1<html> 2<head> 3<title>pcre2syntax specification</title> 4</head> 5<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB"> 6<h1>pcre2syntax 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">PCRE2 REGULAR EXPRESSION SYNTAX SUMMARY</a> 17<li><a name="TOC2" href="#SEC2">QUOTING</a> 18<li><a name="TOC3" href="#SEC3">ESCAPED CHARACTERS</a> 19<li><a name="TOC4" href="#SEC4">CHARACTER TYPES</a> 20<li><a name="TOC5" href="#SEC5">GENERAL CATEGORY PROPERTIES FOR \p and \P</a> 21<li><a name="TOC6" href="#SEC6">PCRE2 SPECIAL CATEGORY PROPERTIES FOR \p and \P</a> 22<li><a name="TOC7" href="#SEC7">BINARY PROPERTIES FOR \p AND \P</a> 23<li><a name="TOC8" href="#SEC8">SCRIPT MATCHING WITH \p AND \P</a> 24<li><a name="TOC9" href="#SEC9">THE BIDI_CLASS PROPERTY FOR \p AND \P</a> 25<li><a name="TOC10" href="#SEC10">CHARACTER CLASSES</a> 26<li><a name="TOC11" href="#SEC11">QUANTIFIERS</a> 27<li><a name="TOC12" href="#SEC12">ANCHORS AND SIMPLE ASSERTIONS</a> 28<li><a name="TOC13" href="#SEC13">REPORTED MATCH POINT SETTING</a> 29<li><a name="TOC14" href="#SEC14">ALTERNATION</a> 30<li><a name="TOC15" href="#SEC15">CAPTURING</a> 31<li><a name="TOC16" href="#SEC16">ATOMIC GROUPS</a> 32<li><a name="TOC17" href="#SEC17">COMMENT</a> 33<li><a name="TOC18" href="#SEC18">OPTION SETTING</a> 34<li><a name="TOC19" href="#SEC19">NEWLINE CONVENTION</a> 35<li><a name="TOC20" href="#SEC20">WHAT \R MATCHES</a> 36<li><a name="TOC21" href="#SEC21">LOOKAHEAD AND LOOKBEHIND ASSERTIONS</a> 37<li><a name="TOC22" href="#SEC22">NON-ATOMIC LOOKAROUND ASSERTIONS</a> 38<li><a name="TOC23" href="#SEC23">SCRIPT RUNS</a> 39<li><a name="TOC24" href="#SEC24">BACKREFERENCES</a> 40<li><a name="TOC25" href="#SEC25">SUBROUTINE REFERENCES (POSSIBLY RECURSIVE)</a> 41<li><a name="TOC26" href="#SEC26">CONDITIONAL PATTERNS</a> 42<li><a name="TOC27" href="#SEC27">BACKTRACKING CONTROL</a> 43<li><a name="TOC28" href="#SEC28">CALLOUTS</a> 44<li><a name="TOC29" href="#SEC29">SEE ALSO</a> 45<li><a name="TOC30" href="#SEC30">AUTHOR</a> 46<li><a name="TOC31" href="#SEC31">REVISION</a> 47</ul> 48<br><a name="SEC1" href="#TOC1">PCRE2 REGULAR EXPRESSION SYNTAX SUMMARY</a><br> 49<P> 50The full syntax and semantics of the regular expressions that are supported by 51PCRE2 are described in the 52<a href="pcre2pattern.html"><b>pcre2pattern</b></a> 53documentation. This document contains a quick-reference summary of the syntax. 54</P> 55<br><a name="SEC2" href="#TOC1">QUOTING</a><br> 56<P> 57<pre> 58 \x where x is non-alphanumeric is a literal x 59 \Q...\E treat enclosed characters as literal 60</PRE> 61</P> 62<br><a name="SEC3" href="#TOC1">ESCAPED CHARACTERS</a><br> 63<P> 64This table applies to ASCII and Unicode environments. An unrecognized escape 65sequence causes an error. 66<pre> 67 \a alarm, that is, the BEL character (hex 07) 68 \cx "control-x", where x is any ASCII printing character 69 \e escape (hex 1B) 70 \f form feed (hex 0C) 71 \n newline (hex 0A) 72 \r carriage return (hex 0D) 73 \t tab (hex 09) 74 \0dd character with octal code 0dd 75 \ddd character with octal code ddd, or backreference 76 \o{ddd..} character with octal code ddd.. 77 \N{U+hh..} character with Unicode code point hh.. (Unicode mode only) 78 \xhh character with hex code hh 79 \x{hh..} character with hex code hh.. 80</pre> 81If PCRE2_ALT_BSUX or PCRE2_EXTRA_ALT_BSUX is set ("ALT_BSUX mode"), the 82following are also recognized: 83<pre> 84 \U the character "U" 85 \uhhhh character with hex code hhhh 86 \u{hh..