1=encoding utf8 2 3=head1 NAME 4 5perl5220delta - what is new for perl v5.22.0 6 7=head1 DESCRIPTION 8 9This document describes differences between the 5.20.0 release and the 5.22.0 10release. 11 12If you are upgrading from an earlier release such as 5.18.0, first read 13L<perl5200delta>, which describes differences between 5.18.0 and 5.20.0. 14 15=head1 Core Enhancements 16 17=head2 New bitwise operators 18 19A new experimental facility has been added that makes the four standard 20bitwise operators (C<& | ^ ~>) treat their operands consistently as 21numbers, and introduces four new dotted operators (C<&. |. ^. ~.>) that 22treat their operands consistently as strings. The same applies to the 23assignment variants (C<&= |= ^= &.= |.= ^.=>). 24 25To use this, enable the "bitwise" feature and disable the 26"experimental::bitwise" warnings category. See L<perlop/Bitwise String 27Operators> for details. 28L<[GH #14348]|https://github.com/Perl/perl5/issues/14348>. 29 30=head2 New double-diamond operator 31 32C<<< <<>> >>> is like C<< <> >> but uses three-argument C<open> to open 33each file in C<@ARGV>. This means that each element of C<@ARGV> will be treated 34as an actual file name, and C<"|foo"> won't be treated as a pipe open. 35 36=head2 New C<\b> boundaries in regular expressions 37 38=head3 C<qr/\b{gcb}/> 39 40C<gcb> stands for Grapheme Cluster Boundary. It is a Unicode property 41that finds the boundary between sequences of characters that look like a 42single character to a native speaker of a language. Perl has long had 43the ability to deal with these through the C<\X> regular escape 44sequence. Now, there is an alternative way of handling these. See 45L<perlrebackslash/\b{}, \b, \B{}, \B> for details. 46 47=head3 C<qr/\b{wb}/> 48 49C<wb> stands for Word Boundary. It is a Unicode property 50that finds the boundary between words. This is similar to the plain 51C<\b> (without braces) but is more suitable for natural language 52processing. It knows, for example, that apostrophes can occur in the 53middle of words. See L<perlrebackslash/\b{}, \b, \B{}, \B> for details. 54 55=head3 C<qr/\b{sb}/> 56 57C<sb> stands for Sentence Boundary. It is a Unicode property 58to aid in parsing natural language sentences. 59See L<perlrebackslash/\b{}, \b, \B{}, \B> for details. 60 61=head2 Non-Capturing Regular Expression Flag 62 63Regular expressions now support a C</n> flag that disables capturing 64and filling in C<$1>, C<$2>, etc inside of groups: 65 66 "hello" =~ /(hi|hello)/n; # $1 is not set 67 68This is equivalent to putting C<?:> at the beginning of every capturing group. 69 70See L<perlre/"n"> for more information. 71 72=head2 C<use re 'strict'> 73 74This applies stricter syntax rules to regular expression patterns 75compiled within its scope. This will hopefully alert you to typos and 76other unintentional behavior that backwards-compatibility issues prevent 77us from reporting in normal regular expression compilations. Because the 78behavior of this is subject to change in future Perl releases as we gain 79experience, using this pragma will raise a warning of category 80C<experimental::re_strict>. 81See L<'strict' in re|re/'strict' mode>. 82 83=head2 Unicode 7.0 (with correction) is now supported 84 85For details on what is in this release, see 86L<http://www.unicode.org/versions/Unicode7.0.0/>. 87The version of Unicode 7.0 that comes with Perl includes 88a correction dealing with glyph shaping in Arabic 89(see L<http://www.unicode.org/errata/#current_errata>). 90 91 92=head2 S<C<use locale>> can restrict which locale categories are affected 93 94It is now possible to pass a parameter to S<C<use locale>> to specify 95a subset of locale categories to be locale-aware, with the remaining 96ones unaffected. See L<perllocale/The "use locale" pragma> for details. 97 98=head2 Perl now supports POSIX 2008 locale currency additions 99 100On platforms that are able to handle POSIX.1-2008, the 101hash returned by 102L<C<POSIX::localeconv()>|perllocale/The localeconv function> 103includes the international currency fields added by that version of the 104POSIX standard. These are 105C<int_n_cs_precedes>, 106C<int_n_sep_by_space>, 107C<int_n_sign_posn>, 108C<int_p_cs_precedes>, 109C<int_p_sep_by_space>, 110and 111C<int_p_sign_posn>. 112 113=head2 Better heuristics on older platforms for determining locale UTF-8ness 114 115On platforms that implement neither the C99 standard nor the POSIX 2001 116standard, determining if the current locale is UTF-8 or not depends on 117heuristics. These are improved in this release. 118 119=head2 Aliasing via reference 120 121Variables and subroutines can now be aliased by assigning to a reference: 122 123 \$c = \$d; 124 \&x = \&y; 125 126Aliasing can also be accomplished 127by using a backslash before a C<foreach> iterator variable; this is 128perhaps the most useful idiom this feature provides: 129 130 foreach \%hash (@array_of_hash_refs) { ... } 131 132This feature is experimental and must be enabled via S<C<use feature 133'refaliasing'>>. It will warn unless the C<experimental::refaliasing> 134warnings category is disabled. 135 136See L<perlref/Assigning to References> 137 138=head2 C<prototype> with no arguments 139 140C<prototype()> with no arguments now infers C<$_>. 141L<[GH #14376]|https://github.com/Perl/perl5/issues/14376>. 142 143=head2 New C<:const> subroutine attribute 144 145The C<const> attribute can be applied to an anonymous subroutine. It 146causes the new sub to be executed immediately whenever one is created 147(I<i.e.> when the C<sub> expression is evaluated). Its value is captured 148and used to create a new constant subroutine that is returned. This 149feature is experimental. See L<perlsub/Constant Functions>. 150 151=head2 C<fileno> now works on directory handles 152 153When the relevant support is available in the operating system, the 154C<fileno> builtin now works on directory handles, yielding the 155underlying file descriptor in the same way as for filehandles. On 156operating systems without such support, C<fileno> on a directory handle 157continues to return the undefined value, as before, but also sets C<$!> to 158indicate that the operation is not supported. 159 160Currently, this uses either a C<dd_fd> member in the OS C<DIR> 161structure, or a C<dirfd(3)> function as specified by POSIX.1-2008. 162 163=head2 List form of pipe open implemented for Win32 164 165The list form of pipe: 166 167 open my $fh, "-|", "program", @arguments; 168 169is now implemented on Win32. It has the same limitations as C<system 170LIST> on Win32, since the Win32 API doesn't accept program arguments 171as a list. 172 173=head2 Assignment to list repetition 174 175C<(...) x ...> can now be used within a list that is assigned to, as long 176as the left-hand side is a valid lvalue. This allows S<C<(undef,undef,$foo) 177= that_function()>> to be written as S<C<((undef)x2, $foo) = that_function()>>. 178 179=head2 Infinity and NaN (not-a-number) handling improved 180 181Floating point values are able to hold the special values infinity, negative 182infinity, and NaN (not-a-number). Now we more robustly recognize and 183propagate the value in computations, and on output normalize them to the strings 184C<Inf>, C<-Inf>, and C<NaN>. 185 186See also the L<POSIX> enhancements. 187 188=head2 Floating point parsing has been improved 189 190Parsing and printing of floating point values has been improved. 191 192As a completely new feature, hexadecimal floating point literals 193(like C<0x1.23p-4>) are now supported, and they can be output with 194S<C<printf "%a">>. See L<perldata/Scalar value constructors> for more 195details. 196 197=head2 Packing infinity or not-a-number into a character is now fatal 198 199Before, when trying to pack infinity or not-a-number into a 200(signed) character, Perl would warn, and assumed you tried to 201pack C<< 0xFF >>; if you gave it as an argument to C<< chr >>, 202C<< U+FFFD >> was returned. 203 204But now, all such actions (C<< pack >>, C<< chr >>, and C<< print '%c' >>) 205result in a fatal error. 206 207=head2 Experimental C Backtrace API 208 209Perl now supports (via a C level API) retrieving 210the C level backtrace (similar to what symbolic debuggers like gdb do). 211 212The backtrace returns the stack trace of the C call frames, 213with the symbol names (function names), the object names (like "perl"), 214and if it can, also the source code locations (file:line). 215 216The supported platforms are Linux and OS X (some *BSD might work at 217least partly, but they have not yet been tested). 218 219The feature needs to be enabled with C<Configure -Dusecbacktrace>. 220 221See L<perlhacktips/"C backtrace"> for more information. 222 223=head1 Security 224 225=head2 Perl is now compiled with C<-fstack-protector-strong> if available 226 227Perl has been compiled with the anti-stack-smashing option 228C<-fstack-protector> since 5.10.1. Now Perl uses the newer variant 229called C<-fstack-protector-strong>, if available. 230 231=head2 The L<Safe> module could allow outside packages to be replaced 232 233Critical bugfix: outside packages could be replaced. L<Safe> has 234been patched to 2.38 to address this. 235 236=head2 Perl is now always compiled with C<-D_FORTIFY_SOURCE=2> if available 237 238The 'code hardening' option called C<_FORTIFY_SOURCE>, available in 239gcc 4.*, is now always used for compiling Perl, if available. 240 241Note that this isn't necessarily a huge step since in many platforms 242the step had already been taken several years ago: many Linux 243distributions (like Fedora) have been using this option for Perl, 244and OS X has enforced the same for many years. 245 246=head1 Incompatible Changes 247 248=head2 Subroutine signatures moved before attributes 249 250The experimental sub signatures feature, as introduced in 5.20, parsed 251signatures after attributes. In this release, following feedback from users 252of the experimental feature, the positioning has been moved such that 253signatures occur after the subroutine name (if any) and before the attribute 254list (if any). 255 256=head2 C<&> and C<\&> prototypes accepts only subs 257 258The C<&> prototype character now accepts only anonymous subs (C<sub 259{...}>), things beginning with C<\&>, or an explicit C<undef>. Formerly 260it erroneously also allowed references to arrays, hashes, and lists. 261L<[GH #2776]|https://github.com/Perl/perl5/issues/2776>. 262L<[GH #14186]|https://github.com/Perl/perl5/issues/14186>. 263L<[GH #14353]|https://github.com/Perl/perl5/issues/14353>. 264 265In addition, the C<\&> prototype was allowing subroutine calls, whereas 266now it only allows subroutines: C<&foo> is still permitted as an argument, 267while C<&foo()> and C<foo()> no longer are. 268L<[GH #10633]|https://github.com/Perl/perl5/issues/10633>. 269 270=head2 C<use encoding> is now lexical 271 272The L<encoding> pragma's effect is now limited to lexical scope. This 273pragma is deprecated, but in the meantime, it could adversely affect 274unrelated modules that are included in the same program; this change 275fixes that. 276 277=head2 List slices returning empty lists 278 279List slices now return an empty list only if the original list was empty 280(or if there are no indices). Formerly, a list slice would return an empty 281list if all indices fell outside the original list; now it returns a list 282of C<undef> values in that case. 283L<[GH #12335]|https://github.com/Perl/perl5/issues/12335>. 284 285=head2 C<\N{}> with a sequence of multiple spaces is now a fatal error 286 287E.g. S<C<\N{TOOE<nbsp>E<nbsp>MANY SPACES}>> or S<C<\N{TRAILING SPACE }>>. 288This has been deprecated since v5.18. 289 290=head2 S<C<use UNIVERSAL '...'>> is now a fatal error 291 292Importing functions from C<UNIVERSAL> has been deprecated since v5.12, and 293is now a fatal error. S<C<use UNIVERSAL>> without any arguments is still 294allowed. 295 296=head2 In double-quotish C<\cI<X>>, I<X> must now be a printable ASCII character 297 298In prior releases, failure to do this raised a deprecation warning. 299 300=head2 Splitting the tokens C<(?> and C<(*> in regular expressions is now a fatal compilation error. 301 302These had been deprecated since v5.18. 303 304=head2 C<qr/foo/x> now ignores all Unicode pattern white space 305 306The C</x> regular expression modifier allows the pattern to contain 307white space and comments (both of which are ignored) for improved 308readability. Until now, not all the white space characters that Unicode 309designates for this purpose were handled. The additional ones now 310recognized are: 311 312 U+0085 NEXT LINE 313 U+200E LEFT-TO-RIGHT MARK 314 U+200F RIGHT-TO-LEFT MARK 315 U+2028 LINE SEPARATOR 316 U+2029 PARAGRAPH SEPARATOR 317 318The use of these characters with C</x> outside bracketed character 319classes and when not preceded by a backslash has raised a deprecation 320warning since v5.18. Now they will be ignored. 321 322=head2 Comment lines within S<C<(?[ ])>> are now ended only by a C<\n> 323 324S<C<(?[ ])>> is an experimental feature, introduced in v5.18. It operates 325as if C</x> is always enabled. But there was a difference: comment 326lines (following a C<#> character) were terminated by anything matching 327C<\R> which includes all vertical whitespace, such as form feeds. For 328consistency, this is now changed to match what terminates comment lines 329outside S<C<(?[ ])>>, namely a C<\n> (even if escaped), which is the 330same as what terminates a heredoc string and formats. 331 332=head2 C<(?[...])> operators now follow standard Perl precedence 333 334This experimental feature allows set operations in regular expression patterns. 335Prior to this, the intersection operator had the same precedence as the other 336binary operators. Now it has higher precedence. This could lead to different 337outcomes than existing code expects (though the documentation has always noted 338that this change might happen, recommending fully parenthesizing the 339expressions). See L<perlrecharclass/Extended Bracketed Character Classes>. 340 341=head2 Omitting C<%> and C<@> on hash and array names is no longer permitted 342 343Really old Perl let you omit the C<@> on array names and the C<%> on hash 344names in some spots. This has issued a deprecation warning since Perl 3455.000, and is no longer permitted. 346 347=head2 C<"$!"