102	PCRE_UTF8_ERR21
				2103	</pre>
				2104	The first byte of a character has the value 0xfe or 0xff. These values can
				2105	never occur in a valid UTF-8 string.
				2106	</P>
				2107	<br><a name="SEC17" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
				2108	<P>
				2109	<b>int pcre_copy_substring(const char <i>subject</i>, int <i>ovector</i>,</b>
				2110	<b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
				2111	<b>int <i>buffersize</i>);</b>
				2112	</P>
				2113	<P>
				2114	<b>int pcre_get_substring(const char <i>subject</i>, int <i>ovector</i>,</b>
				2115	<b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
				2116	<b>const char **<i>stringptr</i>);</b>
				2117	</P>
				2118	<P>
				2119	<b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
				2120	<b>int <i>ovector</i>, int <i>stringcount</i>, const char **<i>listptr</i>);</b>
				2121	</P>
				2122	<P>
				2123	Captured substrings can be accessed directly by using the offsets returned by
				2124	<b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
				2125	<b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
				2126	<b>pcre_get_substring_list()</b> are provided for extracting captured substrings
				2127	as new, separate, zero-terminated strings. These functions identify substrings
				2128	by number. The next section describes functions for extracting named
				2129	substrings.
				2130	</P>
				2131	<P>
				2132	A substring that contains a binary zero is correctly extracted and has a
				2133	further zero added on the end, but the result is not, of course, a C string.
				2134	However, you can process such a string by referring to the length that is
				2135	returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
				2136	Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
				2137	for handling strings containing binary zeros, because the end of the final
				2138	string is not independently indicated.
				2139	</P>
				2140	<P>
				2141	The first three arguments are the same for all three of these functions:
				2142	<i>subject</i> is the subject string that has just been successfully matched,
				2143	<i>ovector</i> is a pointer to the vector of integer offsets that was passed to
				2144	<b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
				2145	captured by the match, including the substring that matched the entire regular
				2146	expression. This is the value returned by <b>pcre_exec()</b> if it is greater
				2147	than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
				2148	space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
				2149	number of elements in the vector divided by three.
				2150	</P>
				2151	<P>
				2152	The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
				2153	extract a single substring, whose number is given as <i>stringnumber</i>. A
				2154	value of zero extracts the substring that matched the entire pattern, whereas
				2155	higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
				2156	the string is placed in <i>buffer</i>, whose length is given by
				2157	<i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
				2158	obtained via <b>pcre_malloc</b>, and its address is returned via
				2159	<i>stringptr</i>. The yield of the function is the length of the string, not
				2160	including the terminating zero, or one of these error codes:
				2161	<pre>
				2162	PCRE_ERROR_NOMEMORY (-6)
				2163	</pre>
				2164	The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
				2165	memory failed for <b>pcre_get_substring()</b>.
				2166	<pre>
				2167	PCRE_ERROR_NOSUBSTRING (-7)
				2168	</pre>
				2169	There is no substring whose number is <i>stringnumber</i>.
				2170	</P>
				2171	<P>
				2172	The <b>pcre_get_substring_list()</b> function extracts all available substrings
				2173	and builds a list of pointers to them. All this is done in a single block of
				2174	memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
				2175	is returned via <i>listptr</i>, which is also the start of the list of string
				2176	pointers. The end of the list is marked by a NULL pointer. The yield of the
				2177	function is zero if all went well, or the error code
				2178	<pre>
				2179	PCRE_ERROR_NOMEMORY (-6)
				2180	</pre>
				2181	if the attempt to get the memory block failed.
				2182	</P>
				2183	<P>
				2184	When any of these functions encounter a substring that is unset, which can
				2185	happen when capturing subpattern number <i>n+1</i> matches some part of the
				2186	subject, but subpattern <i>n</i> has not been used at all, they return an empty
				2187	string. This can be distinguished from a genuine zero-length substring by
				2188	inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
				2189	substrings.
				2190	</P>
				2191	<P>
				2192	The two convenience functions <b>pcre_free_substring()</b> and
				2193	<b>pcre_free_substring_list()</b> can be used to free the memory returned by
				2194	a previous call of <b>pcre_get_substring()</b> or
				2195	<b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
				2196	the function pointed to by <b>pcre_free</b>, which of course could be called
				2197	directly from a C program. However, PCRE is used in some situations where it is
				2198	linked via a special interface to another programming language that cannot use
				2199	<b>pcre_free</b> directly; it is for these cases that the functions are
				2200	provided.