} character with hex code hh.. but only for EXTRA_ALT_BSUX 87</pre> 88When \x is not followed by {, from zero to two hexadecimal digits are read, 89but in ALT_BSUX mode \x must be followed by two hexadecimal digits to be 90recognized as a hexadecimal escape; otherwise it matches a literal "x". 91Likewise, if \u (in ALT_BSUX mode) is not followed by four hexadecimal digits 92or (in EXTRA_ALT_BSUX mode) a sequence of hex digits in curly brackets, it 93matches a literal "u". 94</P> 95<P> 96Note that \0dd is always an octal code. The treatment of backslash followed by 97a non-zero digit is complicated; for details see the section 98<a href="pcre2pattern.html#digitsafterbackslash">"Non-printing characters"</a> 99in the 100<a href="pcre2pattern.html"><b>pcre2pattern</b></a> 101documentation, where details of escape processing in EBCDIC environments are 102also given. \N{U+hh..} is synonymous with \x{hh..} in PCRE2 but is not 103supported in EBCDIC environments. Note that \N not followed by an opening 104curly bracket has a different meaning (see below). 105</P> 106<br><a name="SEC4" href="#TOC1">CHARACTER TYPES</a><br> 107<P> 108<pre> 109 . any character except newline; 110 in dotall mode, any character whatsoever 111 \C one code unit, even in UTF mode (best avoided) 112 \d a decimal digit 113 \D a character that is not a decimal digit 114 \h a horizontal white space character 115 \H a character that is not a horizontal white space character 116 \N a character that is not a newline 117 \p{<i>xx</i>} a character with the <i>xx</i> property 118 \P{<i>xx</i>} a character without the <i>xx</i> property 119 \R a newline sequence 120 \s a white space character 121 \S a character that is not a white space character 122 \v a vertical white space character 123 \V a character that is not a vertical white space character 124 \w a "word" character 125 \W a "non-word" character 126 \X a Unicode extended grapheme cluster 127</pre> 128\C is dangerous because it may leave the current matching point in the middle 129of a UTF-8 or UTF-16 character. The application can lock out the use of \C by 130setting the PCRE2_NEVER_BACKSLASH_C option. It is also possible to build PCRE2 131with the use of \C permanently disabled. 132</P> 133<P> 134By default, \d, \s, and \w match only ASCII characters, even in UTF-8 mode 135or in the 16-bit and 32-bit libraries. However, if locale-specific matching is 136happening, \s and \w may also match characters with code points in the range 137128-255. If the PCRE2_UCP option is set, the behaviour of these escape 138sequences is changed to use Unicode properties and they match many more 139characters. 140</P> 141<P> 142Property descriptions in \p and \P are matched caselessly; hyphens, 143underscores, and white space are ignored, in accordance with Unicode's "loose 144matching" rules. 145</P> 146<br><a name="SEC5" href="#TOC1">GENERAL CATEGORY PROPERTIES FOR \p and \P</a><br> 147<P> 148<pre> 149 C Other 150 Cc Control 151 Cf Format 152 Cn Unassigned 153 Co Private use 154 Cs Surrogate 155 156 L Letter 157 Ll Lower case letter 158 Lm Modifier letter 159 Lo Other letter 160 Lt Title case letter 161 Lu Upper case letter 162 Lc Ll, Lu, or Lt 163 L& Ll, Lu, or Lt 164 165 M Mark 166 Mc Spacing mark 167 Me Enclosing mark 168 Mn Non-spacing mark 169 170 N Number 171 Nd Decimal number 172 Nl Letter number 173 No Other number 174 175 P Punctuation 176 Pc Connector punctuation 177 Pd Dash punctuation 178 Pe Close punctuation 179 Pf Final punctuation 180 Pi Initial punctuation 181 Po Other punctuation 182 Ps Open punctuation 183 184 S Symbol 185 Sc Currency symbol 186 Sk Modifier symbol 187 Sm Mathematical symbol 188 So Other symbol 189 190 