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> 8Return to the <a href="index.html">PCRE index page</a>. 9</p> 10<p> 11This page is part of the PCRE HTML documentation. It was generated automatically 12from the original man page. If there is any nonsense in it, please consult the 13man page, in case the conversion went wrong. 14<br> 15<ul> 16<li><a name="TOC1" href="#SEC1">PCRE NATIVE API</a> 17<li><a name="TOC2" href="#SEC2">PCRE API OVERVIEW</a> 18<li><a name="TOC3" href="#SEC3">NEWLINES</a> 19<li><a name="TOC4" href="#SEC4">MULTITHREADING</a> 20<li><a name="TOC5" href="#SEC5">SAVING PRECOMPILED PATTERNS FOR LATER USE</a> 21<li><a name="TOC6" href="#SEC6">CHECKING BUILD-TIME OPTIONS</a> 22<li><a name="TOC7" href="#SEC7">COMPILING A PATTERN</a> 23<li><a name="TOC8" href="#SEC8">COMPILATION ERROR CODES</a> 24<li><a name="TOC9" href="#SEC9">STUDYING A PATTERN</a> 25<li><a name="TOC10" href="#SEC10">LOCALE SUPPORT</a> 26<li><a name="TOC11" href="#SEC11">INFORMATION ABOUT A PATTERN</a> 27<li><a name="TOC12" href="#SEC12">OBSOLETE INFO FUNCTION</a> 28<li><a name="TOC13" href="#SEC13">REFERENCE COUNTS</a> 29<li><a name="TOC14" href="#SEC14">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a> 30<li><a name="TOC15" href="#SEC15">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a> 31<li><a name="TOC16" href="#SEC16">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a> 32<li><a name="TOC17" href="#SEC17">DUPLICATE SUBPATTERN NAMES</a> 33<li><a name="TOC18" href="#SEC18">FINDING ALL POSSIBLE MATCHES</a> 34<li><a name="TOC19" href="#SEC19">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a> 35<li><a name="TOC20" href="#SEC20">SEE ALSO</a> 36<li><a name="TOC21" href="#SEC21">AUTHOR</a> 37<li><a name="TOC22" href="#SEC22">REVISION</a> 38</ul> 39<br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br> 40<P> 41<b>#include <pcre.h></b> 42</P> 43<P> 44<b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b> 45<b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b> 46<b>const unsigned char *<i>tableptr</i>);</b> 47</P> 48<P> 49<b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b> 50<b>int *<i>errorcodeptr</i>,</b> 51<b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b> 52<b>const unsigned char *<i>tableptr</i>);</b> 53</P> 54<P> 55<b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b> 56<b>const char **<i>errptr</i>);</b> 57</P> 58<P> 59<b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b> 60<b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b> 61<b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b> 62</P> 63<P> 64<b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b> 65<b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b> 66<b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b> 67<b>int *<i>workspace</i>, int <i>wscount</i>);</b> 68</P> 69<P> 70<b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b> 71<b>const char *<i>subject</i>, int *<i>ovector</i>,</b> 72<b>int <i>stringcount</i>, const char *<i>stringname</i>,</b> 73<b>char *<i>buffer</i>, int <i>buffersize</i>);</b> 74</P> 75<P> 76<b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b> 77<b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b> 78<b>int <i>buffersize</i>);</b> 79</P> 80<P> 81<b>int pcre_get_named_substring(const pcre *<i>code</i>,</b> 82<b>const char *<i>subject</i>, int *<i>ovector</i>,</b> 83<b>int <i>stringcount</i>, const char *<i>stringname</i>,</b> 84<b>const char **<i>stringptr</i>);</b> 85</P> 86<P> 87<b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b> 88<b>const char *<i>name</i>);</b> 89</P> 90<P> 91<b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b> 92<b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b> 93</P> 94<P> 95<b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b> 96<b>int <i>stringcount</i>, int <i>stringnumber</i>,</b> 97<b>const char **<i>stringptr</i>);</b> 98</P> 99<P> 100<b>int pcre_get_substring_list(const char *<i>subject</i>,</b> 101<b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b> 102</P> 103<P> 104<b>void pcre_free_substring(const char *<i>stringptr</i>);</b> 105</P> 106<P> 107<b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b> 108</P> 109<P> 110<b>const unsigned char *pcre_maketables(void);</b> 111</P> 112<P> 113<b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b> 114<b>int <i>what</i>, void *<i>where</i>);</b> 115</P> 116<P> 117<b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b> 118<b>*<i>firstcharptr</i>);</b> 119</P> 120<P> 121<b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b> 122</P> 123<P> 124<b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b> 125</P> 126<P> 127<b>char *pcre_version(void);</b> 128</P> 129<P> 130<b>void *(*pcre_malloc)(size_t);</b> 131</P> 132<P> 133<b>void (*pcre_free)(void *);</b> 134</P> 135<P> 136<b>void *(*pcre_stack_malloc)(size_t);</b> 137</P> 138<P> 139<b>void (*pcre_stack_free)(void *);</b> 140</P> 141<P> 142<b>int (*pcre_callout)(pcre_callout_block *);</b> 143</P> 144<br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br> 145<P> 146PCRE has its own native API, which is described in this document. There are 147also some wrapper functions that correspond to the POSIX regular expression 148API. These are described in the 149<a href="pcreposix.html"><b>pcreposix</b></a> 150documentation. Both of these APIs define a set of C function calls. A C++ 151wrapper is distributed with PCRE. It is documented in the 152<a href="pcrecpp.html"><b>pcrecpp</b></a> 153page. 154</P> 155<P> 156The native API C function prototypes are defined in the header file 157<b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>. 158It can normally be accessed by adding <b>-lpcre</b> to the command for linking 159an application that uses PCRE. The header file defines the macros PCRE_MAJOR 160and PCRE_MINOR to contain the major and minor release numbers for the library. 161Applications can use these to include support for different releases of PCRE. 162</P> 163<P> 164The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>, 165and <b>pcre_exec()</b> are used for compiling and matching regular expressions 166in a Perl-compatible manner. A sample program that demonstrates the simplest 167way of using them is provided in the file called <i>pcredemo.c</i> in the PCRE 168source distribution. A listing of this program is given in the 169<a href="pcredemo.html"><b>pcredemo</b></a> 170documentation, and the 171<a href="pcresample.html"><b>pcresample</b></a> 172documentation describes how to compile and run it. 173</P> 174<P> 175A second matching function, <b>pcre_dfa_exec()</b>, which is not 176Perl-compatible, is also provided. This uses a different algorithm for the 177matching. The alternative algorithm finds all possible matches (at a given 178point in the subject), and scans the subject just once (unless there are 179lookbehind assertions). However, this algorithm does not return captured 180substrings. A description of the two matching algorithms and their advantages 181and disadvantages is given in the 182<a href="pcrematching.html"><b>pcrematching</b></a> 183documentation. 184</P> 185<P> 186In addition to the main compiling and matching functions, there are convenience 187functions for extracting captured substrings from a subject string that is 188matched by <b>pcre_exec()</b>. They are: 189<pre> 190 <b>pcre_copy_substring()</b> 191 <b>pcre_copy_named_substring()</b> 192 <b>pcre_get_substring()</b> 193 <b>pcre_get_named_substring()</b> 194 <b>pcre_get_substring_list()</b> 195 <b>pcre_get_stringnumber()</b> 196 <b>pcre_get_stringtable_entries()</b> 197</pre> 198<b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also 199provided, to free the memory used for extracted strings. 200</P> 201<P> 202The function <b>pcre_maketables()</b> is used to build a set of character tables 203in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>, 204or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for 205specialist use. Most commonly, no special tables are passed, in which case 206internal tables that are generated when PCRE is built are used. 207</P> 208<P> 209The function <b>pcre_fullinfo()</b> is used to find out information about a 210compiled pattern; <b>pcre_info()</b> is an obsolete version that returns only 211some of the available information, but is retained for backwards compatibility. 212The function <b>pcre_version()</b> returns a pointer to a string containing the 213version of PCRE and its date of release. 214</P> 215<P> 216The function <b>pcre_refcount()</b> maintains a reference count in a data block 217containing a compiled pattern. This is provided for the benefit of 218object-oriented applications. 219</P> 220<P> 221The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain 222the entry points of the standard <b>malloc()</b> and <b>free()</b> functions, 223respectively. PCRE calls the memory management functions via these variables, 224so a calling program can replace them if it wishes to intercept the calls. This 225should be done before calling any PCRE functions. 226</P> 227<P> 228The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also 229indirections to memory management functions. These special functions are used 230only when PCRE is compiled to use the heap for remembering data, instead of 231recursive function calls, when running the <b>pcre_exec()</b> function. See the 232<a href="pcrebuild.html"><b>pcrebuild</b></a> 233documentation for details of how to do this. It is a non-standard way of 234building PCRE, for use in environments that have limited stacks. Because of the 235greater use of memory management, it runs more slowly. Separate functions are 236provided so that special-purpose external code can be used for this case. When 237used, these functions are always called in a stack-like manner (last obtained, 238first freed), and always for memory blocks of the same size. There is a 239discussion about PCRE's stack usage in the 240<a href="pcrestack.html"><b>pcrestack</b></a> 241documentation. 242</P> 243<P> 244The global variable <b>pcre_callout</b> initially contains NULL. It can be set 245by the caller to a "callout" function, which PCRE will then call at specified 246points during a matching operation. Details are given in the 247<a href="pcrecallout.html"><b>pcrecallout</b></a> 248documentation. 249<a name="newlines"></a></P> 250<br><a name="SEC3" href="#TOC1">NEWLINES</a><br> 251<P> 252PCRE supports five different conventions for indicating line breaks in 253strings: a single CR (carriage return) character, a single LF (linefeed) 254character, the two-character sequence CRLF, any of the three preceding, or any 255Unicode newline sequence. The Unicode newline sequences are the three just 256mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed, 257U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS 258(paragraph separator, U+2029). 259</P> 260<P> 261Each of the first three conventions is used by at least one operating system as 262its standard newline sequence. When PCRE is built, a default can be specified. 263The default default is LF, which is the Unix standard. When PCRE is run, the 264default can be overridden, either when a pattern is compiled, or when it is 265matched. 266</P> 267<P> 268At compile time, the newline convention can be specified by the <i>options</i> 269argument of <b>pcre_compile()</b>, or it can be specified by special text at the 270start of the pattern itself; this overrides any other settings. See the 271<a href="pcrepattern.html"><b>pcrepattern</b></a> 272page for details of the special character sequences. 273</P> 274<P> 275In the PCRE documentation the word "newline" is used to mean "the character or 276pair of characters that indicate a line break". The choice of newline 277convention affects the handling of the dot, circumflex, and dollar 278metacharacters, the handling of #-comments in /x mode, and, when CRLF is a 279recognized line ending sequence, the match position advancement for a 280non-anchored pattern. There is more detail about this in the 281<a href="#execoptions">section on <b>pcre_exec()</b> options</a> 282below. 