> text is now in English outside the scope of C<use locale> 348 349Previously, the text, unlike almost everything else, always came out 350based on the current underlying locale of the program. (Also affected 351on some systems is C<"$^E">.) For programs that are unprepared to 352handle locale differences, this can cause garbage text to be displayed. 353It's better to display text that is translatable via some tool than 354garbage text which is much harder to figure out. 355 356=head2 C<"$!"> text will be returned in UTF-8 when appropriate 357 358The stringification of C<$!> and C<$^E> will have the UTF-8 flag set 359when the text is actually non-ASCII UTF-8. This will enable programs 360that are set up to be locale-aware to properly output messages in the 361user's native language. Code that needs to continue the 5.20 and 362earlier behavior can do the stringification within the scopes of both 363S<C<use bytes>> and S<C<use locale ":messages">>. Within these two 364scopes, no other Perl operations will 365be affected by locale; only C<$!> and C<$^E> stringification. The 366C<bytes> pragma causes the UTF-8 flag to not be set, just as in previous 367Perl releases. This resolves 368L<[GH #12035]|https://github.com/Perl/perl5/issues/12035>. 369 370=head2 Support for C<?PATTERN?> without explicit operator has been removed 371 372The C<m?PATTERN?> construct, which allows matching a regex only once, 373previously had an alternative form that was written directly with a question 374mark delimiter, omitting the explicit C<m> operator. This usage has produced 375a deprecation warning since 5.14.0. It is now a syntax error, so that the 376question mark can be available for use in new operators. 377 378=head2 C<defined(@array)> and C<defined(%hash)> are now fatal errors 379 380These have been deprecated since v5.6.1 and have raised deprecation 381warnings since v5.16. 382 383=head2 Using a hash or an array as a reference are now fatal errors 384 385For example, C<< %foo->{"bar"} >> now causes a fatal compilation 386error. These have been deprecated since before v5.8, and have raised 387deprecation warnings since then. 388 389=head2 Changes to the C<*> prototype 390 391The C<*> character in a subroutine's prototype used to allow barewords to take 392precedence over most, but not all, subroutine names. It was never 393consistent and exhibited buggy behavior. 394 395Now it has been changed, so subroutines always take precedence over barewords, 396which brings it into conformity with similarly prototyped built-in functions: 397 398 sub splat(*) { ... } 399 sub foo { ... } 400 splat(foo); # now always splat(foo()) 401 splat(bar); # still splat('bar') as before 402 close(foo); # close(foo()) 403 close(bar); # close('bar') 404 405=head1 Deprecations 406 407=head2 Setting C<${^ENCODING}> to anything but C<undef> 408 409This variable allows Perl scripts to be written in an encoding other than 410ASCII or UTF-8. However, it affects all modules globally, leading 411to wrong answers and segmentation faults. New scripts should be written 412in UTF-8; old scripts should be converted to UTF-8, which is easily done 413with the L<piconv> utility. 414 415=head2 Use of non-graphic characters in single-character variable names 416 417The syntax for single-character variable names is more lenient than 418for longer variable names, allowing the one-character name to be a 419punctuation character or even invisible (a non-graphic). Perl v5.20 420deprecated the ASCII-range controls as such a name. Now, all 421non-graphic characters that formerly were allowed are deprecated. 422The practical effect of this occurs only when not under C<S<use 423utf8>>, and affects just the C1 controls (code points 0x80 through 4240xFF), NO-BREAK SPACE, and SOFT HYPHEN. 425 426=head2 Inlining of C<sub () { $var }> with observable side-effects 427 428In many cases Perl makes S<C<sub () { $var }>> into an inlinable constant 429subroutine, capturing the value of C<$var> at the time the C<sub> expression 430is evaluated. This can break the closure behavior in those cases where 431C<$var> is subsequently modified, since the subroutine won't return the 432changed value. (Note that this all only applies to anonymous subroutines 433with an empty prototype (S<C<sub ()>>).) 434 435This usage is now deprecated in those cases where the variable could be 436modified elsewhere. Perl detects those cases and emits a deprecation 437warning. Such code will likely change in the future and stop producing a 438constant. 439 440If your variable is only modified in the place where it is declared, then 441Perl will continue to make the sub inlinable with no warnings. 442 443 sub make_constant { 444 my $var = shift; 445 return sub () { $var }; # fine 446 } 447 448 sub make_constant_deprecated { 449 my $var; 450 $var = shift; 451 return sub () { $var }; # deprecated 452 } 453 454 sub make_constant_deprecated2 { 455 my $var = shift; 456 log_that_value($var); # could modify $var 457 return sub () { $var }; # deprecated 458 } 459 460In the second example above, detecting that C<$var> is assigned to only once 461is too hard to detect. That it happens in a spot other than the C<my> 462declaration is enough for Perl to find it suspicious. 463 464This deprecation warning happens only for a simple variable for the body of 465the sub. (A C<BEGIN> block or C<use> statement inside the sub is ignored, 466because it does not become part of the sub's body.) For more complex 467cases, such as S<C<sub () { do_something() if 0; $var }>> the behavior has 468changed such that inlining does not happen if the variable is modifiable 469elsewhere. Such cases should be rare. 470 471=head2 Use of multiple C</x> regexp modifiers 472 473It is now deprecated to say something like any of the following: 474 475 qr/foo/xx; 476 /(?xax:foo)/; 477 use re qw(/amxx); 478 479That is, now C<x> should only occur once in any string of contiguous 480regular expression pattern modifiers. We do not believe there are any 481occurrences of this in all of CPAN. This is in preparation for a future 482Perl release having C</xx> permit white-space for readability in 483bracketed character classes (those enclosed in square brackets: 484C<[...]>). 485 486=head2 Using a NO-BREAK space in a character alias for C<\N{...}> is now deprecated 487 488This non-graphic character is essentially indistinguishable from a 489regular space, and so should not be allowed. See 490L<charnames/CUSTOM ALIASES>. 491 492=head2 A literal C<"{"> should now be escaped in a pattern 493 494If you want a literal left curly bracket (also called a left brace) in a 495regular expression pattern, you should now escape it by either 496preceding it with a backslash (C<"\{">) or enclosing it within square 497brackets C<"[{]">, or by using C<\Q>; otherwise a deprecation warning 498will be raised. This was first announced as forthcoming in the v5.16 499release; it will allow future extensions to the language to happen. 500 501=head2 Making all warnings fatal is discouraged 502 503The documentation for L<fatal warnings|warnings/Fatal Warnings> notes that 504C<< use warnings FATAL => 'all' >> is discouraged, and provides stronger 505language about the risks of fatal warnings in general. 506 507=head1 Performance Enhancements 508 509=over 4 510 511=item * 512 513If a method or class name is known at compile time, a hash is precomputed 514to speed up run-time method lookup. Also, compound method names like 515C<SUPER::new> are parsed at compile time, to save having to parse them at 516run time. 517 518=item * 519 520Array and hash lookups (especially nested ones) that use only constants 521or simple variables as keys, are now considerably faster. See 522L</Internal Changes> for more details. 523 524=item * 525 526C<(...)x1>, C<("constant")x0> and C<($scalar)x0> are now optimised in list 527context. If the right-hand argument is a constant 1, the repetition 528operator disappears. If the right-hand argument is a constant 0, the whole 529expression is optimised to the empty list, so long as the left-hand 530argument is a simple scalar or constant. (That is, C<(foo())x0> is not 531subject to this optimisation.) 532 533=item * 534 535C<substr> assignment is now optimised into 4-argument C<substr> at the end 536of a subroutine (or as the argument to C<return>). Previously, this 537optimisation only happened in void context. 538 539=item * 540 541In C<"\L...">, C<"\Q...">, etc., the extra "stringify" op is now optimised 542away, making these just as fast as C<lcfirst>, C<quotemeta>, etc. 543 544=item * 545 546Assignment to an empty list is now sometimes faster. In particular, it 547never calls C<FETCH> on tied arguments on the right-hand side, whereas it 548used to sometimes. 549 550=item * 551 552There is a performance improvement of up to 20% when C<length> is applied to 553a non-magical, non-tied string, and either C<use bytes> is in scope or the 554string doesn't use UTF-8 internally. 555 556=item * 557 558On most perl builds with 64-bit integers, memory usage for non-magical, 559non-tied scalars containing only a floating point value has been reduced 560by between 8 and 32 bytes, depending on OS. 561 562=item * 563 564In C<@array = split>, the assignment can be optimized away, so that C<split> 565writes directly to the array. This optimisation was happening only for 566package arrays other than C<@_>, and only sometimes. Now this 567optimisation happens almost all the time. 568 569=item * 570 571C<join> is now subject to constant folding. So for example 572S<C<join "-", "a", "b">> is converted at compile-time to C<"a-b">. 573Moreover, C<join> with a scalar or constant for the separator and a 574single-item list to join is simplified to a stringification, and the 575separator doesn't even get evaluated. 576 577=item * 578 579C<qq(@array)> is implemented using two ops: a stringify op and a join op. 580If the C<qq> contains nothing but a single array, the stringification is 581optimized away. 582 583=item * 584 585S<C<our $var>> and S<C<our($s,@a,%h)>> in void context are no longer evaluated at 586run time. Even a whole sequence of S<C<our $foo;>> statements will simply be 587skipped over. The same applies to C<state> variables. 588 589=item * 590 591Many internal functions have been refactored to improve performance and reduce 592their memory footprints. 593L<[GH #13659]|https://github.com/Perl/perl5/issues/13659> 594L<[GH #13856]|https://github.com/Perl/perl5/issues/13856> 595L<[GH #13874]|https://github.com/Perl/perl5/issues/13874> 596 597=item * 598 599C<-T> and C<-B> filetests will return sooner when an empty file is detected. 600L<[GH #13686]|https://github.com/Perl/perl5/issues/13686> 601 602=item * 603 604Hash lookups where the key is a constant are faster. 605 606=item * 607 608Subroutines with an empty prototype and a body containing just C<undef> are now 609eligible for inlining. 610L<[GH #14077]|https://github.com/Perl/perl5/issues/14077> 611 612=item * 613 614Subroutines in packages no longer need to be stored in typeglobs: 615declaring a subroutine will now put a simple sub reference directly in the 616stash if possible, saving memory. The typeglob still notionally exists, 617so accessing it will cause the stash entry to be upgraded to a typeglob 618(I<i.e.> this is just an internal implementation detail). 619This optimization does not currently apply to XSUBs or exported 620subroutines, and method calls will undo it, since they cache things in 621typeglobs. 622L<[GH #13392]|https://github.com/Perl/perl5/issues/13392> 623 624=item * 625 626The functions C<utf8::native_to_unicode()> and C<utf8::unicode_to_native()> 627(see L<utf8>) are now optimized out on ASCII platforms. There is now not even 628a minimal performance hit in writing code portable between ASCII and EBCDIC 629platforms. 630 631=item * 632 633Win32 Perl uses 8 KB less of per-process memory than before for every perl 634process, because some data is now memory mapped from disk and shared 635between processes from the same perl binary. 636 637=back 638 639=head1 Modules and Pragmata 640 641=head2 Updated Modules and Pragmata 642 643Many of the libraries distributed with perl have been upgraded since v5.20.0. 644For a complete list of changes, run: 645 646 corelist --diff 5.20.0 5.22.0 647 648You can substitute your favorite version in place of 5.20.0, too. 649 650Some notable changes include: 651 652=over 4 653 654=item * 655 656L<Archive::Tar> has been upgraded to version 2.04. 657 658Tests can now be run in parallel. 659 660=item * 661 662L<attributes> has been upgraded to version 0.27. 663 664The usage of C<memEQs> in the XS has been corrected. 665L<[GH #14072]|https://github.com/Perl/perl5/issues/14072> 666 667Avoid reading beyond the end of a buffer. [perl #122629] 668 669=item * 670 671L<B> has been upgraded to version 1.58. 672 673It provides a new C<B::safename> function, based on the existing 674C<< B::GV->SAFENAME >>, that converts C<\cOPEN> to C<^OPEN>. 675 676Nulled COPs are now of class C<B::COP>, rather than C<B::OP>. 677 678C<B::REGEXP> objects now provide a C<qr_anoncv> method for accessing the 679implicit CV associated with C<qr//> things containing code blocks, and a 680C<compflags> method that returns the pertinent flags originating from the 681C<qr//blahblah> op. 682 683C<B::PMOP> now provides a C<pmregexp> method returning a C<B::REGEXP> object. 684Two new classes, C<B::PADNAME> and C<B::PADNAMELIST>, have been introduced. 685 686A bug where, after an ithread creation or pseudofork, special/immortal SVs in 687the child ithread/pseudoprocess did not have the correct class of 688C<B::SPECIAL>, has been fixed. 689The C<id> and C<outid> PADLIST methods have been added. 690 691=item * 692 693L<B::Concise> has been upgraded to version 0.996. 694 695Null ops that are part of the execution chain are now given sequence 696numbers. 697 698Private flags for nulled ops are now dumped with mnemonics as they would be 699for the non-nulled counterparts. 700 701=item * 702 703L<B::Deparse> has been upgraded to version 1.35. 704 705It now deparses C<+sub : attr { ... }> correctly at the start of a 706statement. Without the initial C<+>, C<sub> would be a statement label. 707 708C<BEGIN> blocks are now emitted in the right place most of the time, but 709the change unfortunately introduced a regression, in that C<BEGIN> blocks 710occurring just before the end of the enclosing block may appear below it 711instead. 712 713C<B::Deparse> no longer puts erroneous C<local> here and there, such as for 714C<LIST = tr/a//d>. [perl #119815] 715 716Adjacent C<use> statements are no longer accidentally nested if one 717contains a C<do> block. [perl #115066] 718 719Parenthesised arrays in lists passed to C<\> are now correctly deparsed 720with parentheses (I<e.g.