Z Separator 191 Zl Line separator 192 Zp Paragraph separator 193 Zs Space separator 194</PRE> 195</P> 196<br><a name="SEC6" href="#TOC1">PCRE2 SPECIAL CATEGORY PROPERTIES FOR \p and \P</a><br> 197<P> 198<pre> 199 Xan Alphanumeric: union of properties L and N 200 Xps POSIX space: property Z or tab, NL, VT, FF, CR 201 Xsp Perl space: property Z or tab, NL, VT, FF, CR 202 Xuc Univerally-named character: one that can be 203 represented by a Universal Character Name 204 Xwd Perl word: property Xan or underscore 205</pre> 206Perl and POSIX space are now the same. Perl added VT to its space character set 207at release 5.18. 208</P> 209<br><a name="SEC7" href="#TOC1">BINARY PROPERTIES FOR \p AND \P</a><br> 210<P> 211Unicode defines a number of binary properties, that is, properties whose only 212values are true or false. You can obtain a list of those that are recognized by 213\p and \P, along with their abbreviations, by running this command: 214<pre> 215 pcre2test -LP 216</PRE> 217</P> 218<br><a name="SEC8" href="#TOC1">SCRIPT MATCHING WITH \p AND \P</a><br> 219<P> 220Many script names and their 4-letter abbreviations are recognized in 221\p{sc:...} or \p{scx:...} items, or on their own with \p (and also \P of 222course). You can obtain a list of these scripts by running this command: 223<pre> 224 pcre2test -LS 225</PRE> 226</P> 227<br><a name="SEC9" href="#TOC1">THE BIDI_CLASS PROPERTY FOR \p AND \P</a><br> 228<P> 229<pre> 230 \p{Bidi_Class:<class>} matches a character with the given class 231 \p{BC:<class>} matches a character with the given class 232</pre> 233The recognized classes are: 234<pre> 235 AL Arabic letter 236 AN Arabic number 237 B paragraph separator 238 BN boundary neutral 239 CS common separator 240 EN European number 241 ES European separator 242 ET European terminator 243 FSI first strong isolate 244 L left-to-right 245 LRE left-to-right embedding 246 LRI left-to-right isolate 247 LRO left-to-right override 248 NSM non-spacing mark 249 ON other neutral 250 PDF pop directional format 251 PDI pop directional isolate 252 R right-to-left 253 RLE right-to-left embedding 254 RLI right-to-left isolate 255 RLO right-to-left override 256 S segment separator 257 WS which space 258</PRE> 259</P> 260<br><a name="SEC10" href="#TOC1">CHARACTER CLASSES</a><br> 261<P> 262<pre> 263 [...] positive character class 264 [^...] negative character class 265 [x-y] range (can be used for hex characters) 266 [[:xxx:]] positive POSIX named set 267 [[:^xxx:]] negative POSIX named set 268 269 alnum alphanumeric 270 alpha alphabetic 271 ascii 0-127 272 blank space or tab 273 cntrl control character 274 digit decimal digit 275 graph printing, excluding space 276 lower lower case letter 277 print printing, including space 278 punct printing, excluding alphanumeric 279 space white space 280 upper upper case letter 281 word same as \w 282 xdigit hexadecimal digit 283</pre> 284In PCRE2, POSIX character set names recognize only ASCII characters by default, 285but some of them use Unicode properties if PCRE2_UCP is set. You can use 286\Q...\E inside a character class. 287</P> 288<br><a name="SEC11" href="#TOC1">QUANTIFIERS</a><br> 289<P> 290<pre> 291 ? 0 or 1, greedy 292 ?+ 0 or 1, possessive 293 ?? 0 or 1, lazy 294 * 0 or more, greedy 295 *+ 0 or more, possessive 296 *? 0 or more, lazy 297 + 1 or more, greedy 298 ++ 1 or more, possessive 299 +? 