283</P> 284<P> 285The choice of newline convention does not affect the interpretation of 286the \n or \r escape sequences, nor does it affect what \R matches, which is 287controlled in a similar way, but by separate options. 288</P> 289<br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br> 290<P> 291The PCRE functions can be used in multi-threading applications, with the 292proviso that the memory management functions pointed to by <b>pcre_malloc</b>, 293<b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the 294callout function pointed to by <b>pcre_callout</b>, are shared by all threads. 295</P> 296<P> 297The compiled form of a regular expression is not altered during matching, so 298the same compiled pattern can safely be used by several threads at once. 299</P> 300<br><a name="SEC5" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br> 301<P> 302The compiled form of a regular expression can be saved and re-used at a later 303time, possibly by a different program, and even on a host other than the one on 304which it was compiled. Details are given in the 305<a href="pcreprecompile.html"><b>pcreprecompile</b></a> 306documentation. However, compiling a regular expression with one version of PCRE 307for use with a different version is not guaranteed to work and may cause 308crashes. 309</P> 310<br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br> 311<P> 312<b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b> 313</P> 314<P> 315The function <b>pcre_config()</b> makes it possible for a PCRE client to 316discover which optional features have been compiled into the PCRE library. The 317<a href="pcrebuild.html"><b>pcrebuild</b></a> 318documentation has more details about these optional features. 319</P> 320<P> 321The first argument for <b>pcre_config()</b> is an integer, specifying which 322information is required; the second argument is a pointer to a variable into 323which the information is placed. The following information is available: 324<pre> 325 PCRE_CONFIG_UTF8 326</pre> 327The output is an integer that is set to one if UTF-8 support is available; 328otherwise it is set to zero. 329<pre> 330 PCRE_CONFIG_UNICODE_PROPERTIES 331</pre> 332The output is an integer that is set to one if support for Unicode character 333properties is available; otherwise it is set to zero. 334<pre> 335 PCRE_CONFIG_NEWLINE 336</pre> 337The output is an integer whose value specifies the default character sequence 338that is recognized as meaning "newline". The four values that are supported 339are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. 340Though they are derived from ASCII, the same values are returned in EBCDIC 341environments. The default should normally correspond to the standard sequence 342for your operating system. 343<pre> 344 PCRE_CONFIG_BSR 345</pre> 346The output is an integer whose value indicates what character sequences the \R 347escape sequence matches by default. A value of 0 means that \R matches any 348Unicode line ending sequence; a value of 1 means that \R matches only CR, LF, 349or CRLF. The default can be overridden when a pattern is compiled or matched. 350<pre> 351 PCRE_CONFIG_LINK_SIZE 352</pre> 353The output is an integer that contains the number of bytes used for internal 354linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values 355allow larger regular expressions to be compiled, at the expense of slower 356matching. The default value of 2 is sufficient for all but the most massive 357patterns, since it allows the compiled pattern to be up to 64K in size. 358<pre> 359 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD 360</pre> 361The output is an integer that contains the threshold above which the POSIX 362interface uses <b>malloc()</b> for output vectors. Further details are given in 363the 364<a href="pcreposix.html"><b>pcreposix</b></a> 365documentation. 366<pre> 367 PCRE_CONFIG_MATCH_LIMIT 368</pre> 369The output is a long integer that gives the default limit for the number of 370internal matching function calls in a <b>pcre_exec()</b> execution. Further 371details are given with <b>pcre_exec()</b> below. 372<pre> 373 PCRE_CONFIG_MATCH_LIMIT_RECURSION 374</pre> 375The output is a long integer that gives the default limit for the depth of 376recursion when calling the internal matching function in a <b>pcre_exec()</b> 377execution. Further details are given with <b>pcre_exec()</b> below. 378<pre> 379 PCRE_CONFIG_STACKRECURSE 380</pre> 381The output is an integer that is set to one if internal recursion when running 382<b>pcre_exec()</b> is implemented by recursive function calls that use the stack 383to remember their state. This is the usual way that PCRE is compiled. The 384output is zero if PCRE was compiled to use blocks of data on the heap instead 385of recursive function calls. In this case, <b>pcre_stack_malloc</b> and 386<b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus 387avoiding the use of the stack. 388</P> 389<br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br> 390<P> 391<b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b> 392<b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b> 393<b>const unsigned char *<i>tableptr</i>);</b> 394<b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b> 395<b>int *<i>errorcodeptr</i>,</b> 396<b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b> 397<b>const unsigned char *<i>tableptr</i>);</b> 398</P> 399<P> 400Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be 401called to compile a pattern into an internal form. The only difference between 402the two interfaces is that <b>pcre_compile2()</b> has an additional argument, 403<i>errorcodeptr</i>, via which a numerical error code can be returned. To avoid 404too much repetition, we refer just to <b>pcre_compile()</b> below, but the 405information applies equally to <b>pcre_compile2()</b>. 406</P> 407<P> 408The pattern is a C string terminated by a binary zero, and is passed in the 409<i>pattern</i> argument. A pointer to a single block of memory that is obtained 410via <b>pcre_malloc</b> is returned. This contains the compiled code and related 411data. The <b>pcre</b> type is defined for the returned block; this is a typedef 412for a structure whose contents are not externally defined. It is up to the 413caller to free the memory (via <b>pcre_free</b>) when it is no longer required. 414</P> 415<P> 416Although the compiled code of a PCRE regex is relocatable, that is, it does not 417depend on memory location, the complete <b>pcre</b> data block is not 418fully relocatable, because it may contain a copy of the <i>tableptr</i> 419argument, which is an address (see below). 420</P> 421<P> 422The <i>options</i> argument contains various bit settings that affect the 423compilation. It should be zero if no options are required. The available 424options are described below. Some of them (in particular, those that are 425compatible with Perl, but some others as well) can also be set and unset from 426within the pattern (see the detailed description in the 427<a href="pcrepattern.html"><b>pcrepattern</b></a> 428documentation). For those options that can be different in different parts of 429the pattern, the contents of the <i>options</i> argument specifies their 430settings at the start of compilation and execution. The PCRE_ANCHORED, 431PCRE_BSR_<i>xxx</i>, and PCRE_NEWLINE_<i>xxx</i> options can be set at the time 432of matching as well as at compile time. 433</P> 434<P> 435If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately. 436Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns 437NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual 438error message. This is a static string that is part of the library. You must 439not try to free it. The byte offset from the start of the pattern to the 440character that was being processed when the error was discovered is placed in 441the variable pointed to by <i>erroffset</i>, which must not be NULL. If it is, 442an immediate error is given. Some errors are not detected until checks are 443carried out when the whole pattern has been scanned; in this case the offset is 444set to the end of the pattern. 445</P> 446<P> 447If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the 448<i>errorcodeptr</i> argument is not NULL, a non-zero error code number is 449returned via this argument in the event of an error. This is in addition to the 450textual error message. Error codes and messages are listed below. 451</P> 452<P> 453If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of 454character tables that are built when PCRE is compiled, using the default C 455locale. Otherwise, <i>tableptr</i> must be an address that is the result of a 456call to <b>pcre_maketables()</b>. This value is stored with the compiled 457pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is 458passed to it. For more discussion, see the section on locale support below. 459</P> 460<P> 461This code fragment shows a typical straightforward call to <b>pcre_compile()</b>: 462<pre> 463 pcre *re; 464 const char *error; 465 int erroffset; 466 re = pcre_compile( 467 "^A.*Z", /* the pattern */ 468 0, /* default options */ 469 &error, /* for error message */ 470 &erroffset, /* for error offset */ 471 NULL); /* use default character tables */ 472</pre> 473The following names for option bits are defined in the <b>pcre.h</b> header 474file: 475<pre> 476 PCRE_ANCHORED 477</pre> 478If this bit is set, the pattern is forced to be "anchored", that is, it is 479constrained to match only at the first matching point in the string that is 480being searched (the "subject string"). This effect can also be achieved by 481appropriate constructs in the pattern itself, which is the only way to do it in 482Perl. 483<pre> 484 PCRE_AUTO_CALLOUT 485</pre> 486If this bit is set, <b>pcre_compile()</b> automatically inserts callout items, 487all with number 255, before each pattern item. For discussion of the callout 488facility, see the 489<a href="pcrecallout.html"><b>pcrecallout</b></a> 490documentation. 491<pre> 492 PCRE_BSR_ANYCRLF 493 PCRE_BSR_UNICODE 494</pre> 495These options (which are mutually exclusive) control what the \R escape 496sequence matches. The choice is either to match only CR, LF, or CRLF, or to 497match any Unicode newline sequence. The default is specified when PCRE is 498built. It can be overridden from within the pattern, or by setting an option 499when a compiled pattern is matched. 500<pre> 501 PCRE_CASELESS 502</pre> 503If this bit is set, letters in the pattern match both upper and lower case 504letters. It is equivalent to Perl's /i option, and it can be changed within a 505pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the 506concept of case for characters whose values are less than 128, so caseless 507matching is always possible. For characters with higher values, the concept of 508case is supported if PCRE is compiled with Unicode property support, but not 509otherwise. If you want to use caseless matching for characters 128 and above, 510you must ensure that PCRE is compiled with Unicode property support as well as 511with UTF-8 support. 512<pre> 513 PCRE_DOLLAR_ENDONLY 514</pre> 515If this bit is set, a dollar metacharacter in the pattern matches only at the 516end of the subject string. Without this option, a dollar also matches 517immediately before a newline at the end of the string (but not before any other 518newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set. 519There is no equivalent to this option in Perl, and no way to set it within a 520pattern. 521<pre> 522 PCRE_DOTALL 523</pre> 524If this bit is set, a dot metacharater in the pattern matches all characters, 525including those that indicate newline. Without it, a dot does not match when 526the current position is at a newline. This option is equivalent to Perl's /s 527option, and it can be changed within a pattern by a (?s) option setting. A 528negative class such as [^a] always matches newline characters, independent of 529the setting of this option. 530<pre> 531 PCRE_DUPNAMES 532</pre> 533If this bit is set, names used to identify capturing subpatterns need not be 534unique. This can be helpful for certain types of pattern when it is known that 535only one instance of the named subpattern can ever be matched. There are more 536details of named subpatterns below; see also the 537<a href="pcrepattern.html"><b>pcrepattern</b></a> 538documentation. 