>, C<\(@a, (@b), @c)> now retains the parentheses 721around @b), thus preserving the flattening behavior of referenced 722parenthesised arrays. Formerly, it only worked for one array: C<\(@a)>. 723 724C<local our> is now deparsed correctly, with the C<our> included. 725 726C<for($foo; !$bar; $baz) {...}> was deparsed without the C<!> (or C<not>). 727This has been fixed. 728 729Core keywords that conflict with lexical subroutines are now deparsed with 730the C<CORE::> prefix. 731 732C<foreach state $x (...) {...}> now deparses correctly with C<state> and 733not C<my>. 734 735C<our @array = split(...)> now deparses correctly with C<our> in those 736cases where the assignment is optimized away. 737 738It now deparses C<our(I<LIST>)> and typed lexical (C<my Dog $spot>) correctly. 739 740Deparse C<$#_> as that instead of as C<$#{_}>. 741L<[GH #14545]|https://github.com/Perl/perl5/issues/14545> 742 743BEGIN blocks at the end of the enclosing scope are now deparsed in the 744right place. [perl #77452] 745 746BEGIN blocks were sometimes deparsed as __ANON__, but are now always called 747BEGIN. 748 749Lexical subroutines are now fully deparsed. [perl #116553] 750 751C<Anything =~ y///r> with C</r> no longer omits the left-hand operand. 752 753The op trees that make up regexp code blocks are now deparsed for real. 754Formerly, the original string that made up the regular expression was used. 755That caused problems with C<qr/(?{E<lt>E<lt>heredoc})/> and multiline code blocks, 756which were deparsed incorrectly. [perl #123217] [perl #115256] 757 758C<$;> at the end of a statement no longer loses its semicolon. 759[perl #123357] 760 761Some cases of subroutine declarations stored in the stash in shorthand form 762were being omitted. 763 764Non-ASCII characters are now consistently escaped in strings, instead of 765some of the time. (There are still outstanding problems with regular 766expressions and identifiers that have not been fixed.) 767 768When prototype sub calls are deparsed with C<&> (I<e.g.>, under the B<-P> 769option), C<scalar> is now added where appropriate, to force the scalar 770context implied by the prototype. 771 772C<require(foo())>, C<do(foo())>, C<goto(foo())> and similar constructs with 773loop controls are now deparsed correctly. The outer parentheses are not 774optional. 775 776Whitespace is no longer escaped in regular expressions, because it was 777getting erroneously escaped within C<(?x:...)> sections. 778 779C<sub foo { foo() }> is now deparsed with those mandatory parentheses. 780 781C</@array/> is now deparsed as a regular expression, and not just 782C<@array>. 783 784C</@{-}/>, C</@{+}/> and C<$#{1}> are now deparsed with the braces, which 785are mandatory in these cases. 786 787In deparsing feature bundles, C<B::Deparse> was emitting C<no feature;> first 788instead of C<no feature ':all';>. This has been fixed. 789 790C<chdir FH> is now deparsed without quotation marks. 791 792C<\my @a> is now deparsed without parentheses. (Parenthese would flatten 793the array.) 794 795C<system> and C<exec> followed by a block are now deparsed correctly. 796Formerly there was an erroneous C<do> before the block. 797 798C<< use constant QR =E<gt> qr/.../flags >> followed by C<"" =~ QR> is no longer 799without the flags. 800 801Deparsing C<BEGIN { undef &foo }> with the B<-w> switch enabled started to 802emit 'uninitialized' warnings in Perl 5.14. This has been fixed. 803 804Deparsing calls to subs with a C<(;+)> prototype resulted in an infinite 805loop. The C<(;$>) C<(_)> and C<(;_)> prototypes were given the wrong 806precedence, causing C<foo($aE<lt>$b)> to be deparsed without the parentheses. 807 808Deparse now provides a defined state sub in inner subs. 809 810=item * 811 812L<B::Op_private> has been added. 813 814L<B::Op_private> provides detailed information about the flags used in the 815C<op_private> field of perl opcodes. 816 817=item * 818 819L<bigint>, L<bignum>, L<bigrat> have been upgraded to version 0.39. 820 821Document in CAVEATS that using strings as numbers won't always invoke 822the big number overloading, and how to invoke it. [rt.perl.org #123064] 823 824=item * 825 826L<Carp> has been upgraded to version 1.36. 827 828C<Carp::Heavy> now ignores version mismatches with Carp if Carp is newer 829than 1.12, since C<Carp::Heavy>'s guts were merged into Carp at that 830point. 831L<[GH #13708]|https://github.com/Perl/perl5/issues/13708> 832 833Carp now handles non-ASCII platforms better. 834 835Off-by-one error fix for Perl E<lt> 5.14. 836 837=item * 838 839L<constant> has been upgraded to version 1.33. 840 841It now accepts fully-qualified constant names, allowing constants to be defined 842in packages other than the caller. 843 844=item * 845 846L<CPAN> has been upgraded to version 2.11. 847 848Add support for C<Cwd::getdcwd()> and introduce workaround for a misbehavior 849seen on Strawberry Perl 5.20.1. 850 851Fix C<chdir()> after building dependencies bug. 852 853Introduce experimental support for plugins/hooks. 854 855Integrate the C<App::Cpan> sources. 856 857Do not check recursion on optional dependencies. 858 859Sanity check F<META.yml> to contain a hash. 860L<[cpan #95271]|https://rt.cpan.org/Ticket/Display.html?id=95271> 861 862=item * 863 864L<CPAN::Meta::Requirements> has been upgraded to version 2.132. 865 866Works around limitations in C<version::vpp> detecting v-string magic and adds 867support for forthcoming L<ExtUtils::MakeMaker> bootstrap F<version.pm> for 868Perls older than 5.10.0. 869 870=item * 871 872L<Data::Dumper> has been upgraded to version 2.158. 873 874Fixes CVE-2014-4330 by adding a configuration variable/option to limit 875recursion when dumping deep data structures. 876 877Changes to resolve Coverity issues. 878XS dumps incorrectly stored the name of code references stored in a 879GLOB. 880L<[GH #13911]|https://github.com/Perl/perl5/issues/13911> 881 882=item * 883 884L<DynaLoader> has been upgraded to version 1.32. 885 886Remove C<dl_nonlazy> global if unused in Dynaloader. [perl #122926] 887 888=item * 889 890L<Encode> has been upgraded to version 2.72. 891 892C<piconv> now has better error handling when the encoding name is nonexistent, 893and a build breakage when upgrading L<Encode> in perl-5.8.2 and earlier has 894been fixed. 895 896Building in C++ mode on Windows now works. 897 898=item * 899 900L<Errno> has been upgraded to version 1.23. 901 902Add C<-P> to the preprocessor command-line on GCC 5. GCC added extra 903line directives, breaking parsing of error code definitions. [rt.perl.org 904#123784] 905 906=item * 907 908L<experimental> has been upgraded to version 0.013. 909 910Hardcodes features for Perls older than 5.15.7. 911 912=item * 913 914L<ExtUtils::CBuilder> has been upgraded to version 0.280221. 915 916Fixes a regression on Android. 917L<[GH #14064]|https://github.com/Perl/perl5/issues/14064> 918 919=item * 920 921L<ExtUtils::Manifest> has been upgraded to version 1.70. 922 923Fixes a bug with C<maniread()>'s handling of quoted filenames and improves 924C<manifind()> to follow symlinks. 925L<[GH #14003]|https://github.com/Perl/perl5/issues/14003> 926 927=item * 928 929L<ExtUtils::ParseXS> has been upgraded to version 3.28. 930 931Only declare C<file> unused if we actually define it. 932Improve generated C<RETVAL> code generation to avoid repeated 933references to C<ST(0)>. [perl #123278] 934Broaden and document the C</OBJ$/> to C</REF$/> typemap optimization 935for the C<DESTROY> method. [perl #123418] 936 937=item * 938 939L<Fcntl> has been upgraded to version 1.13. 940 941Add support for the Linux pipe buffer size C<fcntl()> commands. 942 943=item * 944 945L<File::Find> has been upgraded to version 1.29. 946 947C<find()> and C<finddepth()> will now warn if passed inappropriate or 948misspelled options. 949 950=item * 951 952L<File::Glob> has been upgraded to version 1.24. 953 954Avoid C<SvIV()> expanding to call C<get_sv()> three times in a few 955places. [perl #123606] 956 957=item * 958 959L<HTTP::Tiny> has been upgraded to version 0.054. 960 961C<keep_alive> is now fork-safe and thread-safe. 962 963=item * 964 965L<IO> has been upgraded to version 1.35. 966 967The XS implementation has been fixed for the sake of older Perls. 968 969=item * 970 971L<IO::Socket> has been upgraded to version 1.38. 972 973Document the limitations of the C<connected()> method. [perl #123096] 974 975=item * 976 977L<IO::Socket::IP> has been upgraded to version 0.37. 978 979A better fix for subclassing C<connect()>. 980L<[cpan #95983]|https://rt.cpan.org/Ticket/Display.html?id=95983> 981L<[cpan #97050]|https://rt.cpan.org/Ticket/Display.html?id=97050> 982 983Implements Timeout for C<connect()>. 984L<[cpan #92075]|https://rt.cpan.org/Ticket/Display.html?id=92075> 985 986=item * 987 988The libnet collection of modules has been upgraded to version 3.05. 989 990Support for IPv6 and SSL to C<Net::FTP>, C<Net::NNTP>, C<Net::POP3> and C<Net::SMTP>. 991Improvements in C<Net::SMTP> authentication. 992 993=item * 994 995L<Locale::Codes> has been upgraded to version 3.34. 996 997Fixed a bug in the scripts used to extract data from spreadsheets that 998prevented the SHP currency code from being found. 999L<[cpan #94229]|https://rt.cpan.org/Ticket/Display.html?id=94229> 1000 1001New codes have been added. 1002 1003=item * 1004 1005L<Math::BigInt> has been upgraded to version 1.9997. 1006 1007Synchronize POD changes from the CPAN release. 1008C<< Math::BigFloat->blog(x) >> would sometimes return C<blog(2*x)> when 1009the accuracy was greater than 70 digits. 1010The result of C<< Math::BigFloat->bdiv() >> in list context now 1011satisfies C<< x = quotient * divisor + remainder >>. 1012 1013Correct handling of subclasses. 1014L<[cpan #96254]|https://rt.cpan.org/Ticket/Display.html?id=96254> 1015L<[cpan #96329]|https://rt.cpan.org/Ticket/Display.html?id=96329> 1016 1017=item * 1018 1019L<Module::Metadata> has been upgraded to version 1.000026. 1020 1021Support installations on older perls with an L<ExtUtils::MakeMaker> earlier 1022than 6.63_03 1023 1024=item * 1025 1026L<overload> has been upgraded to version 1.26. 1027 1028A redundant C<ref $sub> check has been removed. 1029 1030=item * 1031 1032The PathTools module collection has been upgraded to version 3.56. 1033 1034A warning from the B<gcc> compiler is now avoided when building the XS. 1035 1036Don't turn leading C<//> into C</> on Cygwin. [perl #122635] 1037 1038=item * 1039 1040L<perl5db.pl> has been upgraded to version 1.49. 1041 1042The debugger would cause an assertion failure. 1043L<[GH #14605]|https://github.com/Perl/perl5/issues/14605> 1044 1045C<fork()> in the debugger under C<tmux> will now create a new window for 1046the forked process. L<[GH #13602]|https://github.com/Perl/perl5/issues/13602> 1047 1048The debugger now saves the current working directory on startup and 1049restores it when you restart your program with C<R> or C<rerun>. 1050L<[GH #13691]|https://github.com/Perl/perl5/issues/13691> 1051 1052=item * 1053 1054L<PerlIO::scalar> has been upgraded to version 0.22. 1055 1056Reading from a position well past the end of the scalar now correctly 1057returns end of file. [perl #123443] 1058 1059Seeking to a negative position still fails, but no longer leaves the 1060file position set to a negation location. 1061 1062C<eof()> on a C<PerlIO::scalar> handle now properly returns true when 1063the file position is past the 2GB mark on 32-bit systems. 1064 1065Attempting to write at file positions impossible for the platform now 1066fail early rather than wrapping at 4GB. 1067 1068=item * 1069 1070L<Pod::Perldoc> has been upgraded to version 3.25. 1071 1072Filehandles opened for reading or writing now have C<:encoding(UTF-8)> set. 1073L<[cpan #98019]|https://rt.cpan.org/Ticket/Display.html?id=98019> 1074 1075=item * 1076 1077L<POSIX> has been upgraded to version 1.53. 1078 1079The C99 math functions and constants (for example C<acosh>, C<isinf>, C<isnan>, C<round>, 1080C<trunc>; C<M_E>, C<M_SQRT2>, C<M_PI>) have been added. 1081 1082C<POSIX::tmpnam()> now produces a deprecation warning. [perl #122005] 1083 1084=item * 1085 1086L<Safe> has been upgraded to version 2.39. 1087 1088C<reval> was not propagating void context properly. 1089 1090=item * 1091 1092Scalar-List-Utils has been upgraded to version 1.41. 1093 1094A new module, L<Sub::Util>, has been added, containing functions related to 1095CODE refs, including C<subname> (inspired by C<Sub::Identity>) and C<set_subname> 1096(copied and renamed from C<Sub::Name>). 1097The use of C<GetMagic> in C<List::Util::reduce()> has also been fixed. 1098L<[cpan #63211]|https://rt.cpan.org/Ticket/Display.html?id=63211> 1099 1100=item * 1101 1102L<SDBM_File> has been upgraded to version 1.13. 1103 1104Simplified the build process. [perl #123413] 1105 1106=item * 1107 1108L<Time::Piece> has been upgraded to version 1.29. 1109 1110When pretty printing negative C<Time::Seconds>, the "minus" is no longer lost. 1111 1112=item * 1113 1114L<Unicode::Collate> has been upgraded to version 1.12. 1115 1116Version 0.67's improved discontiguous contractions is invalidated by default 1117and is supported as a parameter C<long_contraction>. 1118 1119=item * 1120 1121L<Unicode::Normalize> has been upgraded to version 1.18. 1122 1123The XSUB implementation has been removed in favor of pure Perl. 1124 1125=item * 1126 1127L<Unicode::UCD> has been upgraded to version 0.61. 1128 1129A new function L<property_values()|Unicode::UCD/prop_values()> 1130has been added to return a given property's possible values. 1131 1132A new function L<charprop()|Unicode::UCD/charprop()> 1133has been added to return the value of a given property for a given code 1134point. 1135 1136A new function L<charprops_all()|Unicode::UCD/charprops_all()> 1137has been added to return the values of all Unicode properties for a 1138given code point. 1139 1140A bug has been fixed so that L<propaliases()|Unicode::UCD/prop_aliases()> 1141returns the correct short and long names for the Perl extensions where 1142it was incorrect. 1143 1144A bug has been fixed so that 1145L<prop_value_aliases()|Unicode::UCD/prop_value_aliases()> 1146returns C<undef> instead of a wrong result for properties that are Perl 1147extensions. 1148 1149This module now works on EBCDIC platforms. 