1 or more, lazy 300 {n} exactly n 301 {n,m} at least n, no more than m, greedy 302 {n,m}+ at least n, no more than m, possessive 303 {n,m}? at least n, no more than m, lazy 304 {n,} n or more, greedy 305 {n,}+ n or more, possessive 306 {n,}? n or more, lazy 307</PRE> 308</P> 309<br><a name="SEC12" href="#TOC1">ANCHORS AND SIMPLE ASSERTIONS</a><br> 310<P> 311<pre> 312 \b word boundary 313 \B not a word boundary 314 ^ start of subject 315 also after an internal newline in multiline mode 316 (after any newline if PCRE2_ALT_CIRCUMFLEX is set) 317 \A start of subject 318 $ end of subject 319 also before newline at end of subject 320 also before internal newline in multiline mode 321 \Z end of subject 322 also before newline at end of subject 323 \z end of subject 324 \G first matching position in subject 325</PRE> 326</P> 327<br><a name="SEC13" href="#TOC1">REPORTED MATCH POINT SETTING</a><br> 328<P> 329<pre> 330 \K set reported start of match 331</pre> 332From release 10.38 \K is not permitted by default in lookaround assertions, 333for compatibility with Perl. However, if the PCRE2_EXTRA_ALLOW_LOOKAROUND_BSK 334option is set, the previous behaviour is re-enabled. When this option is set, 335\K is honoured in positive assertions, but ignored in negative ones. 336</P> 337<br><a name="SEC14" href="#TOC1">ALTERNATION</a><br> 338<P> 339<pre> 340 expr|expr|expr... 341</PRE> 342</P> 343<br><a name="SEC15" href="#TOC1">CAPTURING</a><br> 344<P> 345<pre> 346 (...) capture group 347 (?<name>...) named capture group (Perl) 348 (?'name'...) named capture group (Perl) 349 (?P<name>...) named capture group (Python) 350 (?:...) non-capture group 351 (?|...) non-capture group; reset group numbers for 352 capture groups in each alternative 353</pre> 354In non-UTF modes, names may contain underscores and ASCII letters and digits; 355in UTF modes, any Unicode letters and Unicode decimal digits are permitted. In 356both cases, a name must not start with a digit. 357</P> 358<br><a name="SEC16" href="#TOC1">ATOMIC GROUPS</a><br> 359<P> 360<pre> 361 (?>...) atomic non-capture group 362 (*atomic:...) atomic non-capture group 363</PRE> 364</P> 365<br><a name="SEC17" href="#TOC1">COMMENT</a><br> 366<P> 367<pre> 368 (?#....) comment (not nestable) 369</PRE> 370</P> 371<br><a name="SEC18" href="#TOC1">OPTION SETTING</a><br> 372<P> 373Changes of these options within a group are automatically cancelled at the end 374of the group. 375<pre> 376 (?i) caseless 377 (?J) allow duplicate named groups 378 (?m) multiline 379 (?n) no auto capture 380 (?s) single line (dotall) 381 (?U) default ungreedy (lazy) 382 (?x) extended: ignore white space except in classes 383 (?xx) as (?x) but also ignore space and tab in classes 384 (?-...) unset option(s) 385 (?^) unset imnsx options 386</pre> 387Unsetting x or xx unsets both. Several options may be set at once, and a 388mixture of setting and unsetting such as (?i-x) is allowed, but there may be 389only one hyphen. Setting (but no unsetting) is allowed after (?^ for example 390(?^in). An option setting may appear at the start of a non-capture group, for 391example (?i:...). 392</P> 393<P> 394The following are recognized only at the very start of a pattern or after one 395of the newline or \R options with similar syntax. More than one of them may 396appear. For the first three, d is a decimal number. 397<pre> 398 (*LIMIT_DEPTH=d) set the backtracking limit to d 399 (*LIMIT_HEAP=d) set the heap size limit to d * 1024 bytes 400 (*LIMIT_MATCH=d) set the match limit to d 401 (*NOTEMPTY) set PCRE2_NOTEMPTY when matching 402 (*NOTEMPTY_ATSTART) set PCRE2_NOTEMPTY_ATSTART when matching 403 (*NO_AUTO_POSSESS) no auto-possessification (PCRE2_NO_AUTO_POSSESS) 404 (*NO_DOTSTAR_ANCHOR) no .