539<pre> 540 PCRE_EXTENDED 541</pre> 542If this bit is set, whitespace data characters in the pattern are totally 543ignored except when escaped or inside a character class. Whitespace does not 544include the VT character (code 11). In addition, characters between an 545unescaped # outside a character class and the next newline, inclusive, are also 546ignored. This is equivalent to Perl's /x option, and it can be changed within a 547pattern by a (?x) option setting. 548</P> 549<P> 550This option makes it possible to include comments inside complicated patterns. 551Note, however, that this applies only to data characters. Whitespace characters 552may never appear within special character sequences in a pattern, for example 553within the sequence (?( which introduces a conditional subpattern. 554<pre> 555 PCRE_EXTRA 556</pre> 557This option was invented in order to turn on additional functionality of PCRE 558that is incompatible with Perl, but it is currently of very little use. When 559set, any backslash in a pattern that is followed by a letter that has no 560special meaning causes an error, thus reserving these combinations for future 561expansion. By default, as in Perl, a backslash followed by a letter with no 562special meaning is treated as a literal. (Perl can, however, be persuaded to 563give a warning for this.) There are at present no other features controlled by 564this option. It can also be set by a (?X) option setting within a pattern. 565<pre> 566 PCRE_FIRSTLINE 567</pre> 568If this option is set, an unanchored pattern is required to match before or at 569the first newline in the subject string, though the matched text may continue 570over the newline. 571<pre> 572 PCRE_JAVASCRIPT_COMPAT 573</pre> 574If this option is set, PCRE's behaviour is changed in some ways so that it is 575compatible with JavaScript rather than Perl. The changes are as follows: 576</P> 577<P> 578(1) A lone closing square bracket in a pattern causes a compile-time error, 579because this is illegal in JavaScript (by default it is treated as a data 580character). Thus, the pattern AB]CD becomes illegal when this option is set. 581</P> 582<P> 583(2) At run time, a back reference to an unset subpattern group matches an empty 584string (by default this causes the current matching alternative to fail). A 585pattern such as (\1)(a) succeeds when this option is set (assuming it can find 586an "a" in the subject), whereas it fails by default, for Perl compatibility. 587<pre> 588 PCRE_MULTILINE 589</pre> 590By default, PCRE treats the subject string as consisting of a single line of 591characters (even if it actually contains newlines). The "start of line" 592metacharacter (^) matches only at the start of the string, while the "end of 593line" metacharacter ($) matches only at the end of the string, or before a 594terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as 595Perl. 596</P> 597<P> 598When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs 599match immediately following or immediately before internal newlines in the 600subject string, respectively, as well as at the very start and end. This is 601equivalent to Perl's /m option, and it can be changed within a pattern by a 602(?m) option setting. If there are no newlines in a subject string, or no 603occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect. 604<pre> 605 PCRE_NEWLINE_CR 606 PCRE_NEWLINE_LF 607 PCRE_NEWLINE_CRLF 608 PCRE_NEWLINE_ANYCRLF 609 PCRE_NEWLINE_ANY 610</pre> 611These options override the default newline definition that was chosen when PCRE 612was built. Setting the first or the second specifies that a newline is 613indicated by a single character (CR or LF, respectively). Setting 614PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character 615CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three 616preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies 617that any Unicode newline sequence should be recognized. The Unicode newline 618sequences are the three just mentioned, plus the single characters VT (vertical 619tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line 620separator, U+2028), and PS (paragraph separator, U+2029). The last two are 621recognized only in UTF-8 mode. 622</P> 623<P> 624The newline setting in the options word uses three bits that are treated 625as a number, giving eight possibilities. Currently only six are used (default 626plus the five values above). This means that if you set more than one newline 627option, the combination may or may not be sensible. For example, 628PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but 629other combinations may yield unused numbers and cause an error. 630</P> 631<P> 632The only time that a line break is specially recognized when compiling a 633pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character 634class is encountered. This indicates a comment that lasts until after the next 635line break sequence. In other circumstances, line break sequences are treated 636as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated 637as whitespace characters and are therefore ignored. 638</P> 639<P> 640The newline option that is set at compile time becomes the default that is used 641for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden. 642<pre> 643 PCRE_NO_AUTO_CAPTURE 644</pre> 645If this option is set, it disables the use of numbered capturing parentheses in 646the pattern. Any opening parenthesis that is not followed by ? behaves as if it 647were followed by ?: but named parentheses can still be used for capturing (and 648they acquire numbers in the usual way). There is no equivalent of this option 649in Perl. 650<pre> 651 PCRE_UNGREEDY 652</pre> 653This option inverts the "greediness" of the quantifiers so that they are not 654greedy by default, but become greedy if followed by "?". It is not compatible 655with Perl. It can also be set by a (?U) option setting within the pattern. 656<pre> 657 PCRE_UTF8 658</pre> 659This option causes PCRE to regard both the pattern and the subject as strings 660of UTF-8 characters instead of single-byte character strings. However, it is 661available only when PCRE is built to include UTF-8 support. If not, the use 662of this option provokes an error. Details of how this option changes the 663behaviour of PCRE are given in the 664<a href="pcre.html#utf8support">section on UTF-8 support</a> 665in the main 666<a href="pcre.html"><b>pcre</b></a> 667page. 668<pre> 669 PCRE_NO_UTF8_CHECK 670</pre> 671When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is 672automatically checked. There is a discussion about the 673<a href="pcre.html#utf8strings">validity of UTF-8 strings</a> 674in the main 675<a href="pcre.html"><b>pcre</b></a> 676page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_compile()</b> 677returns an error. If you already know that your pattern is valid, and you want 678to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK 679option. When it is set, the effect of passing an invalid UTF-8 string as a 680pattern is undefined. It may cause your program to crash. Note that this option 681can also be passed to <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, to suppress 682the UTF-8 validity checking of subject strings. 683</P> 684<br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br> 685<P> 686The following table lists the error codes than may be returned by 687<b>pcre_compile2()</b>, along with the error messages that may be returned by 688both compiling functions. As PCRE has developed, some error codes have fallen 689out of use. To avoid confusion, they have not been re-used. 690<pre> 691 0 no error 692 1 \ at end of pattern 693 2 \c at end of pattern 694 3 unrecognized character follows \ 695 4 numbers out of order in {} quantifier 696 5 number too big in {} quantifier 697 6 missing terminating ] for character class 698 7 invalid escape sequence in character class 699 8 range out of order in character class 700 9 nothing to repeat 701 10 [this code is not in use] 702 11 internal error: unexpected repeat 703 12 unrecognized character after (? or (?- 704 13 POSIX named classes are supported only within a class 705 14 missing ) 706 15 reference to non-existent subpattern 707 16 erroffset passed as NULL 708 17 unknown option bit(s) set 709 18 missing ) after comment 710 19 [this code is not in use] 711 20 regular expression is too large 712 21 failed to get memory 713 22 unmatched parentheses 714 23 internal error: code overflow 715 24 unrecognized character after (?< 716 25 lookbehind assertion is not fixed length 717 26 malformed number or name after (?( 718 27 conditional group contains more than two branches 719 28 assertion expected after (?( 720 29 (?R or (?[+-]digits must be followed by ) 721 30 unknown POSIX class name 722 31 POSIX collating elements are not supported 723 32 this version of PCRE is not compiled with PCRE_UTF8 support 724 33 [this code is not in use] 725 34 character value in \x{...} sequence is too large 726 35 invalid condition (?(0) 727 36 \C not allowed in lookbehind assertion 728 37 PCRE does not support \L, \l, \N, \U, or \u 729 38 number after (?C is > 255 730 39 closing ) for (?C expected 731 40 recursive call could loop indefinitely 732 41 unrecognized character after (?P 733 42 syntax error in subpattern name (missing terminator) 734 43 two named subpatterns have the same name 735 44 invalid UTF-8 string 736 45 support for \P, \p, and \X has not been compiled 737 46 malformed \P or \p sequence 738 47 unknown property name after \P or \p 739 48 subpattern name is too long (maximum 32 characters) 740 49 too many named subpatterns (maximum 10000) 741 50 [this code is not in use] 742 51 octal value is greater than \377 (not in UTF-8 mode) 743 52 internal error: overran compiling workspace 744 53 internal error: previously-checked referenced subpattern not found 745 54 DEFINE group contains more than one branch 746 55 repeating a DEFINE group is not allowed 747 56 inconsistent NEWLINE options 748 57 \g is not followed by a braced, angle-bracketed, or quoted 749 name/number or by a plain number 750 58 a numbered reference must not be zero 751 59 (*VERB) with an argument is not supported 752 60 (*VERB) not recognized 753 61 number is too big 754 62 subpattern name expected 755 63 digit expected after (?+ 756 64 ] is an invalid data character in JavaScript compatibility mode 757</pre> 758The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may 759be used if the limits were changed when PCRE was built. 760</P> 761<br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br> 762<P> 763<b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b> 764<b>const char **<i>errptr</i>);</b> 765</P> 766<P> 767If a compiled pattern is going to be used several times, it is worth spending 768more time analyzing it in order to speed up the time taken for matching. The 769function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first 770argument. If studying the pattern produces additional information that will 771help speed up matching, <b>pcre_study()</b> returns a pointer to a 772<b>pcre_extra</b> block, in which the <i>study_data</i> field points to the 773results of the study. 774</P> 775<P> 776The returned value from <b>pcre_study()</b> can be passed directly to 777<b>pcre_exec()</b> or <b>pcre_dfa_exec()</b>. However, a <b>pcre_extra</b> block 778also contains other fields that can be set by the caller before the block is 779passed; these are described 780<a href="#extradata">below</a> 781in the section on matching a pattern. 782</P> 783<P> 784If studying the pattern does not produce any useful information, 785<b>pcre_study()</b> returns NULL. In that circumstance, if the calling program 786wants to pass any of the other fields to <b>pcre_exec()</b> or 787<b>pcre_dfa_exec()</b>, it must set up its own <b>pcre_extra</b> block. 