1150 1151=item * 1152 1153L<utf8> has been upgraded to version 1.17 1154 1155A mismatch between the documentation and the code in C<utf8::downgrade()> 1156was fixed in favor of the documentation. The optional second argument 1157is now correctly treated as a perl boolean (true/false semantics) and 1158not as an integer. 1159 1160=item * 1161 1162L<version> has been upgraded to version 0.9909. 1163 1164Numerous changes. See the F<Changes> file in the CPAN distribution for 1165details. 1166 1167=item * 1168 1169L<Win32> has been upgraded to version 0.51. 1170 1171C<GetOSName()> now supports Windows 8.1, and building in C++ mode now works. 1172 1173=item * 1174 1175L<Win32API::File> has been upgraded to version 0.1202 1176 1177Building in C++ mode now works. 1178 1179=item * 1180 1181L<XSLoader> has been upgraded to version 0.20. 1182 1183Allow XSLoader to load modules from a different namespace. 1184[perl #122455] 1185 1186=back 1187 1188=head2 Removed Modules and Pragmata 1189 1190The following modules (and associated modules) have been removed from the core 1191perl distribution: 1192 1193=over 4 1194 1195=item * 1196 1197L<CGI> 1198 1199=item * 1200 1201L<Module::Build> 1202 1203=back 1204 1205=head1 Documentation 1206 1207=head2 New Documentation 1208 1209=head3 L<perlunicook> 1210 1211This document, by Tom Christiansen, provides examples of handling Unicode in 1212Perl. 1213 1214=head2 Changes to Existing Documentation 1215 1216=head3 L<perlaix> 1217 1218=over 4 1219 1220=item * 1221 1222A note on long doubles has been added. 1223 1224=back 1225 1226 1227=head3 L<perlapi> 1228 1229=over 4 1230 1231=item * 1232 1233Note that C<SvSetSV> doesn't do set magic. 1234 1235=item * 1236 1237C<sv_usepvn_flags> - fix documentation to mention the use of C<Newx> instead of 1238C<malloc>. 1239 1240L<[GH #13835]|https://github.com/Perl/perl5/issues/13835> 1241 1242=item * 1243 1244Clarify where C<NUL> may be embedded or is required to terminate a string. 1245 1246=item * 1247 1248Some documentation that was previously missing due to formatting errors is 1249now included. 1250 1251=item * 1252 1253Entries are now organized into groups rather than by the file where they 1254are found. 1255 1256=item * 1257 1258Alphabetical sorting of entries is now done consistently (automatically 1259by the POD generator) to make entries easier to find when scanning. 1260 1261=back 1262 1263=head3 L<perldata> 1264 1265=over 4 1266 1267=item * 1268 1269The syntax of single-character variable names has been brought 1270up-to-date and more fully explained. 1271 1272=item * 1273 1274Hexadecimal floating point numbers are described, as are infinity and 1275NaN. 1276 1277=back 1278 1279=head3 L<perlebcdic> 1280 1281=over 4 1282 1283=item * 1284 1285This document has been significantly updated in the light of recent 1286improvements to EBCDIC support. 1287 1288=back 1289 1290=head3 L<perlfilter> 1291 1292=over 4 1293 1294=item * 1295 1296Added a L<LIMITATIONS|perlfilter/LIMITATIONS> section. 1297 1298=back 1299 1300 1301=head3 L<perlfunc> 1302 1303=over 4 1304 1305=item * 1306 1307Mention that C<study()> is currently a no-op. 1308 1309=item * 1310 1311Calling C<delete> or C<exists> on array values is now described as "strongly 1312discouraged" rather than "deprecated". 1313 1314=item * 1315 1316Improve documentation of C<< our >>. 1317 1318=item * 1319 1320C<-l> now notes that it will return false if symlinks aren't supported by the 1321file system. 1322L<[GH #13695]|https://github.com/Perl/perl5/issues/13695> 1323 1324=item * 1325 1326Note that C<exec LIST> and C<system LIST> may fall back to the shell on 1327Win32. Only the indirect-object syntax C<exec PROGRAM LIST> and 1328C<system PROGRAM LIST> will reliably avoid using the shell. 1329 1330This has also been noted in L<perlport>. 1331 1332L<[GH #13907]|https://github.com/Perl/perl5/issues/13907> 1333 1334=back 1335 1336=head3 L<perlguts> 1337 1338=over 4 1339 1340=item * 1341 1342The OOK example has been updated to account for COW changes and a change in the 1343storage of the offset. 1344 1345=item * 1346 1347Details on C level symbols and libperl.t added. 1348 1349=item * 1350 1351Information on Unicode handling has been added 1352 1353=item * 1354 1355Information on EBCDIC handling has been added 1356 1357=back 1358 1359=head3 L<perlhack> 1360 1361=over 4 1362 1363=item * 1364 1365A note has been added about running on platforms with non-ASCII 1366character sets 1367 1368=item * 1369 1370A note has been added about performance testing 1371 1372=back 1373 1374=head3 L<perlhacktips> 1375 1376=over 4 1377 1378=item * 1379 1380Documentation has been added illustrating the perils of assuming that 1381there is no change to the contents of static memory pointed to by the 1382return values of Perl's wrappers for C library functions. 1383 1384=item * 1385 1386Replacements for C<tmpfile>, C<atoi>, C<strtol>, and C<strtoul> are now 1387recommended. 1388 1389=item * 1390 1391Updated documentation for the C<test.valgrind> C<make> target. 1392L<[GH #13658]|https://github.com/Perl/perl5/issues/13658> 1393 1394=item * 1395 1396Information is given about writing test files portably to non-ASCII 1397platforms. 1398 1399=item * 1400 1401A note has been added about how to get a C language stack backtrace. 1402 1403=back 1404 1405=head3 L<perlhpux> 1406 1407=over 4 1408 1409=item * 1410 1411Note that the message "Redeclaration of "sendpath" with a different 1412storage class specifier" is harmless. 1413 1414=back 1415 1416=head3 L<perllocale> 1417 1418=over 4 1419 1420=item * 1421 1422Updated for the enhancements in v5.22, along with some clarifications. 1423 1424=back 1425 1426=head3 L<perlmodstyle> 1427 1428=over 4 1429 1430=item * 1431 1432Instead of pointing to the module list, we are now pointing to 1433L<PrePAN|http://prepan.org/>. 1434 1435=back 1436 1437=head3 L<perlop> 1438 1439=over 4 1440 1441=item * 1442 1443Updated for the enhancements in v5.22, along with some clarifications. 1444 1445=back 1446 1447=head3 L<perlpodspec> 1448 1449=over 4 1450 1451=item * 1452 1453The specification of the pod language is changing so that the default 1454encoding of pods that aren't in UTF-8 (unless otherwise indicated) is 1455CP1252 instead of ISO 8859-1 (Latin1). 1456 1457=back 1458 1459=head3 L<perlpolicy> 1460 1461=over 4 1462 1463=item * 1464 1465We now have a code of conduct for the I<< p5p >> mailing list, as documented 1466in L<< perlpolicy/STANDARDS OF CONDUCT >>. 1467 1468=item * 1469 1470The conditions for marking an experimental feature as non-experimental are now 1471set out. 1472 1473=item * 1474 1475Clarification has been made as to what sorts of changes are permissible in 1476maintenance releases. 1477 1478=back 1479 1480=head3 L<perlport> 1481 1482=over 4 1483 1484=item * 1485 1486Out-of-date VMS-specific information has been fixed and/or simplified. 1487 1488=item * 1489 1490Notes about EBCDIC have been added. 1491 1492=back 1493 1494=head3 L<perlre> 1495 1496=over 4 1497 1498=item * 1499 1500The description of the C</x> modifier has been clarified to note that 1501comments cannot be continued onto the next line by escaping them; and 1502there is now a list of all the characters that are considered whitespace 1503by this modifier. 1504 1505=item * 1506 1507The new C</n> modifier is described. 1508 1509=item * 1510 1511A note has been added on how to make bracketed character class ranges 1512portable to non-ASCII machines. 1513 1514=back 1515 1516=head3 L<perlrebackslash> 1517 1518=over 4 1519 1520=item * 1521 1522Added documentation of C<\b{sb}>, C<\b{wb}>, C<\b{gcb}>, and C<\b{g}>. 1523 1524=back 1525 1526=head3 L<perlrecharclass> 1527 1528=over 4 1529 1530=item * 1531 1532Clarifications have been added to L<perlrecharclass/Character Ranges> 1533to the effect C<[A-Z]>, C<[a-z]>, C<[0-9]> and 1534any subranges thereof in regular expression bracketed character classes 1535are guaranteed to match exactly what a naive English speaker would 1536expect them to match, even on platforms (such as EBCDIC) where perl 1537has to do extra work to accomplish this. 1538 1539=item * 1540 1541The documentation of Bracketed Character Classes has been expanded to cover the 1542improvements in C<qr/[\N{named sequence}]/> (see under L</Selected Bug Fixes>). 1543 1544=back 1545 1546=head3 L<perlref> 1547 1548=over 4 1549 1550=item * 1551 1552A new section has been added 1553L<Assigning to References|perlref/Assigning to References> 1554 1555=back 1556 1557=head3 L<perlsec> 1558 1559=over 4 1560 1561=item * 1562 1563Comments added on algorithmic complexity and tied hashes. 1564 1565=back 1566 1567=head3 L<perlsyn> 1568 1569=over 4 1570 1571=item * 1572 1573An ambiguity in the documentation of the C<...> statement has been corrected. 1574L<[GH #14054]|https://github.com/Perl/perl5/issues/14054> 1575 1576=item * 1577 1578The empty conditional in C<< for >> and C<< while >> is now documented 1579in L<< perlsyn >>. 1580 1581=back 1582 1583=head3 L<perlunicode> 1584 1585=over 4 1586 1587=item * 1588 1589This has had extensive revisions to bring it up-to-date with current 1590Unicode support and to make it more readable. Notable is that Unicode 15917.0 changed what it should do with non-characters. Perl retains the old 1592way of handling for reasons of backward compatibility. See 1593L<perlunicode/Noncharacter code points>. 1594 1595=back 1596 1597=head3 L<perluniintro> 1598 1599=over 4 1600 1601=item * 1602 1603Advice for how to make sure your strings and regular expression patterns are 1604interpreted as Unicode has been updated. 1605 1606=back 1607 1608=head3 L<perlvar> 1609 1610=over 4 1611 1612=item * 1613 1614C<$]> is no longer listed as being deprecated. Instead, discussion has 1615been added on the advantages and disadvantages of using it versus 1616C<$^V>. C<$OLD_PERL_VERSION> was re-added to the documentation as the long 1617form of C<$]>. 1618 1619=item * 1620 1621C<${^ENCODING}> is now marked as deprecated. 1622 1623=item * 1624 1625The entry for C<%^H> has been clarified to indicate it can only handle 1626simple values. 1627 1628=back 1629 1630=head3 L<perlvms> 1631 1632=over 4 1633 1634=item * 1635 1636Out-of-date and/or incorrect material has been removed. 1637 1638=item * 1639 1640Updated documentation on environment and shell interaction in VMS. 1641 1642=back 1643 1644=head3 L<perlxs> 1645 1646=over 4 1647 1648=item * 1649 1650Added a discussion of locale issues in XS code. 1651 1652=back 1653 1654=head1 Diagnostics 1655 1656The following additions or changes have been made to diagnostic output, 1657including warnings and fatal error messages. For the complete list of 1658diagnostic messages, see L<perldiag>. 1659 1660=head2 New Diagnostics 1661 1662=head3 New Errors 1663 1664=over 4 1665 1666=item * 1667 1668L<Bad symbol for scalar|perldiag/"Bad symbol for scalar"> 1669 1670(P) An internal request asked to add a scalar entry to something that 1671wasn't a symbol table entry. 1672 1673=item * 1674 1675L<Can't use a hash as a reference|perldiag/"Can't use a hash as a reference"> 1676 1677(F) You tried to use a hash as a reference, as in 1678C<< %foo->{"bar"} >> or C<< %$ref->{"hello"} >>. Versions of perl E<lt>= 5.6.1 1679used to allow this syntax, but shouldn't have. 1680 1681=item * 1682 1683L<Can't use an array as a reference|perldiag/"Can't use an array as a reference"> 1684 1685(F) You tried to use an array as a reference, as in 1686C<< @foo->[23] >> or C<< @$ref->[99] >>. Versions of perl E<lt>= 5.6.1 used to 1687allow this syntax, but shouldn't have. 1688 1689=item * 1690 1691L<Can't use 'defined(@array)' (Maybe you should just omit the defined()?)|perldiag/"Can't use 'defined(@array)' (Maybe you should just omit the defined()?)"> 1692 1693(F) C<defined()> is not useful on arrays because it 1694checks for an undefined I<scalar> value. If you want to see if the 1695array is empty, just use S<C<if (@array) { # not empty }>> for example. 1696 1697=item * 1698 1699L<Can't use 'defined(%hash)' (Maybe you should just omit the defined()?)|perldiag/"Can't use 'defined(%hash)' (Maybe you should just omit the defined()?)"> 1700 1701(F) C<defined()> is not usually right on hashes. 1702 1703Although S<C<defined %hash>> is false on a plain not-yet-used hash, it 1704becomes true in several non-obvious circumstances, including iterators, 1705weak references, stash names, even remaining true after S<C<undef %hash>>. 1706These things make S<C<defined %hash>> fairly useless in practice, so it now 1707generates a fatal error. 1708 1709If a check for non-empty is what you wanted then just put it in boolean 1710context (see L<perldata/Scalar values>): 1711 1712 if (%hash) { 1713 # not empty 1714 } 1715 1716If you had S<C<defined %Foo::Bar::QUUX>> to check whether such a package 1717variable exists then that's never really been reliable, and isn't 1718a good way to enquire about the features of a package, or whether 1719it's loaded, etc. 1720 1721=item * 1722 1723L<Cannot chr %f|perldiag/"Cannot chr %f"> 1724 1725(F) You passed an invalid number (like an infinity or not-a-number) to 1726C<chr>. 1727 1728=item * 1729 1730L<Cannot compress %f in pack|perldiag/"Cannot compress %f in pack"> 1731 1732(F) You tried converting an infinity or not-a-number to an unsigned 1733character, which makes no sense. 1734 1735=item * 1736 1737L<Cannot pack %f with '%c'|perldiag/"Cannot pack %f with '%c'"> 1738 1739(F) You tried converting an infinity or not-a-number to a character, 1740which makes no sense. 1741 1742=item * 1743 1744L<Cannot print %f with '%c'|perldiag/"Cannot printf %f with '%c'"> 1745 1746(F) You tried printing an infinity or not-a-number as a character (C<%c>), 1747which makes no sense. Maybe you meant C<'%s'>, or just stringifying it? 1748 1749=item * 1750 1751L<charnames alias definitions may not contain a sequence of multiple spaces|perldiag/"charnames alias definitions may not contain a sequence of multiple spaces"> 1752 1753(F) You defined a character name which had multiple space 1754characters in a row. Change them to single spaces. Usually these 1755names are defined in the C<:alias> import argument to C<use charnames>, but 1756they could be defined by a translator installed into C<$^H{charnames}>. 1757See L<charnames/CUSTOM ALIASES>. 1758 1759=item * 1760 1761L<charnames alias definitions may not contain trailing white-space|perldiag/"charnames alias definitions may not contain trailing white-space"> 1762 1763(F) You defined a character name which ended in a space 1764character. Remove the trailing space(s). Usually these names are 1765defined in the C<:alias> import argument to C<use charnames>, but they 1766could be defined by a translator installed into C<$^H{charnames}>. 