				2201	</P>
				2202	<br><a name="SEC18" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
				2203	<P>
				2204	<b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
				2205	<b>const char *<i>name</i>);</b>
				2206	</P>
				2207	<P>
				2208	<b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
				2209	<b>const char <i>subject</i>, int <i>ovector</i>,</b>
				2210	<b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
				2211	<b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
				2212	</P>
				2213	<P>
				2214	<b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
				2215	<b>const char <i>subject</i>, int <i>ovector</i>,</b>
				2216	<b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
				2217	<b>const char **<i>stringptr</i>);</b>
				2218	</P>
				2219	<P>
				2220	To extract a substring by name, you first have to find associated number.
				2221	For example, for this pattern
				2222	<pre>
				2223	(a+)b(?<xxx>\d+)...
				2224	</pre>
				2225	the number of the subpattern called "xxx" is 2. If the name is known to be
				2226	unique (PCRE_DUPNAMES was not set), you can find the number from the name by
				2227	calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
				2228	pattern, and the second is the name. The yield of the function is the
				2229	subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
				2230	that name.
				2231	</P>
				2232	<P>
				2233	Given the number, you can extract the substring directly, or use one of the
				2234	functions described in the previous section. For convenience, there are also
				2235	two functions that do the whole job.
				2236	</P>
				2237	<P>
				2238	Most of the arguments of <b>pcre_copy_named_substring()</b> and
				2239	<b>pcre_get_named_substring()</b> are the same as those for the similarly named
				2240	functions that extract by number. As these are described in the previous
				2241	section, they are not re-described here. There are just two differences:
				2242	</P>
				2243	<P>
				2244	First, instead of a substring number, a substring name is given. Second, there
				2245	is an extra argument, given at the start, which is a pointer to the compiled
				2246	pattern. This is needed in order to gain access to the name-to-number
				2247	translation table.
				2248	</P>
				2249	<P>
				2250	These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
				2251	then call <b>pcre_copy_substring()</b> or <b>pcre_get_substring()</b>, as
				2252	appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
				2253	the behaviour may not be what you want (see the next section).
				2254	</P>
				2255	<P>
				2256	<b>Warning:</b> If the pattern uses the (?\| feature to set up multiple
				2257	subpatterns with the same number, as described in the
				2258	<a href="pcrepattern.html#dupsubpatternnumber">section on duplicate subpattern numbers</a>
				2259	in the
				2260	<a href="pcrepattern.html"><b>pcrepattern</b></a>
				2261	page, you cannot use names to distinguish the different subpatterns, because
				2262	names are not included in the compiled code. The matching process uses only
				2263	numbers. For this reason, the use of different names for subpatterns of the
				2264	same number causes an error at compile time.
				2265	</P>
				2266	<br><a name="SEC19" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
				2267	<P>
				2268	<b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
				2269	<b>const char <i>name</i>, char <i>first</i>, char *<i>last</i>);</b>
				2270	</P>
				2271	<P>
				2272	When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
				2273	are not required to be unique. (Duplicate names are always allowed for
				2274	subpatterns with the same number, created by using the (?\| feature. Indeed, if
				2275	such subpatterns are named, they are required to use the same names.)
				2276	</P>
				2277	<P>
				2278	Normally, patterns with duplicate names are such that in any one match, only
				2279	one of the named subpatterns participates. An example is shown in the
				2280	<a href="pcrepattern.html"><b>pcrepattern</b></a>
				2281	documentation.
				2282	</P>
				2283	<P>
				2284	When duplicates are present, <b>pcre_copy_named_substring()</b> and
				2285	<b>pcre_get_named_substring()</b> return the first substring corresponding to
				2286	the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
				2287	returned; no data is returned. The <b>pcre_get_stringnumber()</b> function
				2288	returns one of the numbers that are associated with the name, but it is not
				2289	defined which it is.
				2290	</P>
				2291	<P>
				2292	If you want to get full details of all captured substrings for a given name,
				2293	you must use the <b>pcre_get_stringtable_entries()</b> function. The first
				2294	argument is the compiled pattern, and the second is the name. The third and
				2295	fourth are pointers to variables which are updated by the function. After it
				2296	has run, they point to the first and last entries in the name-to-number table
				2297	for the given name. The function itself returns the length of each entry, or
				2298	PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
				2299	described above in the section entitled <i>Information about a pattern</i>
				2300	<a href="#infoaboutpattern">above.</a>
				2301	Given all the relevant entries for the name, you can extract each of their
				2302	numbers, and hence the captured data, if any.