788</P> 789<P> 790The second argument of <b>pcre_study()</b> contains option bits. At present, no 791options are defined, and this argument should always be zero. 792</P> 793<P> 794The third argument for <b>pcre_study()</b> is a pointer for an error message. If 795studying succeeds (even if no data is returned), the variable it points to is 796set to NULL. Otherwise it is set to point to a textual error message. This is a 797static string that is part of the library. You must not try to free it. You 798should test the error pointer for NULL after calling <b>pcre_study()</b>, to be 799sure that it has run successfully. 800</P> 801<P> 802This is a typical call to <b>pcre_study</b>(): 803<pre> 804 pcre_extra *pe; 805 pe = pcre_study( 806 re, /* result of pcre_compile() */ 807 0, /* no options exist */ 808 &error); /* set to NULL or points to a message */ 809</pre> 810Studying a pattern does two things: first, a lower bound for the length of 811subject string that is needed to match the pattern is computed. This does not 812mean that there are any strings of that length that match, but it does 813guarantee that no shorter strings match. The value is used by 814<b>pcre_exec()</b> and <b>pcre_dfa_exec()</b> to avoid wasting time by trying to 815match strings that are shorter than the lower bound. You can find out the value 816in a calling program via the <b>pcre_fullinfo()</b> function. 817</P> 818<P> 819Studying a pattern is also useful for non-anchored patterns that do not have a 820single fixed starting character. A bitmap of possible starting bytes is 821created. This speeds up finding a position in the subject at which to start 822matching. 823<a name="localesupport"></a></P> 824<br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br> 825<P> 826PCRE handles caseless matching, and determines whether characters are letters, 827digits, or whatever, by reference to a set of tables, indexed by character 828value. When running in UTF-8 mode, this applies only to characters with codes 829less than 128. Higher-valued codes never match escapes such as \w or \d, but 830can be tested with \p if PCRE is built with Unicode character property 831support. The use of locales with Unicode is discouraged. If you are handling 832characters with codes greater than 128, you should either use UTF-8 and 833Unicode, or use locales, but not try to mix the two. 834</P> 835<P> 836PCRE contains an internal set of tables that are used when the final argument 837of <b>pcre_compile()</b> is NULL. These are sufficient for many applications. 838Normally, the internal tables recognize only ASCII characters. However, when 839PCRE is built, it is possible to cause the internal tables to be rebuilt in the 840default "C" locale of the local system, which may cause them to be different. 841</P> 842<P> 843The internal tables can always be overridden by tables supplied by the 844application that calls PCRE. These may be created in a different locale from 845the default. As more and more applications change to using Unicode, the need 846for this locale support is expected to die away. 847</P> 848<P> 849External tables are built by calling the <b>pcre_maketables()</b> function, 850which has no arguments, in the relevant locale. The result can then be passed 851to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For 852example, to build and use tables that are appropriate for the French locale 853(where accented characters with values greater than 128 are treated as letters), 854the following code could be used: 855<pre> 856 setlocale(LC_CTYPE, "fr_FR"); 857 tables = pcre_maketables(); 858 re = pcre_compile(..., tables); 859</pre> 860The locale name "fr_FR" is used on Linux and other Unix-like systems; if you 861are using Windows, the name for the French locale is "french". 862</P> 863<P> 864When <b>pcre_maketables()</b> runs, the tables are built in memory that is 865obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure 866that the memory containing the tables remains available for as long as it is 867needed. 868</P> 869<P> 870The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled 871pattern, and the same tables are used via this pointer by <b>pcre_study()</b> 872and normally also by <b>pcre_exec()</b>. Thus, by default, for any single 873pattern, compilation, studying and matching all happen in the same locale, but 874different patterns can be compiled in different locales. 875</P> 876<P> 877It is possible to pass a table pointer or NULL (indicating the use of the 878internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose, 879this facility could be used to match a pattern in a different locale from the 880one in which it was compiled. Passing table pointers at run time is discussed 881below in the section on matching a pattern. 882</P> 883<br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br> 884<P> 885<b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b> 886<b>int <i>what</i>, void *<i>where</i>);</b> 887</P> 888<P> 889The <b>pcre_fullinfo()</b> function returns information about a compiled 890pattern. It replaces the obsolete <b>pcre_info()</b> function, which is 891nevertheless retained for backwards compability (and is documented below). 892</P> 893<P> 894The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled 895pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if 896the pattern was not studied. The third argument specifies which piece of 897information is required, and the fourth argument is a pointer to a variable 898to receive the data. The yield of the function is zero for success, or one of 899the following negative numbers: 900<pre> 901 PCRE_ERROR_NULL the argument <i>code</i> was NULL 902 the argument <i>where</i> was NULL 903 PCRE_ERROR_BADMAGIC the "magic number" was not found 904 PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid 905</pre> 906The "magic number" is placed at the start of each compiled pattern as an simple 907check against passing an arbitrary memory pointer. Here is a typical call of 908<b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern: 909<pre> 910 int rc; 911 size_t length; 912 rc = pcre_fullinfo( 913 re, /* result of pcre_compile() */ 914 pe, /* result of pcre_study(), or NULL */ 915 PCRE_INFO_SIZE, /* what is required */ 916 &length); /* where to put the data */ 917</pre> 918The possible values for the third argument are defined in <b>pcre.h</b>, and are 919as follows: 920<pre> 921 PCRE_INFO_BACKREFMAX 922</pre> 923Return the number of the highest back reference in the pattern. The fourth 924argument should point to an <b>int</b> variable. Zero is returned if there are 925no back references. 926<pre> 927 PCRE_INFO_CAPTURECOUNT 928</pre> 929Return the number of capturing subpatterns in the pattern. The fourth argument 930should point to an <b>int</b> variable. 931<pre> 932 PCRE_INFO_DEFAULT_TABLES 933</pre> 934Return a pointer to the internal default character tables within PCRE. The 935fourth argument should point to an <b>unsigned char *</b> variable. This 936information call is provided for internal use by the <b>pcre_study()</b> 937function. External callers can cause PCRE to use its internal tables by passing 938a NULL table pointer. 939<pre> 940 PCRE_INFO_FIRSTBYTE 941</pre> 942Return information about the first byte of any matched string, for a 943non-anchored pattern. The fourth argument should point to an <b>int</b> 944variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is 945still recognized for backwards compatibility.) 946</P> 947<P> 948If there is a fixed first byte, for example, from a pattern such as 949(cat|cow|coyote), its value is returned. Otherwise, if either 950<br> 951<br> 952(a) the pattern was compiled with the PCRE_MULTILINE option, and every branch 953starts with "^", or 954<br> 955<br> 956(b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set 957(if it were set, the pattern would be anchored), 958<br> 959<br> 960-1 is returned, indicating that the pattern matches only at the start of a 961subject string or after any newline within the string. Otherwise -2 is 962returned. For anchored patterns, -2 is returned. 963<pre> 964 PCRE_INFO_FIRSTTABLE 965</pre> 966If the pattern was studied, and this resulted in the construction of a 256-bit 967table indicating a fixed set of bytes for the first byte in any matching 968string, a pointer to the table is returned. Otherwise NULL is returned. The 969fourth argument should point to an <b>unsigned char *</b> variable. 970<pre> 971 PCRE_INFO_HASCRORLF 972</pre> 973Return 1 if the pattern contains any explicit matches for CR or LF characters, 974otherwise 0. The fourth argument should point to an <b>int</b> variable. An 975explicit match is either a literal CR or LF character, or \r or \n. 976<pre> 977 PCRE_INFO_JCHANGED 978</pre> 979Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise 9800. The fourth argument should point to an <b>int</b> variable. (?J) and 981(?-J) set and unset the local PCRE_DUPNAMES option, respectively. 982<pre> 983 PCRE_INFO_LASTLITERAL 984</pre> 985Return the value of the rightmost literal byte that must exist in any matched 986string, other than at its start, if such a byte has been recorded. The fourth 987argument should point to an <b>int</b> variable. If there is no such byte, -1 is 988returned. For anchored patterns, a last literal byte is recorded only if it 989follows something of variable length. For example, for the pattern 990/^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value 991is -1. 992<pre> 993 PCRE_INFO_MINLENGTH 994</pre> 995If the pattern was studied and a minimum length for matching subject strings 996was computed, its value is returned. Otherwise the returned value is -1. The 997value is a number of characters, not bytes (this may be relevant in UTF-8 998mode). The fourth argument should point to an <b>int</b> variable. A 999non-negative value is a lower bound to the length of any matching string. There 1000may not be any strings of that length that do actually match, but every string 1001that does match is at least that long. 1002<pre> 1003 PCRE_INFO_NAMECOUNT 1004 PCRE_INFO_NAMEENTRYSIZE 1005 PCRE_INFO_NAMETABLE 1006</pre> 1007PCRE supports the use of named as well as numbered capturing parentheses. The 1008names are just an additional way of identifying the parentheses, which still 1009acquire numbers. Several convenience functions such as 1010<b>pcre_get_named_substring()</b> are provided for extracting captured 1011substrings by name. It is also possible to extract the data directly, by first 1012converting the name to a number in order to access the correct pointers in the 1013output vector (described with <b>pcre_exec()</b> below). To do the conversion, 1014you need to use the name-to-number map, which is described by these three 1015values. 1016</P> 1017<P> 1018The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives 1019the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each 1020entry; both of these return an <b>int</b> value. The entry size depends on the 1021length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first 1022entry of the table (a pointer to <b>char</b>). The first two bytes of each entry 1023are the number of the capturing parenthesis, most significant byte first. The 1024rest of the entry is the corresponding name, zero terminated. 1025</P> 1026<P> 1027The names are in alphabetical order. Duplicate names may appear if (?| is used 1028to create multiple groups with the same number, as described in the 1029<a href="pcrepattern.html#dupsubpatternnumber">section on duplicate subpattern numbers</a> 1030in the 1031<a href="pcrepattern.html"><b>pcrepattern</b></a> 1032page. Duplicate names for subpatterns with different numbers are permitted only 1033if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the 1034table in the order in which they were found in the pattern. In the absence of 1035(?| this is the order of increasing number; when (?| is used this is not 1036necessarily the case because later subpatterns may have lower numbers. 1037</P> 1038<P> 1039As a simple example of the name/number table, consider the following pattern 1040(assume PCRE_EXTENDED is set, so white space - including newlines - is 1041ignored): 1042<pre> 1043 (?<date> (?<year>(\d\d)?\d\d) - (?<month>\d\d) - (?