1767See L<charnames/CUSTOM ALIASES>. 1768 1769=item * 1770 1771L<:const is not permitted on named subroutines|perldiag/":const is not permitted on named subroutines"> 1772 1773(F) The C<const> attribute causes an anonymous subroutine to be run and 1774its value captured at the time that it is cloned. Named subroutines are 1775not cloned like this, so the attribute does not make sense on them. 1776 1777=item * 1778 1779L<Hexadecimal float: internal error|perldiag/"Hexadecimal float: internal error"> 1780 1781(F) Something went horribly bad in hexadecimal float handling. 1782 1783=item * 1784 1785L<Hexadecimal float: unsupported long double format|perldiag/"Hexadecimal float: unsupported long double format"> 1786 1787(F) You have configured Perl to use long doubles but 1788the internals of the long double format are unknown, 1789therefore the hexadecimal float output is impossible. 1790 1791=item * 1792 1793L<Illegal suidscript|perldiag/"Illegal suidscript"> 1794 1795(F) The script run under suidperl was somehow illegal. 1796 1797=item * 1798 1799L<In '(?...)', the '(' and '?' must be adjacent in regex; marked by S<<-- HERE> in mE<sol>%sE<sol>|perldiag/"In '(?...)', the '(' and '?' must be adjacent in regex; marked by <-- HERE in m/%s/"> 1800 1801(F) The two-character sequence C<"(?"> in 1802this context in a regular expression pattern should be an 1803indivisible token, with nothing intervening between the C<"("> 1804and the C<"?">, but you separated them. 1805 1806=item * 1807 1808L<In '(*VERB...)', the '(' and '*' must be adjacent in regex; marked by S<<-- HERE> in mE<sol>%sE<sol>|perldiag/"In '(*VERB...)', the '(' and '*' must be adjacent in regex; marked by <-- HERE in m/%s/"> 1809 1810(F) The two-character sequence C<"(*"> in 1811this context in a regular expression pattern should be an 1812indivisible token, with nothing intervening between the C<"("> 1813and the C<"*">, but you separated them. 1814 1815=item * 1816 1817L<Invalid quantifier in {,} in regex; marked by <-- HERE in mE<sol>%sE<sol>|perldiag/"Invalid quantifier in {,} in regex; marked by <-- HERE in m/%s/"> 1818 1819(F) The pattern looks like a {min,max} quantifier, but the min or max could not 1820be parsed as a valid number: either it has leading zeroes, or it represents 1821too big a number to cope with. The S<<-- HERE> shows where in the regular 1822expression the problem was discovered. See L<perlre>. 1823 1824=item * 1825 1826L<'%s' is an unknown bound type in regex|perldiag/"'%s' is an unknown bound type in regex; marked by <-- HERE in m/%s/"> 1827 1828(F) You used C<\b{...}> or C<\B{...}> and the C<...> is not known to 1829Perl. The current valid ones are given in 1830L<perlrebackslash/\b{}, \b, \B{}, \B>. 1831 1832=item * 1833 1834L<Missing or undefined argument to require|perldiag/Missing or undefined argument to require> 1835 1836(F) You tried to call C<require> with no argument or with an undefined 1837value as an argument. C<require> expects either a package name or a 1838file-specification as an argument. See L<perlfunc/require>. 1839 1840Formerly, C<require> with no argument or C<undef> warned about a Null filename. 1841 1842=back 1843 1844=head3 New Warnings 1845 1846=over 4 1847 1848=item * 1849 1850L<\C is deprecated in regex|perldiag/"\C is deprecated in regex; marked by <-- HERE in m/%s/"> 1851 1852(D deprecated) The C<< /\C/ >> character class was deprecated in v5.20, and 1853now emits a warning. It is intended that it will become an error in v5.24. 1854This character class matches a single byte even if it appears within a 1855multi-byte character, breaks encapsulation, and can corrupt UTF-8 1856strings. 1857 1858=item * 1859 1860L<"%s" is more clearly written simply as "%s" in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>|perldiag/"%s" is more clearly written simply as "%s" in regex; marked by <-- HERE in mE<sol>%sE<sol>> 1861 1862(W regexp) (only under C<S<use re 'strict'>> or within C<(?[...])>) 1863 1864You specified a character that has the given plainer way of writing it, 1865and which is also portable to platforms running with different character 1866sets. 1867 1868=item * 1869 1870L<Argument "%s" treated as 0 in increment (++)|perldiag/"Argument "%s" treated 1871as 0 in increment (++)"> 1872 1873(W numeric) The indicated string was fed as an argument to the C<++> operator 1874which expects either a number or a string matching C</^[a-zA-Z]*[0-9]*\z/>. 1875See L<perlop/Auto-increment and Auto-decrement> for details. 1876 1877=item * 1878 1879L<Both or neither range ends should be Unicode in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>|perldiag/"Both or neither range ends should be Unicode in regex; marked by <-- HERE in m/%s/"> 1880 1881(W regexp) (only under C<S<use re 'strict'>> or within C<(?[...])>) 1882 1883In a bracketed character class in a regular expression pattern, you 1884had a range which has exactly one end of it specified using C<\N{}>, and 1885the other end is specified using a non-portable mechanism. Perl treats 1886the range as a Unicode range, that is, all the characters in it are 1887considered to be the Unicode characters, and which may be different code 1888points on some platforms Perl runs on. For example, C<[\N{U+06}-\x08]> 1889is treated as if you had instead said C<[\N{U+06}-\N{U+08}]>, that is it 1890matches the characters whose code points in Unicode are 6, 7, and 8. 1891But that C<\x08> might indicate that you meant something different, so 1892the warning gets raised. 1893 1894=item * 1895 1896L<Can't do %s("%s") on non-UTF-8 locale; resolved to "%s".|perldiag/Can't do %s("%s") on non-UTF-8 locale; resolved to "%s".> 1897 1898(W locale) You are 1) running under "C<use locale>"; 2) the current 1899locale is not a UTF-8 one; 3) you tried to do the designated case-change 1900operation on the specified Unicode character; and 4) the result of this 1901operation would mix Unicode and locale rules, which likely conflict. 1902 1903The warnings category C<locale> is new. 1904 1905=item * 1906 1907L<:const is experimental|perldiag/":const is experimental"> 1908 1909(S experimental::const_attr) The C<const> attribute is experimental. 1910If you want to use the feature, disable the warning with C<no warnings 1911'experimental::const_attr'>, but know that in doing so you are taking 1912the risk that your code may break in a future Perl version. 1913 1914=item * 1915 1916L<gmtime(%f) failed|perldiag/"gmtime(%f) failed"> 1917 1918(W overflow) You called C<gmtime> with a number that it could not handle: 1919too large, too small, or NaN. The returned value is C<undef>. 1920 1921=item * 1922 1923L<Hexadecimal float: exponent overflow|perldiag/"Hexadecimal float: exponent overflow"> 1924 1925(W overflow) The hexadecimal floating point has larger exponent 1926than the floating point supports. 1927 1928=item * 1929 1930L<Hexadecimal float: exponent underflow|perldiag/"Hexadecimal float: exponent underflow"> 1931 1932(W overflow) The hexadecimal floating point has smaller exponent 1933than the floating point supports. 1934 1935=item * 1936 1937L<Hexadecimal float: mantissa overflow|perldiag/"Hexadecimal float: mantissa overflow"> 1938 1939(W overflow) The hexadecimal floating point literal had more bits in 1940the mantissa (the part between the C<0x> and the exponent, also known as 1941the fraction or the significand) than the floating point supports. 1942 1943=item * 1944 1945L<Hexadecimal float: precision loss|perldiag/"Hexadecimal float: precision loss"> 1946 1947(W overflow) The hexadecimal floating point had internally more 1948digits than could be output. This can be caused by unsupported 1949long double formats, or by 64-bit integers not being available 1950(needed to retrieve the digits under some configurations). 1951 1952=item * 1953 1954L<Locale '%s' may not work well.%s|perldiag/Locale '%s' may not work well.%s> 1955 1956(W locale) You are using the named locale, which is a non-UTF-8 one, and 1957which perl has determined is not fully compatible with what it can 1958handle. The second C<%s> gives a reason. 1959 1960The warnings category C<locale> is new. 1961 1962=item * 1963 1964L<localtime(%f) failed|perldiag/"localtime(%f) failed"> 1965 1966(W overflow) You called C<localtime> with a number that it could not handle: 1967too large, too small, or NaN. The returned value is C<undef>. 1968 1969=item * 1970 1971L<Negative repeat count does nothing|perldiag/"Negative repeat count does nothing"> 1972 1973(W numeric) You tried to execute the 1974L<C<x>|perlop/Multiplicative Operators> repetition operator fewer than 0 1975times, which doesn't make sense. 1976 1977=item * 1978 1979L<NO-BREAK SPACE in a charnames alias definition is deprecated|perldiag/"NO-BREAK SPACE in a charnames alias definition is deprecated"> 1980 1981(D deprecated) You defined a character name which contained a no-break 1982space character. Change it to a regular space. Usually these names are 1983defined in the C<:alias> import argument to C<use charnames>, but they 1984could be defined by a translator installed into C<$^H{charnames}>. See 1985L<charnames/CUSTOM ALIASES>. 1986 1987=item * 1988 1989L<Non-finite repeat count does nothing|perldiag/"Non-finite repeat count does nothing"> 1990 1991(W numeric) You tried to execute the 1992L<C<x>|perlop/Multiplicative Operators> repetition operator C<Inf> (or 1993C<-Inf>) or NaN times, which doesn't make sense. 1994 1995=item * 1996 1997L<PerlIO layer ':win32' is experimental|perldiag/"PerlIO layer ':win32' is experimental"> 1998 1999(S experimental::win32_perlio) The C<:win32> PerlIO layer is 2000experimental. If you want to take the risk of using this layer, 2001simply disable this warning: 2002 2003 no warnings "experimental::win32_perlio"; 2004 2005=item * 2006 2007L<Ranges of ASCII printables should be some subset of "0-9", "A-Z", or "a-z" in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>|perldiag/"Ranges of ASCII printables should be some subset of "0-9", "A-Z", or "a-z" in regex; marked by <-- HERE in mE<sol>%sE<sol>"> 2008 2009(W regexp) (only under C<S<use re 'strict'>> or within C<(?[...])>) 2010 2011Stricter rules help to find typos and other errors. Perhaps you didn't 2012even intend a range here, if the C<"-"> was meant to be some other 2013character, or should have been escaped (like C<"\-">). If you did 2014intend a range, the one that was used is not portable between ASCII and 2015EBCDIC platforms, and doesn't have an obvious meaning to a casual 2016reader. 2017 2018 [3-7] # OK; Obvious and portable 2019 [d-g] # OK; Obvious and portable 2020 [A-Y] # OK; Obvious and portable 2021 [A-z] # WRONG; Not portable; not clear what is meant 2022 [a-Z] # WRONG; Not portable; not clear what is meant 2023 [%-.] # WRONG; Not portable; not clear what is meant 2024 [\x41-Z] # WRONG; Not portable; not obvious to non-geek 2025 2026(You can force portability by specifying a Unicode range, which means that 2027the endpoints are specified by 2028L<C<\N{...}>|perlrecharclass/Character Ranges>, but the meaning may 2029still not be obvious.) 2030The stricter rules require that ranges that start or stop with an ASCII 2031character that is not a control have all their endpoints be a literal 2032character, and not some escape sequence (like C<"\x41">), and the ranges 2033must be all digits, or all uppercase letters, or all lowercase letters. 2034 2035=item * 2036 2037L<Ranges of digits should be from the same group in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>|perldiag/"Ranges of digits should be from the same group in regex; marked by <-- HERE in m/%s/"> 2038 2039(W regexp) (only under C<S<use re 'strict'>> or within C<(?[...])>) 2040 2041Stricter rules help to find typos and other errors. You included a 2042range, and at least one of the end points is a decimal digit. Under the 2043stricter rules, when this happens, both end points should be digits in 2044the same group of 10 consecutive digits. 2045 2046=item * 2047 2048L<Redundant argument in %s|perldiag/Redundant argument in %s> 2049 2050(W redundant) You called a function with more arguments than were 2051needed, as indicated by information within other arguments you supplied 2052(I<e.g>. a printf format). Currently only emitted when a printf-type format 2053required fewer arguments than were supplied, but might be used in the 2054future for I<e.g.> L<perlfunc/pack>. 2055 2056The warnings category C<< redundant >> is new. See also 2057L<[GH #13534]|https://github.com/Perl/perl5/issues/13534>. 2058 2059=item * 2060 2061L<Replacement list is longer than search list|perldiag/Replacement list is longer than search list> 2062 2063This is not a new diagnostic, but in earlier releases was accidentally 2064not displayed if the transliteration contained wide characters. This is 2065now fixed, so that you may see this diagnostic in places where you 2066previously didn't (but should have). 2067 2068=item * 2069 2070L<Use of \b{} for non-UTF-8 locale is wrong. Assuming a UTF-8 locale|perldiag/"Use of \b{} for non-UTF-8 locale is wrong. Assuming a UTF-8 locale"> 2071 2072(W locale) You are matching a regular expression using locale rules, 2073and a Unicode boundary is being matched, but the locale is not a Unicode 2074one. This doesn't make sense. Perl will continue, assuming a Unicode 2075(UTF-8) locale, but the results could well be wrong except if the locale 2076happens to be ISO-8859-1 (Latin1) where this message is spurious and can 2077be ignored. 2078 2079The warnings category C<locale> is new. 2080 2081=item * 2082 2083L<< Using E<sol>u for '%s' instead of E<sol>%s in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>|perldiag/"Using E<sol>u for '%s' instead of E<sol>%s in regex; marked by <-- HERE in mE<sol>%sE<sol>" >> 2084 2085(W regexp) You used a Unicode boundary (C<\b{...}> or C<\B{...}>) in a 2086portion of a regular expression where the character set modifiers C</a> 2087or C</aa> are in effect. These two modifiers indicate an ASCII 2088interpretation, and this doesn't make sense for a Unicode definition. 2089The generated regular expression will compile so that the boundary uses 2090all of Unicode. No other portion of the regular expression is affected. 2091 2092=item * 2093 2094L<The bitwise feature is experimental|perldiag/"The bitwise feature is experimental"> 2095 2096(S experimental::bitwise) This warning is emitted if you use bitwise 2097operators (C<& | ^ ~ &. |. ^. ~.