				2303	</P>
				2304	<br><a name="SEC20" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
				2305	<P>
				2306	The traditional matching function uses a similar algorithm to Perl, which stops
				2307	when it finds the first match, starting at a given point in the subject. If you
				2308	want to find all possible matches, or the longest possible match, consider
				2309	using the alternative matching function (see below) instead. If you cannot use
				2310	the alternative function, but still need to find all possible matches, you
				2311	can kludge it up by making use of the callout facility, which is described in
				2312	the
				2313	<a href="pcrecallout.html"><b>pcrecallout</b></a>
				2314	documentation.
				2315	</P>
				2316	<P>
				2317	What you have to do is to insert a callout right at the end of the pattern.
				2318	When your callout function is called, extract and save the current matched
				2319	substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
				2320	other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
				2321	will yield PCRE_ERROR_NOMATCH.
				2322	<a name="dfamatch"></a></P>
				2323	<br><a name="SEC21" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
				2324	<P>
				2325	<b>int pcre_dfa_exec(const pcre <i>code</i>, const pcre_extra <i>extra</i>,</b>
				2326	<b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
				2327	<b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
				2328	<b>int *<i>workspace</i>, int <i>wscount</i>);</b>
				2329	</P>
				2330	<P>
				2331	The function <b>pcre_dfa_exec()</b> is called to match a subject string against
				2332	a compiled pattern, using a matching algorithm that scans the subject string
				2333	just once, and does not backtrack. This has different characteristics to the
				2334	normal algorithm, and is not compatible with Perl. Some of the features of PCRE
				2335	patterns are not supported. Nevertheless, there are times when this kind of
				2336	matching can be useful. For a discussion of the two matching algorithms, and a
				2337	list of features that <b>pcre_dfa_exec()</b> does not support, see the
				2338	<a href="pcrematching.html"><b>pcrematching</b></a>
				2339	documentation.
				2340	</P>
				2341	<P>
				2342	The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
				2343	<b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
				2344	different way, and this is described below. The other common arguments are used
				2345	in the same way as for <b>pcre_exec()</b>, so their description is not repeated
				2346	here.
				2347	</P>
				2348	<P>
				2349	The two additional arguments provide workspace for the function. The workspace
				2350	vector should contain at least 20 elements. It is used for keeping track of
				2351	multiple paths through the pattern tree. More workspace will be needed for
				2352	patterns and subjects where there are a lot of potential matches.
				2353	</P>
				2354	<P>
				2355	Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
				2356	<pre>
				2357	int rc;
				2358	int ovector[10];
				2359	int wspace[20];
				2360	rc = pcre_dfa_exec(
				2361	re, /* result of pcre_compile() */
				2362	NULL, /* we didn't study the pattern */
				2363	"some string", /* the subject string */
				2364	11, /* the length of the subject string */
				2365	0, /* start at offset 0 in the subject */
				2366	0, /* default options */
				2367	ovector, /* vector of integers for substring information */
				2368	10, /* number of elements (NOT size in bytes) */
				2369	wspace, /* working space vector */
				2370	20); /* number of elements (NOT size in bytes) */
				2371	</PRE>
				2372	</P>
				2373	<br><b>
				2374	Option bits for <b>pcre_dfa_exec()</b>
				2375	</b><br>
				2376	<P>
				2377	The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
				2378	zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
				2379	PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
				2380	PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
				2381	PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
				2382	All but the last four of these are exactly the same as for <b>pcre_exec()</b>,
				2383	so their description is not repeated here.
				2384	<pre>
				2385	PCRE_PARTIAL_HARD
				2386	PCRE_PARTIAL_SOFT
				2387	</pre>
				2388	These have the same general effect as they do for <b>pcre_exec()</b>, but the
				2389	details are slightly different. When PCRE_PARTIAL_HARD is set for
				2390	<b>pcre_dfa_exec()</b>, it returns PCRE_ERROR_PARTIAL if the end of the subject
				2391	is reached and there is still at least one matching possibility that requires
				2392	additional characters. This happens even if some complete matches have also
				2393	been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
				2394	is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
				2395	there have been no complete matches, but there is still at least one matching
				2396	possibility. The portion of the string that was inspected when the longest
				2397	partial match was found is set as the first matching string in both cases.
				2398	There is a more detailed discussion of partial and multi-segment matching, with
				2399	examples, in the
				2400	<a href="pcrepartial.html"><b>pcrepartial</b></a>
				2401	documentation.
				2402	<pre>
				2403	PCRE_DFA_SHORTEST
				2404	</pre>
				2405	Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
				2406	soon as it has found one match. Because of the way the alternative algorithm
				2407	works, this is necessarily the shortest possible match at the first possible
				2408	matching point in the subject string.