<day>\d\d) ) 1044</pre> 1045There are four named subpatterns, so the table has four entries, and each entry 1046in the table is eight bytes long. The table is as follows, with non-printing 1047bytes shows in hexadecimal, and undefined bytes shown as ??: 1048<pre> 1049 00 01 d a t e 00 ?? 1050 00 05 d a y 00 ?? ?? 1051 00 04 m o n t h 00 1052 00 02 y e a r 00 ?? 1053</pre> 1054When writing code to extract data from named subpatterns using the 1055name-to-number map, remember that the length of the entries is likely to be 1056different for each compiled pattern. 1057<pre> 1058 PCRE_INFO_OKPARTIAL 1059</pre> 1060Return 1 if the pattern can be used for partial matching with 1061<b>pcre_exec()</b>, otherwise 0. The fourth argument should point to an 1062<b>int</b> variable. From release 8.00, this always returns 1, because the 1063restrictions that previously applied to partial matching have been lifted. The 1064<a href="pcrepartial.html"><b>pcrepartial</b></a> 1065documentation gives details of partial matching. 1066<pre> 1067 PCRE_INFO_OPTIONS 1068</pre> 1069Return a copy of the options with which the pattern was compiled. The fourth 1070argument should point to an <b>unsigned long int</b> variable. These option bits 1071are those specified in the call to <b>pcre_compile()</b>, modified by any 1072top-level option settings at the start of the pattern itself. In other words, 1073they are the options that will be in force when matching starts. For example, 1074if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the 1075result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED. 1076</P> 1077<P> 1078A pattern is automatically anchored by PCRE if all of its top-level 1079alternatives begin with one of the following: 1080<pre> 1081 ^ unless PCRE_MULTILINE is set 1082 \A always 1083 \G always 1084 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears 1085</pre> 1086For such patterns, the PCRE_ANCHORED bit is set in the options returned by 1087<b>pcre_fullinfo()</b>. 1088<pre> 1089 PCRE_INFO_SIZE 1090</pre> 1091Return the size of the compiled pattern, that is, the value that was passed as 1092the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to 1093place the compiled data. The fourth argument should point to a <b>size_t</b> 1094variable. 1095<pre> 1096 PCRE_INFO_STUDYSIZE 1097</pre> 1098Return the size of the data block pointed to by the <i>study_data</i> field in 1099a <b>pcre_extra</b> block. That is, it is the value that was passed to 1100<b>pcre_malloc()</b> when PCRE was getting memory into which to place the data 1101created by <b>pcre_study()</b>. If <b>pcre_extra</b> is NULL, or there is no 1102study data, zero is returned. The fourth argument should point to a 1103<b>size_t</b> variable. 1104</P> 1105<br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br> 1106<P> 1107<b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b> 1108<b>*<i>firstcharptr</i>);</b> 1109</P> 1110<P> 1111The <b>pcre_info()</b> function is now obsolete because its interface is too 1112restrictive to return all the available data about a compiled pattern. New 1113programs should use <b>pcre_fullinfo()</b> instead. The yield of 1114<b>pcre_info()</b> is the number of capturing subpatterns, or one of the 1115following negative numbers: 1116<pre> 1117 PCRE_ERROR_NULL the argument <i>code</i> was NULL 1118 PCRE_ERROR_BADMAGIC the "magic number" was not found 1119</pre> 1120If the <i>optptr</i> argument is not NULL, a copy of the options with which the 1121pattern was compiled is placed in the integer it points to (see 1122PCRE_INFO_OPTIONS above). 1123</P> 1124<P> 1125If the pattern is not anchored and the <i>firstcharptr</i> argument is not NULL, 1126it is used to pass back information about the first character of any matched 1127string (see PCRE_INFO_FIRSTBYTE above). 1128</P> 1129<br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br> 1130<P> 1131<b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b> 1132</P> 1133<P> 1134The <b>pcre_refcount()</b> function is used to maintain a reference count in the 1135data block that contains a compiled pattern. It is provided for the benefit of 1136applications that operate in an object-oriented manner, where different parts 1137of the application may be using the same compiled pattern, but you want to free 1138the block when they are all done. 1139</P> 1140<P> 1141When a pattern is compiled, the reference count field is initialized to zero. 1142It is changed only by calling this function, whose action is to add the 1143<i>adjust</i> value (which may be positive or negative) to it. The yield of the 1144function is the new value. However, the value of the count is constrained to 1145lie between 0 and 65535, inclusive. If the new value is outside these limits, 1146it is forced to the appropriate limit value. 1147</P> 1148<P> 1149Except when it is zero, the reference count is not correctly preserved if a 1150pattern is compiled on one host and then transferred to a host whose byte-order 1151is different. (This seems a highly unlikely scenario.) 1152</P> 1153<br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br> 1154<P> 1155<b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b> 1156<b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b> 1157<b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b> 1158</P> 1159<P> 1160The function <b>pcre_exec()</b> is called to match a subject string against a 1161compiled pattern, which is passed in the <i>code</i> argument. If the 1162pattern was studied, the result of the study should be passed in the 1163<i>extra</i> argument. This function is the main matching facility of the 1164library, and it operates in a Perl-like manner. For specialist use there is 1165also an alternative matching function, which is described 1166<a href="#dfamatch">below</a> 1167in the section about the <b>pcre_dfa_exec()</b> function. 1168</P> 1169<P> 1170In most applications, the pattern will have been compiled (and optionally 1171studied) in the same process that calls <b>pcre_exec()</b>. However, it is 1172possible to save compiled patterns and study data, and then use them later 1173in different processes, possibly even on different hosts. For a discussion 1174about this, see the 1175<a href="pcreprecompile.html"><b>pcreprecompile</b></a> 1176documentation. 1177</P> 1178<P> 1179Here is an example of a simple call to <b>pcre_exec()</b>: 1180<pre> 1181 int rc; 1182 int ovector[30]; 1183 rc = pcre_exec( 1184 re, /* result of pcre_compile() */ 1185 NULL, /* we didn't study the pattern */ 1186 "some string", /* the subject string */ 1187 11, /* the length of the subject string */ 1188 0, /* start at offset 0 in the subject */ 1189 0, /* default options */ 1190 ovector, /* vector of integers for substring information */ 1191 30); /* number of elements (NOT size in bytes) */ 1192<a name="extradata"></a></PRE> 1193</P> 1194<br><b> 1195Extra data for <b>pcre_exec()</b> 1196</b><br> 1197<P> 1198If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b> 1199data block. The <b>pcre_study()</b> function returns such a block (when it 1200doesn't return NULL), but you can also create one for yourself, and pass 1201additional information in it. The <b>pcre_extra</b> block contains the following 1202fields (not necessarily in this order): 1203<pre> 1204 unsigned long int <i>flags</i>; 1205 void *<i>study_data</i>; 1206 unsigned long int <i>match_limit</i>; 1207 unsigned long int <i>match_limit_recursion</i>; 1208 void *<i>callout_data</i>; 1209 const unsigned char *<i>tables</i>; 1210</pre> 1211The <i>flags</i> field is a bitmap that specifies which of the other fields 1212are set. The flag bits are: 1213<pre> 1214 PCRE_EXTRA_STUDY_DATA 1215 PCRE_EXTRA_MATCH_LIMIT 1216 PCRE_EXTRA_MATCH_LIMIT_RECURSION 1217 PCRE_EXTRA_CALLOUT_DATA 1218 PCRE_EXTRA_TABLES 1219</pre> 1220Other flag bits should be set to zero. The <i>study_data</i> field is set in the 1221<b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with 1222the appropriate flag bit. You should not set this yourself, but you may add to 1223the block by setting the other fields and their corresponding flag bits. 1224</P> 1225<P> 1226The <i>match_limit</i> field provides a means of preventing PCRE from using up a 1227vast amount of resources when running patterns that are not going to match, 1228but which have a very large number of possibilities in their search trees. The 1229classic example is a pattern that uses nested unlimited repeats. 1230</P> 1231<P> 1232Internally, PCRE uses a function called <b>match()</b> which it calls repeatedly 1233(sometimes recursively). The limit set by <i>match_limit</i> is imposed on the 1234number of times this function is called during a match, which has the effect of 1235limiting the amount of backtracking that can take place. For patterns that are 1236not anchored, the count restarts from zero for each position in the subject 1237string. 1238</P> 1239<P> 1240The default value for the limit can be set when PCRE is built; the default 1241default is 10 million, which handles all but the most extreme cases. You can 1242override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b> 1243block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in 1244the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns 1245PCRE_ERROR_MATCHLIMIT. 1246</P> 1247<P> 1248The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but 1249instead of limiting the total number of times that <b>match()</b> is called, it 1250limits the depth of recursion. The recursion depth is a smaller number than the 1251total number of calls, because not all calls to <b>match()</b> are recursive. 1252This limit is of use only if it is set smaller than <i>match_limit</i>. 1253</P> 1254<P> 1255Limiting the recursion depth limits the amount of stack that can be used, or, 1256when PCRE has been compiled to use memory on the heap instead of the stack, the 1257amount of heap memory that can be used. 1258</P> 1259<P> 1260The default value for <i>match_limit_recursion</i> can be set when PCRE is 1261built; the default default is the same value as the default for 1262<i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b> 1263with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and 1264PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit 1265is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT. 1266</P> 1267<P> 1268The <i>callout_data</i> field is used in conjunction with the "callout" feature, 1269and is described in the 1270<a href="pcrecallout.html"><b>pcrecallout</b></a> 1271documentation. 1272</P> 1273<P> 1274The <i>tables</i> field is used to pass a character tables pointer to 1275<b>pcre_exec()</b>; this overrides the value that is stored with the compiled 1276pattern. A non-NULL value is stored with the compiled pattern only if custom 1277tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument. 1278If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's 1279internal tables to be used. This facility is helpful when re-using patterns 1280that have been saved after compiling with an external set of tables, because 1281the external tables might be at a different address when <b>pcre_exec()</b> is 1282called. See the 1283<a href="pcreprecompile.html"><b>pcreprecompile</b></a> 1284documentation for a discussion of saving compiled patterns for later use. 1285<a name="execoptions"></a></P> 1286<br><b> 1287Option bits for <b>pcre_exec()</b> 1288</b><br> 1289<P> 1290The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be 1291zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>, 1292PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART, 1293PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and 1294PCRE_PARTIAL_HARD. 1295<pre> 1296 PCRE_ANCHORED 1297</pre> 1298The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first 1299matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out 1300to be anchored by virtue of its contents, it cannot be made unachored at 1301matching time. 1302<pre> 1303 PCRE_BSR_ANYCRLF 1304 PCRE_BSR_UNICODE 1305</pre> 1306These options (which are mutually exclusive) control what the \R escape 1307sequence matches. The choice is either to match only CR, LF, or CRLF, or to 1308match any Unicode newline sequence. These options override the choice that was 1309made or defaulted when the pattern was compiled. 