>) with the "bitwise" feature enabled. 2098Simply suppress the warning if you want to use the feature, but know 2099that in doing so you are taking the risk of using an experimental 2100feature which may change or be removed in a future Perl version: 2101 2102 no warnings "experimental::bitwise"; 2103 use feature "bitwise"; 2104 $x |.= $y; 2105 2106=item * 2107 2108L<Unescaped left brace in regex is deprecated, passed through in regex; marked by <-- HERE in mE<sol>%sE<sol>|perldiag/"Unescaped left brace in regex is deprecated, passed through in regex; marked by <-- HERE in m/%s/"> 2109 2110(D deprecated, regexp) You used a literal C<"{"> character in a regular 2111expression pattern. You should change to use C<"\{"> instead, because a future 2112version of Perl (tentatively v5.26) will consider this to be a syntax error. If 2113the pattern delimiters are also braces, any matching right brace 2114(C<"}">) should also be escaped to avoid confusing the parser, for 2115example, 2116 2117 qr{abc\{def\}ghi} 2118 2119=item * 2120 2121L<Use of literal non-graphic characters in variable names is deprecated|perldiag/"Use of literal non-graphic characters in variable names is deprecated"> 2122 2123(D deprecated) Using literal non-graphic (including control) 2124characters in the source to refer to the I<^FOO> variables, like C<$^X> and 2125C<${^GLOBAL_PHASE}> is now deprecated. 2126 2127=item * 2128 2129L<Useless use of attribute "const"|perldiag/Useless use of attribute "const"> 2130 2131(W misc) The C<const> attribute has no effect except 2132on anonymous closure prototypes. You applied it to 2133a subroutine via L<attributes.pm|attributes>. This is only useful 2134inside an attribute handler for an anonymous subroutine. 2135 2136=item * 2137 2138L<Useless use of E<sol>d modifier in transliteration operator|perldiag/"Useless use of /d modifier in transliteration operator"> 2139 2140This is not a new diagnostic, but in earlier releases was accidentally 2141not displayed if the transliteration contained wide characters. This is 2142now fixed, so that you may see this diagnostic in places where you 2143previously didn't (but should have). 2144 2145=item * 2146 2147L<E<quot>use re 'strict'E<quot> is experimental|perldiag/"use re 'strict'" is experimental> 2148 2149(S experimental::re_strict) The things that are different when a regular 2150expression pattern is compiled under C<'strict'> are subject to change 2151in future Perl releases in incompatible ways; there are also proposals 2152to change how to enable strict checking instead of using this subpragma. 2153This means that a pattern that compiles today may not in a future Perl 2154release. This warning is to alert you to that risk. 2155 2156=item * 2157 2158L<Warning: unable to close filehandle properly: %s|perldiag/"Warning: unable to close filehandle properly: %s"> 2159 2160L<Warning: unable to close filehandle %s properly: %s|perldiag/"Warning: unable to close filehandle %s properly: %s"> 2161 2162(S io) Previously, perl silently ignored any errors when doing an implicit 2163close of a filehandle, I<i.e.> where the reference count of the filehandle 2164reached zero and the user's code hadn't already called C<close()>; I<e.g.> 2165 2166 { 2167 open my $fh, '>', $file or die "open: '$file': $!\n"; 2168 print $fh, $data or die; 2169 } # implicit close here 2170 2171In a situation such as disk full, due to buffering, the error may only be 2172detected during the final close, so not checking the result of the close is 2173dangerous. 2174 2175So perl now warns in such situations. 2176 2177=item * 2178 2179L<Wide character (U+%X) in %s|perldiag/"Wide character (U+%X) in %s"> 2180 2181(W locale) While in a single-byte locale (I<i.e.>, a non-UTF-8 2182one), a multi-byte character was encountered. Perl considers this 2183character to be the specified Unicode code point. Combining non-UTF-8 2184locales and Unicode is dangerous. Almost certainly some characters 2185will have two different representations. For example, in the ISO 8859-7 2186(Greek) locale, the code point 0xC3 represents a Capital Gamma. But so 2187also does 0x393. This will make string comparisons unreliable. 2188 2189You likely need to figure out how this multi-byte character got mixed up 2190with your single-byte locale (or perhaps you thought you had a UTF-8 2191locale, but Perl disagrees). 2192 2193The warnings category C<locale> is new. 2194 2195=back 2196 2197=head2 Changes to Existing Diagnostics 2198 2199=over 4 2200 2201=item * 2202 2203<> should be quotes 2204 2205This warning has been changed to 2206L<< <> at require-statement should be quotes|perldiag/"<> at require-statement should be quotes" >> 2207to make the issue more identifiable. 2208 2209=item * 2210 2211L<Argument "%s" isn't numeric%s|perldiag/"Argument "%s" isn't numeric%s"> 2212 2213The L<perldiag> entry for this warning has added this clarifying note: 2214 2215 Note that for the Inf and NaN (infinity and not-a-number) the 2216 definition of "numeric" is somewhat unusual: the strings themselves 2217 (like "Inf") are considered numeric, and anything following them is 2218 considered non-numeric. 2219 2220=item * 2221 2222L<Global symbol "%s" requires explicit package name|perldiag/"Global symbol "%s" requires explicit package name (did you forget to declare "my %s"?)"> 2223 2224This message has had '(did you forget to declare "my %s"?)' appended to it, to 2225make it more helpful to new Perl programmers. 2226L<[GH #13732]|https://github.com/Perl/perl5/issues/13732> 2227 2228=item * 2229 2230'"my" variable &foo::bar can't be in a package' has been reworded to say 2231'subroutine' instead of 'variable'. 2232 2233=item * 2234 2235L<<< \N{} in character class restricted to one character in regex; marked by 2236S<< <-- HERE >> in mE<sol>%sE<sol>|perldiag/"\N{} in inverted character 2237class or as a range end-point is restricted to one character in regex; 2238marked by <-- HERE in m/%s/" >>> 2239 2240This message has had I<character class> changed to I<inverted character 2241class or as a range end-point is> to reflect improvements in 2242C<qr/[\N{named sequence}]/> (see under L</Selected Bug Fixes>). 2243 2244=item * 2245 2246L<panic: frexp|perldiag/"panic: frexp: %f"> 2247 2248This message has had ': C<%f>' appended to it, to show what the offending 2249floating point number is. 2250 2251=item * 2252 2253I<Possible precedence problem on bitwise %c operator> reworded as 2254L<Possible precedence problem on bitwise %s operator|perldiag/"Possible precedence problem on bitwise %s operator">. 2255 2256=item * 2257 2258L<Unsuccessful %s on filename containing newline|perldiag/"Unsuccessful %s on filename containing newline"> 2259 2260This warning is now only produced when the newline is at the end of 2261the filename. 2262 2263=item * 2264 2265"Variable C<%s> will not stay shared" has been changed to say "Subroutine" 2266when it is actually a lexical sub that will not stay shared. 2267 2268=item * 2269 2270L<Variable length lookbehind not implemented in regex mE<sol>%sE<sol>|perldiag/"Variable length lookbehind not implemented in regex m/%s/"> 2271 2272The L<perldiag> entry for this warning has had information about Unicode 2273behavior added. 2274 2275=back 2276 2277=head2 Diagnostic Removals 2278 2279=over 2280 2281=item * 2282 2283"Ambiguous use of -foo resolved as -&foo()" 2284 2285There is actually no ambiguity here, and this impedes the use of negated 2286constants; I<e.g.>, C<-Inf>. 2287 2288=item * 2289 2290"Constant is not a FOO reference" 2291 2292Compile-time checking of constant dereferencing (I<e.g.>, C<< my_constant->() >>) 2293has been removed, since it was not taking overloading into account. 2294L<[GH #9891]|https://github.com/Perl/perl5/issues/9891> 2295L<[GH #14044]|https://github.com/Perl/perl5/issues/14044> 2296 2297=back 2298 2299=head1 Utility Changes 2300 2301=head2 F<find2perl>, F<s2p> and F<a2p> removal 2302 2303=over 4 2304 2305=item * 2306 2307The F<x2p/> directory has been removed from the Perl core. 2308 2309This removes find2perl, s2p and a2p. They have all been released to CPAN as 2310separate distributions (C<App::find2perl>, C<App::s2p>, C<App::a2p>). 2311 2312=back 2313 2314=head2 L<h2ph> 2315 2316=over 4 2317 2318=item * 2319 2320F<h2ph> now handles hexadecimal constants in the compiler's predefined 2321macro definitions, as visible in C<$Config{cppsymbols}>. 2322L<[GH #14491]|https://github.com/Perl/perl5/issues/14491>. 2323 2324=back 2325 2326=head2 L<encguess> 2327 2328=over 4 2329 2330=item * 2331 2332No longer depends on non-core modules. 2333 2334=back 2335 2336=head1 Configuration and Compilation 2337 2338=over 4 2339 2340=item * 2341 2342F<Configure> now checks for C<lrintl()>, C<lroundl()>, C<llrintl()>, and 2343C<llroundl()>. 2344 2345=item * 2346 2347F<Configure> with C<-Dmksymlinks> should now be faster. 2348L<[GH #13890]|https://github.com/Perl/perl5/issues/13890>. 2349 2350=item * 2351 2352The C<pthreads> and C<cl> libraries will be linked by default if present. 2353This allows XS modules that require threading to work on non-threaded 2354perls. Note that you must still pass C<-Dusethreads> if you want a 2355threaded perl. 2356 2357=item * 2358 2359To get more precision and range for floating point numbers one can now 2360use the GCC quadmath library which implements the quadruple precision 2361floating point numbers on x86 and IA-64 platforms. See F<INSTALL> for 2362details. 2363 2364=item * 2365 2366MurmurHash64A and MurmurHash64B can now be configured as the internal hash 2367function. 2368 2369=item * 2370 2371C<make test.valgrind> now supports parallel testing. 2372 2373For example: 2374 2375 TEST_JOBS=9 make test.valgrind 2376 2377See L<perlhacktips/valgrind> for more information. 2378 2379L<[GH #13658]|https://github.com/Perl/perl5/issues/13658> 2380 2381=item * 2382 2383The MAD (Misc Attribute Decoration) build option has been removed 2384 2385This was an unmaintained attempt at preserving 2386the Perl parse tree more faithfully so that automatic conversion of 2387Perl 5 to Perl 6 would have been easier. 2388 2389This build-time configuration option had been unmaintained for years, 2390and had probably seriously diverged on both Perl 5 and Perl 6 sides. 2391 2392=item * 2393 2394A new compilation flag, C<< -DPERL_OP_PARENT >> is available. For details, 2395see the discussion below at L<< /Internal Changes >>. 2396 2397=item * 2398 2399Pathtools no longer tries to load XS on miniperl. This speeds up building perl 2400slightly. 2401 2402=back 2403 2404=head1 Testing 2405 2406=over 4 2407 2408=item * 2409 2410F<t/porting/re_context.t> has been added to test that L<utf8> and its 2411dependencies only use the subset of the C<$1..$n> capture vars that 2412C<Perl_save_re_context()> is hard-coded to localize, because that function 2413has no efficient way of determining at runtime what vars to localize. 2414 2415=item * 2416 2417Tests for performance issues have been added in the file F<t/perf/taint.t>. 2418 2419=item * 2420 2421Some regular expression tests are written in such a way that they will 2422run very slowly if certain optimizations break. These tests have been 2423moved into new files, F<< t/re/speed.t >> and F<< t/re/speed_thr.t >>, 2424and are run with a C<< watchdog() >>. 2425 2426=item * 2427 2428C<< test.pl >> now allows C<< plan skip_all => $reason >>, to make it 2429more compatible with C<< Test::More >>. 2430 2431=item * 2432 2433A new test script, F<op/infnan.t>, has been added to test if infinity and NaN are 2434working correctly. See L</Infinity and NaN (not-a-number) handling improved>. 2435 2436=back 2437 2438=head1 Platform Support 2439 2440=head2 Regained Platforms 2441 2442=over 4 2443 2444=item IRIX and Tru64 platforms are working again. 2445 2446Some C<make test> failures remain: 2447L<[GH #14557]|https://github.com/Perl/perl5/issues/14557> 2448and L<[GH #14727]|https://github.com/Perl/perl5/issues/14727> 2449for IRIX; L<[GH #14629]|https://github.com/Perl/perl5/issues/14629>, 2450L<[cpan #99605]|https://rt.cpan.org/Public/Bug/Display.html?id=99605>, and 2451L<[cpan #104836]|https://rt.cpan.org/Ticket/Display.html?id=104836> for Tru64. 2452 2453=item z/OS running EBCDIC Code Page 1047 2454 2455Core perl now works on this EBCDIC platform. Earlier perls also worked, but, 2456even though support wasn't officially withdrawn, recent perls would not compile 2457and run well. Perl 5.20 would work, but had many bugs which have now been 2458fixed. Many CPAN modules that ship with Perl still fail tests, including 2459C<Pod::Simple>. However the version of C<Pod::Simple> currently on CPAN should work; 2460it was fixed too late to include in Perl 5.22. Work is under way to fix many 2461of the still-broken CPAN modules, which likely will be installed on CPAN when 2462completed, so that you may not have to wait until Perl 5.24 to get a working 2463version. 2464 2465=back 2466 2467=head2 Discontinued Platforms 2468 2469=over 4 2470 2471=item NeXTSTEP/OPENSTEP 2472 2473NeXTSTEP was a proprietary operating system bundled with NeXT's 2474workstations in the early to mid 90s; OPENSTEP was an API specification 2475that provided a NeXTSTEP-like environment on a non-NeXTSTEP system. Both 2476are now long dead, so support for building Perl on them has been removed. 2477 2478=back 2479 2480=head2 Platform-Specific Notes 2481 2482=over 4 2483 2484=item EBCDIC 2485 2486Special handling is required of the perl interpreter on EBCDIC platforms 2487to get C<qr/[i-j]/> to match only C<"i"> and C<"j">, since there are 7 2488characters between the 2489code points for C<"i"> and C<"j">. This special handling had only been 2490invoked when both ends of the range are literals. Now it is also 2491invoked if any of the C<\N{...}> forms for specifying a character by 2492name or Unicode code point is used instead of a literal. See 2493L<perlrecharclass/Character Ranges>. 2494 2495=item HP-UX 2496 2497The archname now distinguishes use64bitint from use64bitall. 2498 2499=item Android 2500 2501Build support has been improved for cross-compiling in general and for 2502Android in particular. 2503 2504=item VMS 2505 2506=over 4 2507 2508=item * 2509 2510When spawning a subprocess without waiting, the return value is now 2511the correct PID. 2512 2513=item * 2514 2515Fix a prototype so linking doesn't fail under the VMS C++ compiler. 2516 2517=item * 2518 2519C<finite>, C<finitel>, and C<isfinite> detection has been added to 2520C<configure.com>, environment handling has had some minor changes, and 2521a fix for legacy feature checking status. 