				2409	<pre>
				2410	PCRE_DFA_RESTART
				2411	</pre>
				2412	When <b>pcre_dfa_exec()</b> returns a partial match, it is possible to call it
				2413	again, with additional subject characters, and have it continue with the same
				2414	match. The PCRE_DFA_RESTART option requests this action; when it is set, the
				2415	<i>workspace</i> and <i>wscount</i> options must reference the same vector as
				2416	before because data about the match so far is left in them after a partial
				2417	match. There is more discussion of this facility in the
				2418	<a href="pcrepartial.html"><b>pcrepartial</b></a>
				2419	documentation.
				2420	</P>
				2421	<br><b>
				2422	Successful returns from <b>pcre_dfa_exec()</b>
				2423	</b><br>
				2424	<P>
				2425	When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
				2426	substring in the subject. Note, however, that all the matches from one run of
				2427	the function start at the same point in the subject. The shorter matches are
				2428	all initial substrings of the longer matches. For example, if the pattern
				2429	<pre>
				2430	<.*>
				2431	</pre>
				2432	is matched against the string
				2433	<pre>
				2434	This is <something> <something else> <something further> no more
				2435	</pre>
				2436	the three matched strings are
				2437	<pre>
				2438	<something>
				2439	<something> <something else>
				2440	<something> <something else> <something further>
				2441	</pre>
				2442	On success, the yield of the function is a number greater than zero, which is
				2443	the number of matched substrings. The substrings themselves are returned in
				2444	<i>ovector</i>. Each string uses two elements; the first is the offset to the
				2445	start, and the second is the offset to the end. In fact, all the strings have
				2446	the same start offset. (Space could have been saved by giving this only once,
				2447	but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
				2448	returns data, even though the meaning of the strings is different.)
				2449	</P>
				2450	<P>
				2451	The strings are returned in reverse order of length; that is, the longest
				2452	matching string is given first. If there were too many matches to fit into
				2453	<i>ovector</i>, the yield of the function is zero, and the vector is filled with
				2454	the longest matches. Unlike <b>pcre_exec()</b>, <b>pcre_dfa_exec()</b> can use
				2455	the entire <i>ovector</i> for returning matched strings.
				2456	</P>
				2457	<br><b>
				2458	Error returns from <b>pcre_dfa_exec()</b>
				2459	</b><br>
				2460	<P>
				2461	The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
				2462	Many of the errors are the same as for <b>pcre_exec()</b>, and these are
				2463	described
				2464	<a href="#errorlist">above.</a>
				2465	There are in addition the following errors that are specific to
				2466	<b>pcre_dfa_exec()</b>:
				2467	<pre>
				2468	PCRE_ERROR_DFA_UITEM (-16)
				2469	</pre>
				2470	This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
				2471	that it does not support, for instance, the use of \C or a back reference.
				2472	<pre>
				2473	PCRE_ERROR_DFA_UCOND (-17)
				2474	</pre>
				2475	This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
				2476	uses a back reference for the condition, or a test for recursion in a specific
				2477	group. These are not supported.
				2478	<pre>
				2479	PCRE_ERROR_DFA_UMLIMIT (-18)
				2480	</pre>
				2481	This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
				2482	block that contains a setting of the <i>match_limit</i> or
				2483	<i>match_limit_recursion</i> fields. This is not supported (these fields are
				2484	meaningless for DFA matching).
				2485	<pre>
				2486	PCRE_ERROR_DFA_WSSIZE (-19)
				2487	</pre>
				2488	This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
				2489	<i>workspace</i> vector.
				2490	<pre>
				2491	PCRE_ERROR_DFA_RECURSE (-20)
				2492	</pre>
				2493	When a recursive subpattern is processed, the matching function calls itself
				2494	recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
				2495	error is given if the output vector is not large enough. This should be
				2496	extremely rare, as a vector of size 1000 is used.
				2497	</P>
				2498	<br><a name="SEC22" href="#TOC1">SEE ALSO</a><br>
				2499	<P>
				2500	<b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
				2501	<b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
				2502	<b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
				2503	</P>
				2504	<br><a name="SEC23" href="#TOC1">AUTHOR</a><br>
				2505	<P>
				2506	Philip Hazel
				2507	<br>
				2508	University Computing Service
				2509	<br>
				2510	Cambridge CB2 3QH, England.
				2511	<br>
				2512	</P>
				2513	<br><a name="SEC24" href="#TOC1">REVISION</a><br>
				2514	<P>
				2515	Last updated: 02 December 2011
				2516	<br>
				2517	Copyright © 1997-2011 University of Cambridge.
				2518	<br>
				2519	<p>
				2520	Return to the <a href="index.html">PCRE index page</a>.
				2521	</p>