1310<pre> 1311 PCRE_NEWLINE_CR 1312 PCRE_NEWLINE_LF 1313 PCRE_NEWLINE_CRLF 1314 PCRE_NEWLINE_ANYCRLF 1315 PCRE_NEWLINE_ANY 1316</pre> 1317These options override the newline definition that was chosen or defaulted when 1318the pattern was compiled. For details, see the description of 1319<b>pcre_compile()</b> above. During matching, the newline choice affects the 1320behaviour of the dot, circumflex, and dollar metacharacters. It may also alter 1321the way the match position is advanced after a match failure for an unanchored 1322pattern. 1323</P> 1324<P> 1325When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a 1326match attempt for an unanchored pattern fails when the current position is at a 1327CRLF sequence, and the pattern contains no explicit matches for CR or LF 1328characters, the match position is advanced by two characters instead of one, in 1329other words, to after the CRLF. 1330</P> 1331<P> 1332The above rule is a compromise that makes the most common cases work as 1333expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not 1334set), it does not match the string "\r\nA" because, after failing at the 1335start, it skips both the CR and the LF before retrying. However, the pattern 1336[\r\n]A does match that string, because it contains an explicit CR or LF 1337reference, and so advances only by one character after the first failure. 1338</P> 1339<P> 1340An explicit match for CR of LF is either a literal appearance of one of those 1341characters, or one of the \r or \n escape sequences. Implicit matches such as 1342[^X] do not count, nor does \s (which includes CR and LF in the characters 1343that it matches). 1344</P> 1345<P> 1346Notwithstanding the above, anomalous effects may still occur when CRLF is a 1347valid newline sequence and explicit \r or \n escapes appear in the pattern. 1348<pre> 1349 PCRE_NOTBOL 1350</pre> 1351This option specifies that first character of the subject string is not the 1352beginning of a line, so the circumflex metacharacter should not match before 1353it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex 1354never to match. This option affects only the behaviour of the circumflex 1355metacharacter. It does not affect \A. 1356<pre> 1357 PCRE_NOTEOL 1358</pre> 1359This option specifies that the end of the subject string is not the end of a 1360line, so the dollar metacharacter should not match it nor (except in multiline 1361mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at 1362compile time) causes dollar never to match. This option affects only the 1363behaviour of the dollar metacharacter. It does not affect \Z or \z. 1364<pre> 1365 PCRE_NOTEMPTY 1366</pre> 1367An empty string is not considered to be a valid match if this option is set. If 1368there are alternatives in the pattern, they are tried. If all the alternatives 1369match the empty string, the entire match fails. For example, if the pattern 1370<pre> 1371 a?b? 1372</pre> 1373is applied to a string not beginning with "a" or "b", it matches an empty 1374string at the start of the subject. With PCRE_NOTEMPTY set, this match is not 1375valid, so PCRE searches further into the string for occurrences of "a" or "b". 1376<pre> 1377 PCRE_NOTEMPTY_ATSTART 1378</pre> 1379This is like PCRE_NOTEMPTY, except that an empty string match that is not at 1380the start of the subject is permitted. If the pattern is anchored, such a match 1381can occur only if the pattern contains \K. 1382</P> 1383<P> 1384Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it 1385does make a special case of a pattern match of the empty string within its 1386<b>split()</b> function, and when using the /g modifier. It is possible to 1387emulate Perl's behaviour after matching a null string by first trying the match 1388again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then 1389if that fails, by advancing the starting offset (see below) and trying an 1390ordinary match again. There is some code that demonstrates how to do this in 1391the 1392<a href="pcredemo.html"><b>pcredemo</b></a> 1393sample program. 1394<pre> 1395 PCRE_NO_START_OPTIMIZE 1396</pre> 1397There are a number of optimizations that <b>pcre_exec()</b> uses at the start of 1398a match, in order to speed up the process. For example, if it is known that a 1399match must start with a specific character, it searches the subject for that 1400character, and fails immediately if it cannot find it, without actually running 1401the main matching function. When callouts are in use, these optimizations can 1402cause them to be skipped. This option disables the "start-up" optimizations, 1403causing performance to suffer, but ensuring that the callouts do occur. 1404<pre> 1405 PCRE_NO_UTF8_CHECK 1406</pre> 1407When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8 1408string is automatically checked when <b>pcre_exec()</b> is subsequently called. 1409The value of <i>startoffset</i> is also checked to ensure that it points to the 1410start of a UTF-8 character. There is a discussion about the validity of UTF-8 1411strings in the 1412<a href="pcre.html#utf8strings">section on UTF-8 support</a> 1413in the main 1414<a href="pcre.html"><b>pcre</b></a> 1415page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_exec()</b> returns 1416the error PCRE_ERROR_BADUTF8. If <i>startoffset</i> contains an invalid value, 1417PCRE_ERROR_BADUTF8_OFFSET is returned. 1418</P> 1419<P> 1420If you already know that your subject is valid, and you want to skip these 1421checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when 1422calling <b>pcre_exec()</b>. You might want to do this for the second and 1423subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find 1424all the matches in a single subject string. However, you should be sure that 1425the value of <i>startoffset</i> points to the start of a UTF-8 character. When 1426PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a 1427subject, or a value of <i>startoffset</i> that does not point to the start of a 1428UTF-8 character, is undefined. Your program may crash. 1429<pre> 1430 PCRE_PARTIAL_HARD 1431 PCRE_PARTIAL_SOFT 1432</pre> 1433These options turn on the partial matching feature. For backwards 1434compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match 1435occurs if the end of the subject string is reached successfully, but there are 1436not enough subject characters to complete the match. If this happens when 1437PCRE_PARTIAL_HARD is set, <b>pcre_exec()</b> immediately returns 1438PCRE_ERROR_PARTIAL. Otherwise, if PCRE_PARTIAL_SOFT is set, matching continues 1439by testing any other alternatives. Only if they all fail is PCRE_ERROR_PARTIAL 1440returned (instead of PCRE_ERROR_NOMATCH). The portion of the string that 1441was inspected when the partial match was found is set as the first matching 1442string. There is a more detailed discussion in the 1443<a href="pcrepartial.html"><b>pcrepartial</b></a> 1444documentation. 1445</P> 1446<br><b> 1447The string to be matched by <b>pcre_exec()</b> 1448</b><br> 1449<P> 1450The subject string is passed to <b>pcre_exec()</b> as a pointer in 1451<i>subject</i>, a length (in bytes) in <i>length</i>, and a starting byte offset 1452in <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of 1453a UTF-8 character. Unlike the pattern string, the subject may contain binary 1454zero bytes. When the starting offset is zero, the search for a match starts at 1455the beginning of the subject, and this is by far the most common case. 1456</P> 1457<P> 1458A non-zero starting offset is useful when searching for another match in the 1459same subject by calling <b>pcre_exec()</b> again after a previous success. 1460Setting <i>startoffset</i> differs from just passing over a shortened string and 1461setting PCRE_NOTBOL in the case of a pattern that begins with any kind of 1462lookbehind. For example, consider the pattern 1463<pre> 1464 \Biss\B 1465</pre> 1466which finds occurrences of "iss" in the middle of words. (\B matches only if 1467the current position in the subject is not a word boundary.) When applied to 1468the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first 1469occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the 1470subject, namely "issipi", it does not match, because \B is always false at the 1471start of the subject, which is deemed to be a word boundary. However, if 1472<b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i> 1473set to 4, it finds the second occurrence of "iss" because it is able to look 1474behind the starting point to discover that it is preceded by a letter. 1475</P> 1476<P> 1477If a non-zero starting offset is passed when the pattern is anchored, one 1478attempt to match at the given offset is made. This can only succeed if the 1479pattern does not require the match to be at the start of the subject. 1480</P> 1481<br><b> 1482How <b>pcre_exec()</b> returns captured substrings 1483</b><br> 1484<P> 1485In general, a pattern matches a certain portion of the subject, and in 1486addition, further substrings from the subject may be picked out by parts of the 1487pattern. Following the usage in Jeffrey Friedl's book, this is called 1488"capturing" in what follows, and the phrase "capturing subpattern" is used for 1489a fragment of a pattern that picks out a substring. PCRE supports several other 1490kinds of parenthesized subpattern that do not cause substrings to be captured. 1491</P> 1492<P> 1493Captured substrings are returned to the caller via a vector of integers whose 1494address is passed in <i>ovector</i>. The number of elements in the vector is 1495passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>: this 1496argument is NOT the size of <i>ovector</i> in bytes. 1497</P> 1498<P> 1499The first two-thirds of the vector is used to pass back captured substrings, 1500each substring using a pair of integers. The remaining third of the vector is 1501used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns, 1502and is not available for passing back information. The number passed in 1503<i>ovecsize</i> should always be a multiple of three. If it is not, it is 1504rounded down. 1505</P> 1506<P> 1507When a match is successful, information about captured substrings is returned 1508in pairs of integers, starting at the beginning of <i>ovector</i>, and 1509continuing up to two-thirds of its length at the most. The first element of 1510each pair is set to the byte offset of the first character in a substring, and 1511the second is set to the byte offset of the first character after the end of a 1512substring. <b>Note</b>: these values are always byte offsets, even in UTF-8 1513mode. They are not character counts. 1514</P> 1515<P> 1516The first pair of integers, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the 1517portion of the subject string matched by the entire pattern. The next pair is 1518used for the first capturing subpattern, and so on. The value returned by 1519<b>pcre_exec()</b> is one more than the highest numbered pair that has been set. 1520For example, if two substrings have been captured, the returned value is 3. If 1521there are no capturing subpatterns, the return value from a successful match is 15221, indicating that just the first pair of offsets has been set. 1523</P> 1524<P> 1525If a capturing subpattern is matched repeatedly, it is the last portion of the 1526string that it matched that is returned. 1527</P> 1528<P> 1529If the vector is too small to hold all the captured substring offsets, it is 1530used as far as possible (up to two-thirds of its length), and the function 1531returns a value of zero. If the substring offsets are not of interest, 1532<b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and 1533<i>ovecsize</i> as zero. However, if the pattern contains back references and 1534the <i>ovector</i> is not big enough to remember the related substrings, PCRE 1535has to get additional memory for use during matching. Thus it is usually 1536advisable to supply an <i>ovector</i>. 