2522 2523=back 2524 2525=item Win32 2526 2527=over 4 2528 2529=item * 2530 2531F<miniperl.exe> is now built with C<-fno-strict-aliasing>, allowing 64-bit 2532builds to complete on GCC 4.8. 2533L<[GH #14556]|https://github.com/Perl/perl5/issues/14556> 2534 2535=item * 2536 2537C<nmake minitest> now works on Win32. Due to dependency issues you 2538need to build C<nmake test-prep> first, and a small number of the 2539tests fail. 2540L<[GH #14318]|https://github.com/Perl/perl5/issues/14318> 2541 2542=item * 2543 2544Perl can now be built in C++ mode on Windows by setting the makefile macro 2545C<USE_CPLUSPLUS> to the value "define". 2546 2547=item * 2548 2549The list form of piped open has been implemented for Win32. Note: unlike 2550C<system LIST> this does not fall back to the shell. 2551L<[GH #13574]|https://github.com/Perl/perl5/issues/13574> 2552 2553=item * 2554 2555New C<DebugSymbols> and C<DebugFull> configuration options added to 2556Windows makefiles. 2557 2558=item * 2559 2560Previously, compiling XS modules (including CPAN ones) using Visual C++ for 2561Win64 resulted in around a dozen warnings per file from F<hv_func.h>. These 2562warnings have been silenced. 2563 2564=item * 2565 2566Support for building without PerlIO has been removed from the Windows 2567makefiles. Non-PerlIO builds were all but deprecated in Perl 5.18.0 and are 2568already not supported by F<Configure> on POSIX systems. 2569 2570=item * 2571 2572Between 2 and 6 milliseconds and seven I/O calls have been saved per attempt 2573to open a perl module for each path in C<@INC>. 2574 2575=item * 2576 2577Intel C builds are now always built with C99 mode on. 2578 2579=item * 2580 2581C<%I64d> is now being used instead of C<%lld> for MinGW. 2582 2583=item * 2584 2585In the experimental C<:win32> layer, a crash in C<open> was fixed. Also 2586opening F</dev/null> (which works under Win32 Perl's default C<:unix> 2587layer) was implemented for C<:win32>. 2588L<[GH #13968]|https://github.com/Perl/perl5/issues/13968> 2589 2590=item * 2591 2592A new makefile option, C<USE_LONG_DOUBLE>, has been added to the Windows 2593dmake makefile for gcc builds only. Set this to "define" if you want perl to 2594use long doubles to give more accuracy and range for floating point numbers. 2595 2596=back 2597 2598=item OpenBSD 2599 2600On OpenBSD, Perl will now default to using the system C<malloc> due to the 2601security features it provides. Perl's own malloc wrapper has been in use 2602since v5.14 due to performance reasons, but the OpenBSD project believes 2603the tradeoff is worth it and would prefer that users who need the speed 2604specifically ask for it. 2605 2606L<[GH #13888]|https://github.com/Perl/perl5/issues/13888>. 2607 2608=item Solaris 2609 2610=over 4 2611 2612=item * 2613 2614We now look for the Sun Studio compiler in both F</opt/solstudio*> and 2615F</opt/solarisstudio*>. 2616 2617=item * 2618 2619Builds on Solaris 10 with C<-Dusedtrace> would fail early since make 2620didn't follow implied dependencies to build C<perldtrace.h>. Added an 2621explicit dependency to C<depend>. 2622L<[GH #13334]|https://github.com/Perl/perl5/issues/13334> 2623 2624=item * 2625 2626C99 options have been cleaned up; hints look for C<solstudio> 2627as well as C<SUNWspro>; and support for native C<setenv> has been added. 2628 2629=back 2630 2631=back 2632 2633=head1 Internal Changes 2634 2635=over 4 2636 2637=item * 2638 2639Experimental support has been added to allow ops in the optree to locate 2640their parent, if any. This is enabled by the non-default build option 2641C<-DPERL_OP_PARENT>. It is envisaged that this will eventually become 2642enabled by default, so XS code which directly accesses the C<op_sibling> 2643field of ops should be updated to be future-proofed. 2644 2645On C<PERL_OP_PARENT> builds, the C<op_sibling> field has been renamed 2646C<op_sibparent> and a new flag, C<op_moresib>, added. On the last op in a 2647sibling chain, C<op_moresib> is false and C<op_sibparent> points to the 2648parent (if any) rather than being C<NULL>. 2649 2650To make existing code work transparently whether using C<PERL_OP_PARENT> 2651or not, a number of new macros and functions have been added that should 2652be used, rather than directly manipulating C<op_sibling>. 2653 2654For the case of just reading C<op_sibling> to determine the next sibling, 2655two new macros have been added. A simple scan through a sibling chain 2656like this: 2657 2658 for (; kid->op_sibling; kid = kid->op_sibling) { ... } 2659 2660should now be written as: 2661 2662 for (; OpHAS_SIBLING(kid); kid = OpSIBLING(kid)) { ... } 2663 2664For altering optrees, a general-purpose function C<op_sibling_splice()> 2665has been added, which allows for manipulation of a chain of sibling ops. 2666By analogy with the Perl function C<splice()>, it allows you to cut out 2667zero or more ops from a sibling chain and replace them with zero or more 2668new ops. It transparently handles all the updating of sibling, parent, 2669op_last pointers etc. 2670 2671If you need to manipulate ops at a lower level, then three new macros, 2672C<OpMORESIB_set>, C<OpLASTSIB_set> and C<OpMAYBESIB_set> are intended to 2673be a low-level portable way to set C<op_sibling> / C<op_sibparent> while 2674also updating C<op_moresib>. The first sets the sibling pointer to a new 2675sibling, the second makes the op the last sibling, and the third 2676conditionally does the first or second action. Note that unlike 2677C<op_sibling_splice()> these macros won't maintain consistency in the 2678parent at the same time (I<e.g.> by updating C<op_first> and C<op_last> where 2679appropriate). 2680 2681A C-level C<Perl_op_parent()> function and a Perl-level C<B::OP::parent()> 2682method have been added. The C function only exists under 2683C<PERL_OP_PARENT> builds (using it is build-time error on vanilla 2684perls). C<B::OP::parent()> exists always, but on a vanilla build it 2685always returns C<NULL>. Under C<PERL_OP_PARENT>, they return the parent 2686of the current op, if any. The variable C<$B::OP::does_parent> allows you 2687to determine whether C<B> supports retrieving an op's parent. 2688 2689C<PERL_OP_PARENT> was introduced in 5.21.2, but the interface was 2690changed considerably in 5.21.11. If you updated your code before the 26915.21.11 changes, it may require further revision. The main changes after 26925.21.2 were: 2693 2694=over 4 2695 2696=item * 2697 2698The C<OP_SIBLING> and C<OP_HAS_SIBLING> macros have been renamed 2699C<OpSIBLING> and C<OpHAS_SIBLING> for consistency with other 2700op-manipulating macros. 2701 2702=item * 2703 2704The C<op_lastsib> field has been renamed C<op_moresib>, and its meaning 2705inverted. 2706 2707=item * 2708 2709The macro C<OpSIBLING_set> has been removed, and has been superseded by 2710C<OpMORESIB_set> I<et al>. 2711 2712=item * 2713 2714The C<op_sibling_splice()> function now accepts a null C<parent> argument 2715where the splicing doesn't affect the first or last ops in the sibling 2716chain 2717 2718=back 2719 2720=item * 2721 2722Macros have been created to allow XS code to better manipulate the POSIX locale 2723category C<LC_NUMERIC>. See L<perlapi/Locale-related functions and macros>. 2724 2725=item * 2726 2727The previous C<atoi> I<et al> replacement function, C<grok_atou>, has now been 2728superseded by C<grok_atoUV>. See L<perlclib> for details. 2729 2730=item * 2731 2732A new function, C<Perl_sv_get_backrefs()>, has been added which allows you 2733retrieve the weak references, if any, which point at an SV. 2734 2735=item * 2736 2737The C<screaminstr()> function has been removed. Although marked as 2738public API, it was undocumented and had no usage in CPAN modules. Calling 2739it has been fatal since 5.17.0. 2740 2741=item * 2742 2743The C<newDEFSVOP()>, C<block_start()>, C<block_end()> and C<intro_my()> 2744functions have been added to the API. 2745 2746=item * 2747 2748The internal C<convert> function in F<op.c> has been renamed 2749C<op_convert_list> and added to the API. 2750 2751=item * 2752 2753The C<sv_magic()> function no longer forbids "ext" magic on read-only 2754values. After all, perl can't know whether the custom magic will modify 2755the SV or not. 2756L<[GH #14202]|https://github.com/Perl/perl5/issues/14202>. 2757 2758=item * 2759 2760Accessing L<perlapi/CvPADLIST> on an XSUB is now forbidden. 2761 2762The C<CvPADLIST> field has been reused for a different internal purpose 2763for XSUBs. So in particular, you can no longer rely on it being NULL as a 2764test of whether a CV is an XSUB. Use C<CvISXSUB()> instead. 2765 2766=item * 2767 2768SVs of type C<SVt_NV> are now sometimes bodiless when the build 2769configuration and platform allow it: specifically, when C<< sizeof(NV) <= 2770sizeof(IV) >>. "Bodiless" means that the NV value is stored directly in 2771the head of an SV, without requiring a separate body to be allocated. This 2772trick has already been used for IVs since 5.9.2 (though in the case of 2773IVs, it is always used, regardless of platform and build configuration). 2774 2775=item * 2776 2777The C<$DB::single>, C<$DB::signal> and C<$DB::trace> variables now have set- and 2778get-magic that stores their values as IVs, and those IVs are used when 2779testing their values in C<pp_dbstate()>. This prevents perl from 2780recursing infinitely if an overloaded object is assigned to any of those 2781variables. 2782L<[GH #14013]|https://github.com/Perl/perl5/issues/14013>. 2783 2784=item * 2785 2786C<Perl_tmps_grow()>, which is marked as public API but is undocumented, has 2787been removed from the public API. This change does not affect XS code that 2788uses the C<EXTEND_MORTAL> macro to pre-extend the mortal stack. 2789 2790=item * 2791 2792Perl's internals no longer sets or uses the C<SVs_PADMY> flag. 2793C<SvPADMY()> now returns a true value for anything not marked C<PADTMP> 2794and C<SVs_PADMY> is now defined as 0. 2795 2796=item * 2797 2798The macros C<SETsv> and C<SETsvUN> have been removed. They were no longer used 2799in the core since commit 6f1401dc2a five years ago, and have not been 2800found present on CPAN. 2801 2802=item * 2803 2804The C<< SvFAKE >> bit (unused on HVs) got informally reserved by 2805David Mitchell for future work on vtables. 2806 2807=item * 2808 2809The C<sv_catpvn_flags()> function accepts C<SV_CATBYTES> and C<SV_CATUTF8> 2810flags, which specify whether the appended string is bytes or UTF-8, 2811respectively. (These flags have in fact been present since 5.16.0, but 2812were formerly not regarded as part of the API.) 2813 2814=item * 2815 2816A new opcode class, C<< METHOP >>, has been introduced. It holds 2817information used at runtime to improve the performance 2818of class/object method calls. 2819 2820C<< OP_METHOD >> and C<< OP_METHOD_NAMED >> have changed from being 2821C<< UNOP/SVOP >> to being C<< METHOP >>. 2822 2823=item * 2824 2825C<cv_name()> is a new API function that can be passed a CV or GV. It 2826returns an SV containing the name of the subroutine, for use in 2827diagnostics. 2828 2829L<[GH #12767]|https://github.com/Perl/perl5/issues/12767> 2830L<[GH #13392]|https://github.com/Perl/perl5/issues/13392> 2831 2832=item * 2833 2834C<cv_set_call_checker_flags()> is a new API function that works like 2835C<cv_set_call_checker()>, except that it allows the caller to specify 2836whether the call checker requires a full GV for reporting the subroutine's 2837name, or whether it could be passed a CV instead. Whatever value is 2838passed will be acceptable to C<cv_name()>. C<cv_set_call_checker()> 2839guarantees there will be a GV, but it may have to create one on the fly, 2840which is inefficient. 2841L<[GH #12767]|https://github.com/Perl/perl5/issues/12767> 2842 2843=item * 2844 2845C<CvGV> (which is not part of the API) is now a more complex macro, which may 2846call a function and reify a GV. For those cases where it has been used as a 2847boolean, C<CvHASGV> has been added, which will return true for CVs that 2848notionally have GVs, but without reifying the GV. C<CvGV> also returns a GV 2849now for lexical subs. 2850L<[GH #13392]|https://github.com/Perl/perl5/issues/13392> 2851 2852=item * 2853 2854The L<perlapi/sync_locale> function has been added to the public API. 2855Changing the program's locale should be avoided by XS code. Nevertheless, 2856certain non-Perl libraries called from XS need to do so, such as C<Gtk>. 2857When this happens, Perl needs to be told that the locale has 2858changed. Use this function to do so, before returning to Perl. 2859 2860=item * 2861 2862The defines and labels for the flags in the C<op_private> field of OPs are now 2863auto-generated from data in F<regen/op_private>. The noticeable effect of this 2864is that some of the flag output of C<Concise> might differ slightly, and the 2865flag output of S<C<perl -Dx>> may differ considerably (they both use the same set 2866of labels now). Also, debugging builds now have a new assertion in 2867C<op_free()> to ensure that the op doesn't have any unrecognized flags set in 2868C<op_private>. 2869 2870=item * 2871 2872The deprecated variable C<PL_sv_objcount> has been removed. 2873 2874=item * 2875 2876Perl now tries to keep the locale category C<LC_NUMERIC> set to "C" 2877except around operations that need it to be set to the program's 2878underlying locale. This protects the many XS modules that cannot cope 2879with the decimal radix character not being a dot. Prior to this 2880release, Perl initialized this category to "C", but a call to 2881C<POSIX::setlocale()> would change it. Now such a call will change the 2882underlying locale of the C<LC_NUMERIC> category for the program, but the 2883locale exposed to XS code will remain "C". There are new macros 2884to manipulate the LC_NUMERIC locale, including 2885C<STORE_LC_NUMERIC_SET_TO_NEEDED> and 2886C<STORE_LC_NUMERIC_FORCE_TO_UNDERLYING>. 2887See L<perlapi/Locale-related functions and macros>. 2888 2889=item * 2890 2891A new macro L<C<isUTF8_CHAR>|perlapi/isUTF8_CHAR> has been written which 2892efficiently determines if the string given by its parameters begins 2893with a well-formed UTF-8 encoded character. 2894 2895=item * 2896 2897The following private API functions had their context parameter removed: 2898C<Perl_cast_ulong>, C<Perl_cast_i32>, C<Perl_cast_iv>, C<Perl_cast_uv>, 2899C<Perl_cv_const_sv>, C<Perl_mg_find>, C<Perl_mg_findext>, C<Perl_mg_magical>, 2900C<Perl_mini_mktime>, C<Perl_my_dirfd>, C<Perl_sv_backoff>, C<Perl_utf8_hop>. 