* anchoring (PCRE2_NO_DOTSTAR_ANCHOR) 405 (*NO_JIT) disable JIT optimization 406 (*NO_START_OPT) no start-match optimization (PCRE2_NO_START_OPTIMIZE) 407 (*UTF) set appropriate UTF mode for the library in use 408 (*UCP) set PCRE2_UCP (use Unicode properties for \d etc) 409</pre> 410Note that LIMIT_DEPTH, LIMIT_HEAP, and LIMIT_MATCH can only reduce the value of 411the limits set by the caller of <b>pcre2_match()</b> or <b>pcre2_dfa_match()</b>, 412not increase them. LIMIT_RECURSION is an obsolete synonym for LIMIT_DEPTH. The 413application can lock out the use of (*UTF) and (*UCP) by setting the 414PCRE2_NEVER_UTF or PCRE2_NEVER_UCP options, respectively, at compile time. 415</P> 416<br><a name="SEC19" href="#TOC1">NEWLINE CONVENTION</a><br> 417<P> 418These are recognized only at the very start of the pattern or after option 419settings with a similar syntax. 420<pre> 421 (*CR) carriage return only 422 (*LF) linefeed only 423 (*CRLF) carriage return followed by linefeed 424 (*ANYCRLF) all three of the above 425 (*ANY) any Unicode newline sequence 426 (*NUL) the NUL character (binary zero) 427</PRE> 428</P> 429<br><a name="SEC20" href="#TOC1">WHAT \R MATCHES</a><br> 430<P> 431These are recognized only at the very start of the pattern or after option 432setting with a similar syntax. 433<pre> 434 (*BSR_ANYCRLF) CR, LF, or CRLF 435 (*BSR_UNICODE) any Unicode newline sequence 436</PRE> 437</P> 438<br><a name="SEC21" href="#TOC1">LOOKAHEAD AND LOOKBEHIND ASSERTIONS</a><br> 439<P> 440<pre> 441 (?=...) ) 442 (*pla:...) ) positive lookahead 443 (*positive_lookahead:...) ) 444 445 (?!...) ) 446 (*nla:...) ) negative lookahead 447 (*negative_lookahead:...) ) 448 449 (?<=...) ) 450 (*plb:...) ) positive lookbehind 451 (*positive_lookbehind:...) ) 452 453 (?<!...) ) 454 (*nlb:...) ) negative lookbehind 455 (*negative_lookbehind:...) ) 456</pre> 457Each top-level branch of a lookbehind must be of a fixed length. 458</P> 459<br><a name="SEC22" href="#TOC1">NON-ATOMIC LOOKAROUND ASSERTIONS</a><br> 460<P> 461These assertions are specific to PCRE2 and are not Perl-compatible. 462<pre> 463 (?*...) ) 464 (*napla:...) ) synonyms 465 (*non_atomic_positive_lookahead:...) ) 466 467 (?<*...) ) 468 (*naplb:...) ) synonyms 469 (*non_atomic_positive_lookbehind:...) ) 470</PRE> 471</P> 472<br><a name="SEC23" href="#TOC1">SCRIPT RUNS</a><br> 473<P> 474<pre> 475 (*script_run:...) ) script run, can be backtracked into 476 (*sr:...) ) 477 478 (*atomic_script_run:...) ) atomic script run 479 (*asr:...) ) 480</PRE> 481</P> 482<br><a name="SEC24" href="#TOC1">BACKREFERENCES</a><br> 483<P> 484<pre> 485 \n reference by number (can be ambiguous) 486 \gn reference by number 487 \g{n} reference by number 488 \g+n relative reference by number (PCRE2 extension) 489 \g-n relative reference by number 490 \g{+n} relative reference by number (PCRE2 extension) 491 \g{-n} relative reference by number 492 \k<name> reference by name (Perl) 493 \k'name' reference by name (Perl) 494 \g{name} reference by name (Perl) 495 \k{name} reference by name (.NET) 496 (?P=name) reference by name (Python) 497</PRE> 498</P> 499<br><a name="SEC25" href="#TOC1">SUBROUTINE REFERENCES (POSSIBLY RECURSIVE)</a><br> 500<P> 501<pre> 502 (?R) recurse whole pattern 503 (?n) call subroutine by absolute number 504 (?+n) call subroutine by relative number 505 (?-n) call subroutine by relative number 506 (?