1<html> 2<head> 3<title>pcre2posix specification</title> 4</head> 5<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB"> 6<h1>pcre2posix man page</h1> 7<p> 8Return to the <a href="index.html">PCRE2 index page</a>. 9</p> 10<p> 11This page is part of the PCRE2 HTML documentation. It was generated 12automatically from the original man page. If there is any nonsense in it, 13please consult the man page, in case the conversion went wrong. 14<br> 15<ul> 16<li><a name="TOC1" href="#SEC1">SYNOPSIS</a> 17<li><a name="TOC2" href="#SEC2">DESCRIPTION</a> 18<li><a name="TOC3" href="#SEC3">USING THE POSIX FUNCTIONS</a> 19<li><a name="TOC4" href="#SEC4">COMPILING A PATTERN</a> 20<li><a name="TOC5" href="#SEC5">MATCHING NEWLINE CHARACTERS</a> 21<li><a name="TOC6" href="#SEC6">MATCHING A PATTERN</a> 22<li><a name="TOC7" href="#SEC7">ERROR MESSAGES</a> 23<li><a name="TOC8" href="#SEC8">MEMORY USAGE</a> 24<li><a name="TOC9" href="#SEC9">AUTHOR</a> 25<li><a name="TOC10" href="#SEC10">REVISION</a> 26</ul> 27<br><a name="SEC1" href="#TOC1">SYNOPSIS</a><br> 28<P> 29<b>#include <pcre2posix.h></b> 30</P> 31<P> 32<b>int pcre2_regcomp(regex_t *<i>preg</i>, const char *<i>pattern</i>,</b> 33<b> int <i>cflags</i>);</b> 34<br> 35<br> 36<b>int pcre2_regexec(const regex_t *<i>preg</i>, const char *<i>string</i>,</b> 37<b> size_t <i>nmatch</i>, regmatch_t <i>pmatch</i>[], int <i>eflags</i>);</b> 38<br> 39<br> 40<b>size_t pcre2_regerror(int <i>errcode</i>, const regex_t *<i>preg</i>,</b> 41<b> char *<i>errbuf</i>, size_t <i>errbuf_size</i>);</b> 42<br> 43<br> 44<b>void pcre2_regfree(regex_t *<i>preg</i>);</b> 45</P> 46<br><a name="SEC2" href="#TOC1">DESCRIPTION</a><br> 47<P> 48This set of functions provides a POSIX-style API for the PCRE2 regular 49expression 8-bit library. There are no POSIX-style wrappers for PCRE2's 16-bit 50and 32-bit libraries. See the 51<a href="pcre2api.html"><b>pcre2api</b></a> 52documentation for a description of PCRE2's native API, which contains much 53additional functionality. 54</P> 55<P> 56<b>IMPORTANT NOTE</b>: The functions described here are NOT thread-safe, and 57should not be used in multi-threaded applications. They are also limited to 58processing subjects that are not bigger than 2GB. Use the native API instead. 59</P> 60<P> 61These functions are wrapper functions that ultimately call the PCRE2 native 62API. Their prototypes are defined in the <b>pcre2posix.h</b> header file, and 63they all have unique names starting with <b>pcre2_</b>. However, the 64<b>pcre2posix.h</b> header also contains macro definitions that convert the 65standard POSIX names such <b>regcomp()</b> into <b>pcre2_regcomp()</b> etc. This 66means that a program can use the usual POSIX names without running the risk of 67accidentally linking with POSIX functions from a different library. 68</P> 69<P> 70On Unix-like systems the PCRE2 POSIX library is called <b>libpcre2-posix</b>, so 71can be accessed by adding <b>-lpcre2-posix</b> to the command for linking an 72application. Because the POSIX functions call the native ones, it is also 73necessary to add <b>-lpcre2-8</b>. 74</P> 75<P> 76On Windows systems, if you are linking to a DLL version of the library, it is 77recommended that <b>PCRE2POSIX_SHARED</b> is defined before including the 78<b>pcre2posix.h</b> header, as it will allow for a more efficient way to 79invoke the functions by adding the <b>__declspec(dllimport)</b> decorator. 80</P> 81<P> 82Although they were not defined as prototypes in <b>pcre2posix.h</b>, releases 8310.33 to 10.36 of the library contained functions with the POSIX names 84<b>regcomp()</b> etc. These simply passed their arguments to the PCRE2 85functions. These functions were provided for backwards compatibility with 86earlier versions of PCRE2, which had only POSIX names. However, this has proved 87troublesome in situations where a program links with several libraries, some of 88which use PCRE2's POSIX interface while others use the real POSIX functions. 89For this reason, the POSIX names have been removed since release 10.37. 90</P> 91<P> 92Calling the header file <b>pcre2posix.h</b> avoids any conflict with other POSIX 93libraries. It can, of course, be renamed or aliased as <b>regex.h</b>, which is 94the "correct" name, if there is no clash. It provides two structure types, 95<i>regex_t</i> for compiled internal forms, and <i>regmatch_t</i> for returning 96captured substrings. It also defines some constants whose names start with 97"REG_"; these are used for setting options and identifying error codes. 