1537</P> 1538<P> 1539The <b>pcre_fullinfo()</b> function can be used to find out how many capturing 1540subpatterns there are in a compiled pattern. The smallest size for 1541<i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to 1542the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3. 1543</P> 1544<P> 1545It is possible for capturing subpattern number <i>n+1</i> to match some part of 1546the subject when subpattern <i>n</i> has not been used at all. For example, if 1547the string "abc" is matched against the pattern (a|(z))(bc) the return from the 1548function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this 1549happens, both values in the offset pairs corresponding to unused subpatterns 1550are set to -1. 1551</P> 1552<P> 1553Offset values that correspond to unused subpatterns at the end of the 1554expression are also set to -1. For example, if the string "abc" is matched 1555against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The 1556return from the function is 2, because the highest used capturing subpattern 1557number is 1. However, you can refer to the offsets for the second and third 1558capturing subpatterns if you wish (assuming the vector is large enough, of 1559course). 1560</P> 1561<P> 1562Some convenience functions are provided for extracting the captured substrings 1563as separate strings. These are described below. 1564<a name="errorlist"></a></P> 1565<br><b> 1566Error return values from <b>pcre_exec()</b> 1567</b><br> 1568<P> 1569If <b>pcre_exec()</b> fails, it returns a negative number. The following are 1570defined in the header file: 1571<pre> 1572 PCRE_ERROR_NOMATCH (-1) 1573</pre> 1574The subject string did not match the pattern. 1575<pre> 1576 PCRE_ERROR_NULL (-2) 1577</pre> 1578Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was 1579NULL and <i>ovecsize</i> was not zero. 1580<pre> 1581 PCRE_ERROR_BADOPTION (-3) 1582</pre> 1583An unrecognized bit was set in the <i>options</i> argument. 1584<pre> 1585 PCRE_ERROR_BADMAGIC (-4) 1586</pre> 1587PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch 1588the case when it is passed a junk pointer and to detect when a pattern that was 1589compiled in an environment of one endianness is run in an environment with the 1590other endianness. This is the error that PCRE gives when the magic number is 1591not present. 1592<pre> 1593 PCRE_ERROR_UNKNOWN_OPCODE (-5) 1594</pre> 1595While running the pattern match, an unknown item was encountered in the 1596compiled pattern. This error could be caused by a bug in PCRE or by overwriting 1597of the compiled pattern. 1598<pre> 1599 PCRE_ERROR_NOMEMORY (-6) 1600</pre> 1601If a pattern contains back references, but the <i>ovector</i> that is passed to 1602<b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE 1603gets a block of memory at the start of matching to use for this purpose. If the 1604call via <b>pcre_malloc()</b> fails, this error is given. The memory is 1605automatically freed at the end of matching. 1606<pre> 1607 PCRE_ERROR_NOSUBSTRING (-7) 1608</pre> 1609This error is used by the <b>pcre_copy_substring()</b>, 1610<b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see 1611below). It is never returned by <b>pcre_exec()</b>. 1612<pre> 1613 PCRE_ERROR_MATCHLIMIT (-8) 1614</pre> 1615The backtracking limit, as specified by the <i>match_limit</i> field in a 1616<b>pcre_extra</b> structure (or defaulted) was reached. See the description 1617above. 1618<pre> 1619 PCRE_ERROR_CALLOUT (-9) 1620</pre> 1621This error is never generated by <b>pcre_exec()</b> itself. It is provided for 1622use by callout functions that want to yield a distinctive error code. See the 1623<a href="pcrecallout.html"><b>pcrecallout</b></a> 1624documentation for details. 1625<pre> 1626 PCRE_ERROR_BADUTF8 (-10) 1627</pre> 1628A string that contains an invalid UTF-8 byte sequence was passed as a subject. 1629<pre> 1630 PCRE_ERROR_BADUTF8_OFFSET (-11) 1631</pre> 1632The UTF-8 byte sequence that was passed as a subject was valid, but the value 1633of <i>startoffset</i> did not point to the beginning of a UTF-8 character. 1634<pre> 1635 PCRE_ERROR_PARTIAL (-12) 1636</pre> 1637The subject string did not match, but it did match partially. See the 1638<a href="pcrepartial.html"><b>pcrepartial</b></a> 1639documentation for details of partial matching. 1640<pre> 1641 PCRE_ERROR_BADPARTIAL (-13) 1642</pre> 1643This code is no longer in use. It was formerly returned when the PCRE_PARTIAL 1644option was used with a compiled pattern containing items that were not 1645supported for partial matching. From release 8.00 onwards, there are no 1646restrictions on partial matching. 1647<pre> 1648 PCRE_ERROR_INTERNAL (-14) 1649</pre> 1650An unexpected internal error has occurred. This error could be caused by a bug 1651in PCRE or by overwriting of the compiled pattern. 1652<pre> 1653 PCRE_ERROR_BADCOUNT (-15) 1654</pre> 1655This error is given if the value of the <i>ovecsize</i> argument is negative. 1656<pre> 1657 PCRE_ERROR_RECURSIONLIMIT (-21) 1658</pre> 1659The internal recursion limit, as specified by the <i>match_limit_recursion</i> 1660field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the 1661description above. 1662<pre> 1663 PCRE_ERROR_BADNEWLINE (-23) 1664</pre> 1665An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given. 1666</P> 1667<P> 1668Error numbers -16 to -20 and -22 are not used by <b>pcre_exec()</b>. 1669</P> 1670<br><a name="SEC15" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br> 1671<P> 1672<b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b> 1673<b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b> 1674<b>int <i>buffersize</i>);</b> 1675</P> 1676<P> 1677<b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b> 1678<b>int <i>stringcount</i>, int <i>stringnumber</i>,</b> 1679<b>const char **<i>stringptr</i>);</b> 1680</P> 1681<P> 1682<b>int pcre_get_substring_list(const char *<i>subject</i>,</b> 1683<b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b> 1684</P> 1685<P> 1686Captured substrings can be accessed directly by using the offsets returned by 1687<b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions 1688<b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and 1689<b>pcre_get_substring_list()</b> are provided for extracting captured substrings 1690as new, separate, zero-terminated strings. These functions identify substrings 1691by number. The next section describes functions for extracting named 1692substrings. 1693</P> 1694<P> 1695A substring that contains a binary zero is correctly extracted and has a 1696further zero added on the end, but the result is not, of course, a C string. 1697However, you can process such a string by referring to the length that is 1698returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>. 1699Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate 1700for handling strings containing binary zeros, because the end of the final 1701string is not independently indicated. 1702</P> 1703<P> 1704The first three arguments are the same for all three of these functions: 1705<i>subject</i> is the subject string that has just been successfully matched, 1706<i>ovector</i> is a pointer to the vector of integer offsets that was passed to 1707<b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were 1708captured by the match, including the substring that matched the entire regular 1709expression. This is the value returned by <b>pcre_exec()</b> if it is greater 1710than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of 1711space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the 1712number of elements in the vector divided by three. 1713</P> 1714<P> 1715The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b> 1716extract a single substring, whose number is given as <i>stringnumber</i>. A 1717value of zero extracts the substring that matched the entire pattern, whereas 1718higher values extract the captured substrings. For <b>pcre_copy_substring()</b>, 1719the string is placed in <i>buffer</i>, whose length is given by 1720<i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is 1721obtained via <b>pcre_malloc</b>, and its address is returned via 1722<i>stringptr</i>. The yield of the function is the length of the string, not 1723including the terminating zero, or one of these error codes: 1724<pre> 1725 PCRE_ERROR_NOMEMORY (-6) 1726</pre> 1727The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get 1728memory failed for <b>pcre_get_substring()</b>. 1729<pre> 1730 PCRE_ERROR_NOSUBSTRING (-7) 1731</pre> 1732There is no substring whose number is <i>stringnumber</i>. 1733</P> 1734<P> 1735The <b>pcre_get_substring_list()</b> function extracts all available substrings 1736and builds a list of pointers to them. All this is done in a single block of 1737memory that is obtained via <b>pcre_malloc</b>. The address of the memory block 1738is returned via <i>listptr</i>, which is also the start of the list of string 1739pointers. The end of the list is marked by a NULL pointer. The yield of the 1740function is zero if all went well, or the error code 1741<pre> 1742 PCRE_ERROR_NOMEMORY (-6) 1743</pre> 1744if the attempt to get the memory block failed. 1745</P> 1746<P> 1747When any of these functions encounter a substring that is unset, which can 1748happen when capturing subpattern number <i>n+1</i> matches some part of the 1749subject, but subpattern <i>n</i> has not been used at all, they return an empty 1750string. This can be distinguished from a genuine zero-length substring by 1751inspecting the appropriate offset in <i>ovector</i>, which is negative for unset 1752substrings. 1753</P> 1754<P> 1755The two convenience functions <b>pcre_free_substring()</b> and 1756<b>pcre_free_substring_list()</b> can be used to free the memory returned by 1757a previous call of <b>pcre_get_substring()</b> or 1758<b>pcre_get_substring_list()</b>, respectively. They do nothing more than call 1759the function pointed to by <b>pcre_free</b>, which of course could be called 1760directly from a C program. However, PCRE is used in some situations where it is 1761linked via a special interface to another programming language that cannot use 1762<b>pcre_free</b> directly; it is for these cases that the functions are 1763provided. 1764</P> 1765<br><a name="SEC16" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br> 1766<P> 1767<b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b> 1768<b>const char *<i>name</i>);</b> 1769</P> 1770<P> 1771<b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b> 1772<b>const char *<i>subject</i>, int *<i>ovector</i>,</b> 1773<b>int <i>stringcount</i>, const char *<i>stringname</i>,</b> 1774<b>char *<i>buffer</i>, int <i>buffersize</i>);</b> 1775</P> 1776<P> 1777<b>int pcre_get_named_substring(const pcre *<i>code</i>,</b> 1778<b>const char *<i>subject</i>, int *<i>ovector</i>,</b> 1779<b>int <i>stringcount</i>, const char *<i>stringname</i>,</b> 1780<b>const char **<i>stringptr</i>);</b> 1781</P> 1782<P> 1783To extract a substring by name, you first have to find associated number. 1784For example, for this pattern 1785<pre> 1786 (a+)b(?<xxx>\d+)... 1787</pre> 1788the number of the subpattern called "xxx" is 2. If the name is known to be 1789unique (PCRE_DUPNAMES was not set), you can find the number from the name by 1790calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled 1791pattern, and the second is the name. The yield of the function is the 1792subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of 1793that name. 1794</P> 1795<P> 1796Given the number, you can extract the substring directly, or use one of the 1797functions described in the previous section. For convenience, there are also 1798two functions that do the whole job. 1799</P> 1800<P> 1801Most of the arguments of <b>pcre_copy_named_substring()</b> and 1802<b>pcre_get_named_substring()</b> are the same as those for the similarly named 1803functions that extract by number. As these are described in the previous 1804section, they are not re-described here. There are just two differences: 1805</P> 1806<P> 1807First, instead of a substring number, a substring name is given. Second, there 1808is an extra argument, given at the start, which is a pointer to the compiled 1809pattern. This is needed in order to gain access to the name-to-number 1810translation table. 