&name) call subroutine by name (Perl) 507 (?P>name) call subroutine by name (Python) 508 \g<name> call subroutine by name (Oniguruma) 509 \g'name' call subroutine by name (Oniguruma) 510 \g<n> call subroutine by absolute number (Oniguruma) 511 \g'n' call subroutine by absolute number (Oniguruma) 512 \g<+n> call subroutine by relative number (PCRE2 extension) 513 \g'+n' call subroutine by relative number (PCRE2 extension) 514 \g<-n> call subroutine by relative number (PCRE2 extension) 515 \g'-n' call subroutine by relative number (PCRE2 extension) 516</PRE> 517</P> 518<br><a name="SEC26" href="#TOC1">CONDITIONAL PATTERNS</a><br> 519<P> 520<pre> 521 (?(condition)yes-pattern) 522 (?(condition)yes-pattern|no-pattern) 523 524 (?(n) absolute reference condition 525 (?(+n) relative reference condition 526 (?(-n) relative reference condition 527 (?(<name>) named reference condition (Perl) 528 (?('name') named reference condition (Perl) 529 (?(name) named reference condition (PCRE2, deprecated) 530 (?(R) overall recursion condition 531 (?(Rn) specific numbered group recursion condition 532 (?(R&name) specific named group recursion condition 533 (?(DEFINE) define groups for reference 534 (?(VERSION[>]=n.m) test PCRE2 version 535 (?(assert) assertion condition 536</pre> 537Note the ambiguity of (?(R) and (?(Rn) which might be named reference 538conditions or recursion tests. Such a condition is interpreted as a reference 539condition if the relevant named group exists. 540</P> 541<br><a name="SEC27" href="#TOC1">BACKTRACKING CONTROL</a><br> 542<P> 543All backtracking control verbs may be in the form (*VERB:NAME). For (*MARK) the 544name is mandatory, for the others it is optional. (*SKIP) changes its behaviour 545if :NAME is present. The others just set a name for passing back to the caller, 546but this is not a name that (*SKIP) can see. The following act immediately they 547are reached: 548<pre> 549 (*ACCEPT) force successful match 550 (*FAIL) force backtrack; synonym (*F) 551 (*MARK:NAME) set name to be passed back; synonym (*:NAME) 552</pre> 553The following act only when a subsequent match failure causes a backtrack to 554reach them. They all force a match failure, but they differ in what happens 555afterwards. Those that advance the start-of-match point do so only if the 556pattern is not anchored. 557<pre> 558 (*COMMIT) overall failure, no advance of starting point 559 (*PRUNE) advance to next starting character 560 (*SKIP) advance to current matching position 561 (*SKIP:NAME) advance to position corresponding to an earlier 562 (*MARK:NAME); if not found, the (*SKIP) is ignored 563 (*THEN) local failure, backtrack to next alternation 564</pre> 565The effect of one of these verbs in a group called as a subroutine is confined 566to the subroutine call. 567</P> 568<br><a name="SEC28" href="#TOC1">CALLOUTS</a><br> 569<P> 570<pre> 571 (?C) callout (assumed number 0) 572 (?Cn) callout with numerical data n 573 (?C"text") callout with string data 574</pre> 575The allowed string delimiters are ` ' " ^ % # $ (which are the same for the 576start and the end), and the starting delimiter { matched with the ending 577delimiter }. To encode the ending delimiter within the string, double it. 578</P> 579<br><a name="SEC29" href="#TOC1">SEE ALSO</a><br> 580<P> 581<b>pcre2pattern</b>(3), <b>pcre2api</b>(3), <b>pcre2callout</b>(3), 582<b>pcre2matching</b>(3), <b>pcre2</b>(3). 583</P> 584<br><a name="SEC30" href="#TOC1">AUTHOR</a><br> 585<P> 586Philip Hazel 587<br> 588Retired from University Computing Service 589<br> 590Cambridge, England. 591<br> 592</P> 593<br><a name="SEC31" href="#TOC1">REVISION</a><br> 594<P> 595Last updated: 12 January 2022 596<br> 597Copyright © 1997-2022 University of Cambridge. 598<br> 599<p> 600Return to the <a href="index.html">PCRE2 index page</a>. 601</p> 602