98</P> 99<br><a name="SEC3" href="#TOC1">USING THE POSIX FUNCTIONS</a><br> 100<P> 101Note that these functions are just POSIX-style wrappers for PCRE2's native API. 102They do not give POSIX regular expression behaviour, and they are not 103thread-safe or even POSIX compatible. 104</P> 105<P> 106Those POSIX option bits that can reasonably be mapped to PCRE2 native options 107have been implemented. In addition, the option REG_EXTENDED is defined with the 108value zero. This has no effect, but since programs that are written to the 109POSIX interface often use it, this makes it easier to slot in PCRE2 as a 110replacement library. Other POSIX options are not even defined. 111</P> 112<P> 113There are also some options that are not defined by POSIX. These have been 114added at the request of users who want to make use of certain PCRE2-specific 115features via the POSIX calling interface or to add BSD or GNU functionality. 116</P> 117<P> 118When PCRE2 is called via these functions, it is only the API that is POSIX-like 119in style. The syntax and semantics of the regular expressions themselves are 120still those of Perl, subject to the setting of various PCRE2 options, as 121described below. "POSIX-like in style" means that the API approximates to the 122POSIX definition; it is not fully POSIX-compatible, and in multi-unit encoding 123domains it is probably even less compatible. 124</P> 125<P> 126The descriptions below use the actual names of the functions, but, as described 127above, the standard POSIX names (without the <b>pcre2_</b> prefix) may also be 128used. 129</P> 130<br><a name="SEC4" href="#TOC1">COMPILING A PATTERN</a><br> 131<P> 132The function <b>pcre2_regcomp()</b> is called to compile a pattern into an 133internal form. By default, the pattern is a C string terminated by a binary 134zero (but see REG_PEND below). The <i>preg</i> argument is a pointer to a 135<b>regex_t</b> structure that is used as a base for storing information about 136the compiled regular expression. It is also used for input when REG_PEND is 137set. The <b>regex_t</b> structure used by <b>pcre2_regcomp()</b> is defined in 138<b>pcre2posix.h</b> and is not the same as the structure used by other libraries 139that provide POSIX-style matching. 140</P> 141<P> 142The argument <i>cflags</i> is either zero, or contains one or more of the bits 143defined by the following macros: 144<pre> 145 REG_DOTALL 146</pre> 147The PCRE2_DOTALL option is set when the regular expression is passed for 148compilation to the native function. Note that REG_DOTALL is not part of the 149POSIX standard. 150<pre> 151 REG_ICASE 152</pre> 153The PCRE2_CASELESS option is set when the regular expression is passed for 154compilation to the native function. 155<pre> 156 REG_NEWLINE 157</pre> 158The PCRE2_MULTILINE option is set when the regular expression is passed for 159compilation to the native function. Note that this does <i>not</i> mimic the 160defined POSIX behaviour for REG_NEWLINE (see the following section). 161<pre> 162 REG_NOSPEC 163</pre> 164The PCRE2_LITERAL option is set when the regular expression is passed for 165compilation to the native function. This disables all meta characters in the 166pattern, causing it to be treated as a literal string. The only other options 167that are allowed with REG_NOSPEC are REG_ICASE, REG_NOSUB, REG_PEND, and 168REG_UTF. Note that REG_NOSPEC is not part of the POSIX standard. 169<pre> 170 REG_NOSUB 171</pre> 172When a pattern that is compiled with this flag is passed to 173<b>pcre2_regexec()</b> for matching, the <i>nmatch</i> and <i>pmatch</i> arguments 174are ignored, and no captured strings are returned. Versions of the PCRE library 175prior to 10.22 used to set the PCRE2_NO_AUTO_CAPTURE compile option, but this 176no longer happens because it disables the use of backreferences. 177<pre> 178 REG_PEND 179</pre> 180If this option is set, the <b>reg_endp</b> field in the <i>preg</i> structure 181(which has the type const char *) must be set to point to the character beyond 182the end of the pattern before calling <b>pcre2_regcomp()</b>. The pattern itself 183may now contain binary zeros, which are treated as data characters. Without 184REG_PEND, a binary zero terminates the pattern and the <b>re_endp</b> field is 185ignored. This is a GNU extension to the POSIX standard and should be used with 186caution in software intended to be portable to other systems. 