1811</P> 1812<P> 1813These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they 1814then call <b>pcre_copy_substring()</b> or <b>pcre_get_substring()</b>, as 1815appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names, 1816the behaviour may not be what you want (see the next section). 1817</P> 1818<P> 1819<b>Warning:</b> If the pattern uses the (?| feature to set up multiple 1820subpatterns with the same number, as described in the 1821<a href="pcrepattern.html#dupsubpatternnumber">section on duplicate subpattern numbers</a> 1822in the 1823<a href="pcrepattern.html"><b>pcrepattern</b></a> 1824page, you cannot use names to distinguish the different subpatterns, because 1825names are not included in the compiled code. The matching process uses only 1826numbers. For this reason, the use of different names for subpatterns of the 1827same number causes an error at compile time. 1828</P> 1829<br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br> 1830<P> 1831<b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b> 1832<b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b> 1833</P> 1834<P> 1835When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns 1836are not required to be unique. (Duplicate names are always allowed for 1837subpatterns with the same number, created by using the (?| feature. Indeed, if 1838such subpatterns are named, they are required to use the same names.) 1839</P> 1840<P> 1841Normally, patterns with duplicate names are such that in any one match, only 1842one of the named subpatterns participates. An example is shown in the 1843<a href="pcrepattern.html"><b>pcrepattern</b></a> 1844documentation. 1845</P> 1846<P> 1847When duplicates are present, <b>pcre_copy_named_substring()</b> and 1848<b>pcre_get_named_substring()</b> return the first substring corresponding to 1849the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is 1850returned; no data is returned. The <b>pcre_get_stringnumber()</b> function 1851returns one of the numbers that are associated with the name, but it is not 1852defined which it is. 1853</P> 1854<P> 1855If you want to get full details of all captured substrings for a given name, 1856you must use the <b>pcre_get_stringtable_entries()</b> function. The first 1857argument is the compiled pattern, and the second is the name. The third and 1858fourth are pointers to variables which are updated by the function. After it 1859has run, they point to the first and last entries in the name-to-number table 1860for the given name. The function itself returns the length of each entry, or 1861PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is 1862described above in the section entitled <i>Information about a pattern</i>. 1863Given all the relevant entries for the name, you can extract each of their 1864numbers, and hence the captured data, if any. 1865</P> 1866<br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br> 1867<P> 1868The traditional matching function uses a similar algorithm to Perl, which stops 1869when it finds the first match, starting at a given point in the subject. If you 1870want to find all possible matches, or the longest possible match, consider 1871using the alternative matching function (see below) instead. If you cannot use 1872the alternative function, but still need to find all possible matches, you 1873can kludge it up by making use of the callout facility, which is described in 1874the 1875<a href="pcrecallout.html"><b>pcrecallout</b></a> 1876documentation. 1877</P> 1878<P> 1879What you have to do is to insert a callout right at the end of the pattern. 1880When your callout function is called, extract and save the current matched 1881substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try 1882other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b> 1883will yield PCRE_ERROR_NOMATCH. 1884<a name="dfamatch"></a></P> 1885<br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br> 1886<P> 1887<b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b> 1888<b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b> 1889<b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b> 1890<b>int *<i>workspace</i>, int <i>wscount</i>);</b> 1891</P> 1892<P> 1893The function <b>pcre_dfa_exec()</b> is called to match a subject string against 1894a compiled pattern, using a matching algorithm that scans the subject string 1895just once, and does not backtrack. This has different characteristics to the 1896normal algorithm, and is not compatible with Perl. Some of the features of PCRE 1897patterns are not supported. Nevertheless, there are times when this kind of 1898matching can be useful. For a discussion of the two matching algorithms, and a 1899list of features that <b>pcre_dfa_exec()</b> does not support, see the 1900<a href="pcrematching.html"><b>pcrematching</b></a> 1901documentation. 1902</P> 1903<P> 1904The arguments for the <b>pcre_dfa_exec()</b> function are the same as for 1905<b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a 1906different way, and this is described below. The other common arguments are used 1907in the same way as for <b>pcre_exec()</b>, so their description is not repeated 1908here. 1909</P> 1910<P> 1911The two additional arguments provide workspace for the function. The workspace 1912vector should contain at least 20 elements. It is used for keeping track of 1913multiple paths through the pattern tree. More workspace will be needed for 1914patterns and subjects where there are a lot of potential matches. 1915</P> 1916<P> 1917Here is an example of a simple call to <b>pcre_dfa_exec()</b>: 1918<pre> 1919 int rc; 1920 int ovector[10]; 1921 int wspace[20]; 1922 rc = pcre_dfa_exec( 1923 re, /* result of pcre_compile() */ 1924 NULL, /* we didn't study the pattern */ 1925 "some string", /* the subject string */ 1926 11, /* the length of the subject string */ 1927 0, /* start at offset 0 in the subject */ 1928 0, /* default options */ 1929 ovector, /* vector of integers for substring information */ 1930 10, /* number of elements (NOT size in bytes) */ 1931 wspace, /* working space vector */ 1932 20); /* number of elements (NOT size in bytes) */ 1933</PRE> 1934</P> 1935<br><b> 1936Option bits for <b>pcre_dfa_exec()</b> 1937</b><br> 1938<P> 1939The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be 1940zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>, 1941PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART, 1942PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, 1943and PCRE_DFA_RESTART. All but the last four of these are exactly the same as 1944for <b>pcre_exec()</b>, so their description is not repeated here. 1945<pre> 1946 PCRE_PARTIAL_HARD 1947 PCRE_PARTIAL_SOFT 1948</pre> 1949These have the same general effect as they do for <b>pcre_exec()</b>, but the 1950details are slightly different. When PCRE_PARTIAL_HARD is set for 1951<b>pcre_dfa_exec()</b>, it returns PCRE_ERROR_PARTIAL if the end of the subject 1952is reached and there is still at least one matching possibility that requires 1953additional characters. This happens even if some complete matches have also 1954been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH 1955is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached, 1956there have been no complete matches, but there is still at least one matching 1957possibility. The portion of the string that was inspected when the longest 1958partial match was found is set as the first matching string in both cases. 1959<pre> 1960 PCRE_DFA_SHORTEST 1961</pre> 1962Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as 1963soon as it has found one match. Because of the way the alternative algorithm 1964works, this is necessarily the shortest possible match at the first possible 1965matching point in the subject string. 1966<pre> 1967 PCRE_DFA_RESTART 1968</pre> 1969When <b>pcre_dfa_exec()</b> returns a partial match, it is possible to call it 1970again, with additional subject characters, and have it continue with the same 1971match. The PCRE_DFA_RESTART option requests this action; when it is set, the 1972<i>workspace</i> and <i>wscount</i> options must reference the same vector as 1973before because data about the match so far is left in them after a partial 1974match. There is more discussion of this facility in the 1975<a href="pcrepartial.html"><b>pcrepartial</b></a> 1976documentation. 1977</P> 1978<br><b> 1979Successful returns from <b>pcre_dfa_exec()</b> 1980</b><br> 1981<P> 1982When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one 1983substring in the subject. Note, however, that all the matches from one run of 1984the function start at the same point in the subject. The shorter matches are 1985all initial substrings of the longer matches. For example, if the pattern 1986<pre> 1987 <.*> 1988</pre> 1989is matched against the string 1990<pre> 1991 This is <something> <something else> <something further> no more 1992</pre> 1993the three matched strings are 1994<pre> 1995 <something> 1996 <something> <something else> 1997 <something> <something else> <something further> 1998</pre> 1999On success, the yield of the function is a number greater than zero, which is 2000the number of matched substrings. The substrings themselves are returned in 2001<i>ovector</i>. Each string uses two elements; the first is the offset to the 2002start, and the second is the offset to the end. In fact, all the strings have 2003the same start offset. (Space could have been saved by giving this only once, 2004but it was decided to retain some compatibility with the way <b>pcre_exec()</b> 2005returns data, even though the meaning of the strings is different.) 2006</P> 2007<P> 2008The strings are returned in reverse order of length; that is, the longest 2009matching string is given first. If there were too many matches to fit into 2010<i>ovector</i>, the yield of the function is zero, and the vector is filled with 2011the longest matches. 2012</P> 2013<br><b> 2014Error returns from <b>pcre_dfa_exec()</b> 2015</b><br> 2016<P> 2017The <b>pcre_dfa_exec()</b> function returns a negative number when it fails. 2018Many of the errors are the same as for <b>pcre_exec()</b>, and these are 2019described 2020<a href="#errorlist">above.</a> 2021There are in addition the following errors that are specific to 2022<b>pcre_dfa_exec()</b>: 2023<pre> 2024 PCRE_ERROR_DFA_UITEM (-16) 2025</pre> 2026This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern 2027that it does not support, for instance, the use of \C or a back reference. 2028<pre> 2029 PCRE_ERROR_DFA_UCOND (-17) 2030</pre> 2031This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that 2032uses a back reference for the condition, or a test for recursion in a specific 2033group. These are not supported. 2034<pre> 2035 PCRE_ERROR_DFA_UMLIMIT (-18) 2036</pre> 2037This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i> 2038block that contains a setting of the <i>match_limit</i> field. This is not 2039supported (it is meaningless). 2040<pre> 2041 PCRE_ERROR_DFA_WSSIZE (-19) 2042</pre> 2043This return is given if <b>pcre_dfa_exec()</b> runs out of space in the 2044<i>workspace</i> vector. 2045<pre> 2046 PCRE_ERROR_DFA_RECURSE (-20) 2047</pre> 2048When a recursive subpattern is processed, the matching function calls itself 2049recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This 2050error is given if the output vector is not large enough. This should be 2051extremely rare, as a vector of size 1000 is used. 2052</P> 2053<br><a name="SEC20" href="#TOC1">SEE ALSO</a><br> 2054<P> 2055<b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3), 2056<b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3), 2057<b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3). 2058</P> 2059<br><a name="SEC21" href="#TOC1">AUTHOR</a><br> 2060<P> 2061Philip Hazel 2062<br> 2063University Computing Service 2064<br> 2065Cambridge CB2 3QH, England. 2066<br> 2067</P> 2068<br><a name="SEC22" href="#TOC1">REVISION</a><br> 2069<P> 2070Last updated: 03 October 2009 2071<br> 2072Copyright © 1997-2009 University of Cambridge. 2073<br> 2074<p> 2075Return to the <a href="index.html">PCRE index page</a>. 2076</p> 2077