187<pre> 188 REG_UCP 189</pre> 190The PCRE2_UCP option is set when the regular expression is passed for 191compilation to the native function. This causes PCRE2 to use Unicode properties 192when matching \d, \w, etc., instead of just recognizing ASCII values. Note 193that REG_UCP is not part of the POSIX standard. 194<pre> 195 REG_UNGREEDY 196</pre> 197The PCRE2_UNGREEDY option is set when the regular expression is passed for 198compilation to the native function. Note that REG_UNGREEDY is not part of the 199POSIX standard. 200<pre> 201 REG_UTF 202</pre> 203The PCRE2_UTF option is set when the regular expression is passed for 204compilation to the native function. This causes the pattern itself and all data 205strings used for matching it to be treated as UTF-8 strings. Note that REG_UTF 206is not part of the POSIX standard. 207</P> 208<P> 209In the absence of these flags, no options are passed to the native function. 210This means that the regex is compiled with PCRE2 default semantics. In 211particular, the way it handles newline characters in the subject string is the 212Perl way, not the POSIX way. Note that setting PCRE2_MULTILINE has only 213<i>some</i> of the effects specified for REG_NEWLINE. It does not affect the way 214newlines are matched by the dot metacharacter (they are not) or by a negative 215class such as [^a] (they are). 216</P> 217<P> 218The yield of <b>pcre2_regcomp()</b> is zero on success, and non-zero otherwise. 219The <i>preg</i> structure is filled in on success, and one other member of the 220structure (as well as <i>re_endp</i>) is public: <i>re_nsub</i> contains the 221number of capturing subpatterns in the regular expression. Various error codes 222are defined in the header file. 223</P> 224<P> 225NOTE: If the yield of <b>pcre2_regcomp()</b> is non-zero, you must not attempt 226to use the contents of the <i>preg</i> structure. If, for example, you pass it 227to <b>pcre2_regexec()</b>, the result is undefined and your program is likely to 228crash. 229</P> 230<br><a name="SEC5" href="#TOC1">MATCHING NEWLINE CHARACTERS</a><br> 231<P> 232This area is not simple, because POSIX and Perl take different views of things. 233It is not possible to get PCRE2 to obey POSIX semantics, but then PCRE2 was 234never intended to be a POSIX engine. The following table lists the different 235possibilities for matching newline characters in Perl and PCRE2: 236<pre> 237 Default Change with 238 239 . matches newline no PCRE2_DOTALL 240 newline matches [^a] yes not changeable 241 $ matches \n at end yes PCRE2_DOLLAR_ENDONLY 242 $ matches \n in middle no PCRE2_MULTILINE 243 ^ matches \n in middle no PCRE2_MULTILINE 244</pre> 245This is the equivalent table for a POSIX-compatible pattern matcher: 246<pre> 247 Default Change with 248 249 . matches newline yes REG_NEWLINE 250 newline matches [^a] yes REG_NEWLINE 251 $ matches \n at end no REG_NEWLINE 252 $ matches \n in middle no REG_NEWLINE 253 ^ matches \n in middle no REG_NEWLINE 254</pre> 255This behaviour is not what happens when PCRE2 is called via its POSIX 256API. By default, PCRE2's behaviour is the same as Perl's, except that there is 257no equivalent for PCRE2_DOLLAR_ENDONLY in Perl. In both PCRE2 and Perl, there 258is no way to stop newline from matching [^a]. 259</P> 260<P> 261Default POSIX newline handling can be obtained by setting PCRE2_DOTALL and 262PCRE2_DOLLAR_ENDONLY when calling <b>pcre2_compile()</b> directly, but there is 263no way to make PCRE2 behave exactly as for the REG_NEWLINE action. When using 264the POSIX API, passing REG_NEWLINE to PCRE2's <b>pcre2_regcomp()</b> function 265causes PCRE2_MULTILINE to be passed to <b>pcre2_compile()</b>, and REG_DOTALL 266passes PCRE2_DOTALL. There is no way to pass PCRE2_DOLLAR_ENDONLY. 267</P> 268<br><a name="SEC6" href="#TOC1">MATCHING A PATTERN</a><br> 269<P> 270The function <b>pcre2_regexec()</b> is called to match a compiled pattern 271<i>preg</i> against a given <i>string</i>, which is by default terminated by a 272zero byte (but see REG_STARTEND below), subject to the options in <i>eflags</i>. 273These can be: 274<pre> 275 REG_NOTBOL 276</pre> 277The PCRE2_NOTBOL option is set when calling the underlying PCRE2 matching 278function. 279<pre> 280 REG_NOTEMPTY 281</pre> 282The PCRE2_NOTEMPTY option is set when calling the underlying PCRE2 matching 283function. Note that REG_NOTEMPTY is not part of the POSIX standard. However, 284setting this option can give more POSIX-like behaviour in some situations. 285<pre> 286 REG_NOTEOL 287</pre> 288The PCRE2_NOTEOL option is set when calling the underlying PCRE2 matching 289function. 290<pre> 291 REG_STARTEND 292</pre> 293When this option is set, the subject string starts at <i>string</i> + 294<i>pmatch[0].rm_so</i> and ends at <i>string</i> + <i>pmatch[0].rm_eo</i>, which 295should point to the first character beyond the string. There may be binary 296zeros within the subject string, and indeed, using REG_STARTEND is the only 297way to pass a subject string that contains a binary zero. 298</P> 299<P> 300Whatever the value of <i>pmatch[0].rm_so</i>, the offsets of the matched string 301and any captured substrings are still given relative to the start of 302<i>string</i> itself. (Before PCRE2 release 10.30 these were given relative to 303<i>string</i> + <i>pmatch[0].rm_so</i>, but this differs from other 304implementations.) 305</P> 306<P> 307This is a BSD extension, compatible with but not specified by IEEE Standard 3081003.2 (POSIX.2), and should be used with caution in software intended to be 309portable to other systems. Note that a non-zero <i>rm_so</i> does not imply 310REG_NOTBOL; REG_STARTEND affects only the location and length of the string, 311not how it is matched. Setting REG_STARTEND and passing <i>pmatch</i> as NULL 312are mutually exclusive; the error REG_INVARG is returned. 313</P> 314<P> 315If the pattern was compiled with the REG_NOSUB flag, no data about any matched 316strings is returned. The <i>nmatch</i> and <i>pmatch</i> arguments of 317<b>pcre2_regexec()</b> are ignored (except possibly as input for REG_STARTEND). 318</P> 319<P> 320The value of <i>nmatch</i> may be zero, and the value <i>pmatch</i> may be NULL 321(unless REG_STARTEND is set); in both these cases no data about any matched 322strings is returned. 323</P> 324<P> 325Otherwise, the portion of the string that was matched, and also any captured 326substrings, are returned via the <i>pmatch</i> argument, which points to an 327array of <i>nmatch</i> structures of type <i>regmatch_t</i>, containing the 328members <i>rm_so</i> and <i>rm_eo</i>. These contain the byte offset to the first 329character of each substring and the offset to the first character after the end 330of each substring, respectively. The 0th element of the vector relates to the 331entire portion of <i>string</i> that was matched; subsequent elements relate to 332the capturing subpatterns of the regular expression. Unused entries in the 333array have both structure members set to -1. 334</P> 335<P> 336<i>regmatch_t</i> as well as the <i>regoff_t</i> typedef it uses are defined in 337<b>pcre2posix.h</b> and are not warranted to have the same size or layout as other 338similarly named types from other libraries that provide POSIX-style matching. 339</P> 340<P> 341A successful match yields a zero return; various error codes are defined in the 342header file, of which REG_NOMATCH is the "expected" failure code. 343</P> 344<br><a name="SEC7" href="#TOC1">ERROR MESSAGES</a><br> 345<P> 346The <b>pcre2_regerror()</b> function maps a non-zero errorcode from either 347<b>pcre2_regcomp()</b> or <b>pcre2_regexec()</b> to a printable message. If 348<i>preg</i> is not NULL, the error should have arisen from the use of that 349structure. A message terminated by a binary zero is placed in <i>errbuf</i>. If 350the buffer is too short, only the first <i>errbuf_size</i> - 1 characters of the 351error message are used. The yield of the function is the size of buffer needed 352to hold the whole message, including the terminating zero. This value is 353greater than <i>errbuf_size</i> if the message was truncated. 354</P> 355<br><a name="SEC8" href="#TOC1">MEMORY USAGE</a><br> 356<P> 357Compiling a regular expression causes memory to be allocated and associated 358with the <i>preg</i> structure. The function <b>pcre2_regfree()</b> frees all 359such memory, after which <i>preg</i> may no longer be used as a compiled 360expression. 361</P> 362<br><a name="SEC9" href="#TOC1">AUTHOR</a><br> 363<P> 364Philip Hazel 365<br> 366Retired from University Computing Service 367<br> 368Cambridge, England. 369<br> 370</P> 371<br><a name="SEC10" href="#TOC1">REVISION</a><br> 372<P> 373Last updated: 19 January 2024 374<br> 375Copyright © 1997-2024 University of Cambridge. 376<br> 377<p> 378Return to the <a href="index.html">PCRE2 index page</a>. 379</p> 380