2901 2902Note that the prefix-less versions of those functions that are part of the 2903public API, such as C<cast_i32()>, remain unaffected. 2904 2905=item * 2906 2907The C<PADNAME> and C<PADNAMELIST> types are now separate types, and no 2908longer simply aliases for SV and AV. 2909L<[GH #14250]|https://github.com/Perl/perl5/issues/14250>. 2910 2911=item * 2912 2913Pad names are now always UTF-8. The C<PadnameUTF8> macro always returns 2914true. Previously, this was effectively the case already, but any support 2915for two different internal representations of pad names has now been 2916removed. 2917 2918=item * 2919 2920A new op class, C<UNOP_AUX>, has been added. This is a subclass of 2921C<UNOP> with an C<op_aux> field added, which points to an array of unions 2922of UV, SV* etc. It is intended for where an op needs to store more data 2923than a simple C<op_sv> or whatever. Currently the only op of this type is 2924C<OP_MULTIDEREF> (see next item). 2925 2926=item * 2927 2928A new op has been added, C<OP_MULTIDEREF>, which performs one or more 2929nested array and hash lookups where the key is a constant or simple 2930variable. For example the expression C<$a[0]{$k}[$i]>, which previously 2931involved ten C<rv2Xv>, C<Xelem>, C<gvsv> and C<const> ops is now performed 2932by a single C<multideref> op. It can also handle C<local>, C<exists> and 2933C<delete>. A non-simple index expression, such as C<[$i+1]> is still done 2934using C<aelem>/C<helem>, and single-level array lookup with a small constant 2935index is still done using C<aelemfast>. 2936 2937=back 2938 2939=head1 Selected Bug Fixes 2940 2941=over 4 2942 2943=item * 2944 2945C<close> now sets C<$!> 2946 2947When an I/O error occurs, the fact that there has been an error is recorded 2948in the handle. C<close> returns false for such a handle. Previously, the 2949value of C<$!> would be untouched by C<close>, so the common convention of 2950writing S<C<close $fh or die $!>> did not work reliably. Now the handle 2951records the value of C<$!>, too, and C<close> restores it. 2952 2953=item * 2954 2955C<no re> now can turn off everything that C<use re> enables 2956 2957Previously, running C<no re> would turn off only a few things. Now it 2958can turn off all the enabled things. For example, the only way to 2959stop debugging, once enabled, was to exit the enclosing block; that is 2960now fixed. 2961 2962=item * 2963 2964C<pack("D", $x)> and C<pack("F", $x)> now zero the padding on x86 long 2965double builds. Under some build options on GCC 4.8 and later, they used 2966to either overwrite the zero-initialized padding, or bypass the 2967initialized buffer entirely. This caused F<op/pack.t> to fail. 2968L<[GH #14554]|https://github.com/Perl/perl5/issues/14554> 2969 2970=item * 2971 2972Extending an array cloned from a parent thread could result in "Modification of 2973a read-only value attempted" errors when attempting to modify the new elements. 2974L<[GH #14605]|https://github.com/Perl/perl5/issues/14605> 2975 2976=item * 2977 2978An assertion failure and subsequent crash with C<< *x=<y> >> has been fixed. 2979L<[GH #14493]|https://github.com/Perl/perl5/issues/14493> 2980 2981=item * 2982 2983A possible crashing/looping bug related to compiling lexical subs has been 2984fixed. 2985L<[GH #14596]|https://github.com/Perl/perl5/issues/14596> 2986 2987=item * 2988 2989UTF-8 now works correctly in function names, in unquoted HERE-document 2990terminators, and in variable names used as array indexes. 2991L<[GH #14601]|https://github.com/Perl/perl5/issues/14601> 2992 2993=item * 2994 2995Repeated global pattern matches in scalar context on large tainted strings were 2996exponentially slow depending on the current match position in the string. 2997L<[GH #14238]|https://github.com/Perl/perl5/issues/14238> 2998 2999=item * 3000 3001Various crashes due to the parser getting confused by syntax errors have been 3002fixed. 3003L<[GH #14496]|https://github.com/Perl/perl5/issues/14496> 3004L<[GH #14497]|https://github.com/Perl/perl5/issues/14497> 3005L<[GH #14548]|https://github.com/Perl/perl5/issues/14548> 3006L<[GH #14564]|https://github.com/Perl/perl5/issues/14564> 3007 3008=item * 3009 3010C<split> in the scope of lexical C<$_> has been fixed not to fail assertions. 3011L<[GH #14483]|https://github.com/Perl/perl5/issues/14483> 3012 3013=item * 3014 3015C<my $x : attr> syntax inside various list operators no longer fails 3016assertions. 3017L<[GH #14500]|https://github.com/Perl/perl5/issues/14500> 3018 3019=item * 3020 3021An C<@> sign in quotes followed by a non-ASCII digit (which is not a valid 3022identifier) would cause the parser to crash, instead of simply trying the 3023C<@> as literal. This has been fixed. 3024L<[GH #14553]|https://github.com/Perl/perl5/issues/14553> 3025 3026=item * 3027 3028C<*bar::=*foo::=*glob_with_hash> has been crashing since Perl 5.14, but no 3029longer does. 3030L<[GH #14512]|https://github.com/Perl/perl5/issues/14512> 3031 3032=item * 3033 3034C<foreach> in scalar context was not pushing an item on to the stack, resulting 3035in bugs. (S<C<print 4, scalar do { foreach(@x){} } + 1>> would print 5.) 3036It has been fixed to return C<undef>. 3037L<[GH #14569]|https://github.com/Perl/perl5/issues/14569> 3038 3039=item * 3040 3041Several cases of data used to store environment variable contents in core C 3042code being potentially overwritten before being used have been fixed. 3043L<[GH #14476]|https://github.com/Perl/perl5/issues/14476> 3044 3045=item * 3046 3047Some patterns starting with C</.*..../> matched against long strings have 3048been slow since v5.8, and some of the form C</.*..../i> have been slow 3049since v5.18. They are now all fast again. 3050L<[GH #14475]|https://github.com/Perl/perl5/issues/14475>. 3051 3052=item * 3053 3054The original visible value of C<$/> is now preserved when it is set to 3055an invalid value. Previously if you set C<$/> to a reference to an 3056array, for example, perl would produce a runtime error and not set 3057C<PL_rs>, but Perl code that checked C<$/> would see the array 3058reference. 3059L<[GH #14245]|https://github.com/Perl/perl5/issues/14245>. 3060 3061=item * 3062 3063In a regular expression pattern, a POSIX class, like C<[:ascii:]>, must 3064be inside a bracketed character class, like C<qr/[[:ascii:]]/>. A 3065warning is issued when something looking like a POSIX class is not 3066inside a bracketed class. That warning wasn't getting generated when 3067the POSIX class was negated: C<[:^ascii:]>. This is now fixed. 3068 3069=item * 3070 3071Perl 5.14.0 introduced a bug whereby S<C<eval { LABEL: }>> would crash. This 3072has been fixed. 3073L<[GH #14438]|https://github.com/Perl/perl5/issues/14438>. 3074 3075=item * 3076 3077Various crashes due to the parser getting confused by syntax errors have 3078been fixed. 3079L<[GH #14421]|https://github.com/Perl/perl5/issues/14421>. 3080L<[GH #14472]|https://github.com/Perl/perl5/issues/14472>. 3081L<[GH #14480]|https://github.com/Perl/perl5/issues/14480>. 3082L<[GH #14447]|https://github.com/Perl/perl5/issues/14447>. 3083 3084=item * 3085 3086Code like C</$a[/> used to read the next line of input and treat it as 3087though it came immediately after the opening bracket. Some invalid code 3088consequently would parse and run, but some code caused crashes, so this is 3089now disallowed. 3090L<[GH #14462]|https://github.com/Perl/perl5/issues/14462>. 3091 3092=item * 3093 3094Fix argument underflow for C<pack>. 3095L<[GH #14525]|https://github.com/Perl/perl5/issues/14525>. 3096 3097=item * 3098 3099Fix handling of non-strict C<\x{}>. Now C<\x{}> is equivalent to C<\x{0}> 3100instead of faulting. 3101 3102=item * 3103 3104C<stat -t> is now no longer treated as stackable, just like C<-t stat>. 3105L<[GH #14499]|https://github.com/Perl/perl5/issues/14499>. 3106 3107=item * 3108 3109The following no longer causes a SEGV: C<qr{x+(y(?0))*}>. 3110 3111=item * 3112 3113Fixed infinite loop in parsing backrefs in regexp patterns. 3114 3115=item * 3116 3117Several minor bug fixes in behavior of Infinity and NaN, including 3118warnings when stringifying Infinity-like or NaN-like strings. For example, 3119"NaNcy" doesn't numify to NaN anymore. 3120 3121=item * 3122 3123A bug in regular expression patterns that could lead to segfaults and 3124other crashes has been fixed. This occurred only in patterns compiled 3125with C</i> while taking into account the current POSIX locale (which usually 3126means they have to be compiled within the scope of C<S<use locale>>), 3127and there must be a string of at least 128 consecutive bytes to match. 3128L<[GH #14389]|https://github.com/Perl/perl5/issues/14389>. 3129 3130=item * 3131 3132C<s///g> now works on very long strings (where there are more than 2 3133billion iterations) instead of dying with 'Substitution loop'. 3134L<[GH #11742]|https://github.com/Perl/perl5/issues/11742>. 3135L<[GH #14190]|https://github.com/Perl/perl5/issues/14190>. 3136 3137=item * 3138 3139C<gmtime> no longer crashes with not-a-number values. 3140L<[GH #14365]|https://github.com/Perl/perl5/issues/14365>. 3141 3142=item * 3143 3144C<\()> (a reference to an empty list), and C<y///> with lexical C<$_> in 3145scope, could both do a bad write past the end of the stack. They have 3146both been fixed to extend the stack first. 3147 3148=item * 3149 3150C<prototype()> with no arguments used to read the previous item on the 3151stack, so S<C<print "foo", prototype()>> would print foo's prototype. 3152It has been fixed to infer C<$_> instead. 3153L<[GH #14376]|https://github.com/Perl/perl5/issues/14376>. 3154 3155=item * 3156 3157Some cases of lexical state subs declared inside predeclared subs could 3158crash, for example when evalling a string including the name of an outer 3159variable, but no longer do. 3160 3161=item * 3162 3163Some cases of nested lexical state subs inside anonymous subs could cause 3164'Bizarre copy' errors or possibly even crashes. 3165 3166=item * 3167 3168When trying to emit warnings, perl's default debugger (F<perl5db.pl>) was 3169sometimes giving 'Undefined subroutine &DB::db_warn called' instead. This 3170bug, which started to occur in Perl 5.18, has been fixed. 3171L<[GH #14400]|https://github.com/Perl/perl5/issues/14400>. 3172 3173=item * 3174 3175Certain syntax errors in substitutions, such as C<< s/${<>{})// >>, would 3176crash, and had done so since Perl 5.10. (In some cases the crash did not 3177start happening till 5.16.) The crash has, of course, been fixed. 3178L<[GH #14391]|https://github.com/Perl/perl5/issues/14391>. 3179 3180=item * 3181 3182Fix a couple of string grow size calculation overflows; in particular, 3183a repeat expression like S<C<33 x ~3>> could cause a large buffer 3184overflow since the new output buffer size was not correctly handled by 3185C<SvGROW()>. An expression like this now properly produces a memory wrap 3186panic. 3187L<[GH #14401]|https://github.com/Perl/perl5/issues/14401>. 3188 3189=item * 3190 3191C<< formline("@...", "a"); >> would crash. The C<FF_CHECKNL> case in 3192C<pp_formline()> didn't set the pointer used to mark the chop position, 3193which led to the C<FF_MORE> case crashing with a segmentation fault. 3194This has been fixed. 3195L<[GH #14388]|https://github.com/Perl/perl5/issues/14388>. 3196 3197=item * 3198 3199A possible buffer overrun and crash when parsing a literal pattern during 3200regular expression compilation has been fixed. 3201L<[GH #14416]|https://github.com/Perl/perl5/issues/14416>. 3202 3203=item * 3204 3205C<fchmod()> and C<futimes()> now set C<$!> when they fail due to being 3206passed a closed file handle. 3207L<[GH #14073]|https://github.com/Perl/perl5/issues/14073>. 3208 3209=item * 3210 3211C<op_free()> and C<scalarvoid()> no longer crash due to a stack overflow 3212when freeing a deeply recursive op tree. 3213L<[GH #11866]|https://github.com/Perl/perl5/issues/11866>. 3214 3215=item * 3216 3217In Perl 5.20.0, C<$^N> accidentally had the internal UTF-8 flag turned off 3218if accessed from a code block within a regular expression, effectively 3219UTF-8-encoding the value. This has been fixed. 3220L<[GH #14211]|https://github.com/Perl/perl5/issues/14211>. 3221 3222=item * 3223 3224A failed C<semctl> call no longer overwrites existing items on the stack, 3225which means that C<(semctl(-1,0,0,0))[0]> no longer gives an 3226"uninitialized" warning. 3227 3228=item * 3229 3230C<else{foo()}> with no space before C<foo> is now better at assigning the 3231right line number to that statement. 3232L<[GH #14070]|https://github.com/Perl/perl5/issues/14070>. 3233 3234=item * 3235 3236Sometimes the assignment in C<@array = split> gets optimised so that C<split> 3237itself writes directly to the array. This caused a bug, preventing this 3238assignment from being used in lvalue context. So 3239C<(@a=split//,"foo")=bar()> was an error. (This bug probably goes back to 3240Perl 3, when the optimisation was added.) It has now been fixed. 3241L<[GH #14183]|https://github.com/Perl/perl5/issues/14183>. 3242 3243=item * 3244 3245When an argument list fails the checks specified by a subroutine 3246signature (which is still an experimental feature), the resulting error 3247messages now give the file and line number of the caller, not of the 3248called subroutine. 3249L<[GH #13643]|https://github.com/Perl/perl5/issues/13643>. 3250 3251=item * 3252 3253The flip-flop operators (C<..> and C<...> in scalar context) used to maintain 3254a separate state for each recursion level (the number of times the 3255enclosing sub was called recursively), contrary to the documentation. Now 3256each closure has one internal state for each flip-flop. 3257L<[GH #14110]|https://github.com/Perl/perl5/issues/14110>. 3258 3259=item * 3260 3261The flip-flop operator (C<..> in scalar context) would return the same 3262scalar each time, unless the containing subroutine was called recursively. 3263Now it always returns a new scalar. 3264L<[GH #14110]|https://github.com/Perl/perl5/issues/14110>. 3265 3266=item * 3267 3268C<use>, C<no>, statement labels, special blocks (C<BEGIN>) and pod are now 3269permitted as the first thing in a C<map> or C<grep> block, the block after 3270C<print> or C<say> (or other functions) returning a handle, and within 3271C<${...}>, C<@{...}>, etc. 3272L<[GH #14088]|https://github.com/Perl/perl5/issues/14088>. 3273 3274=item * 3275 3276The repetition operator C<x> now propagates lvalue context to its left-hand 3277argument when used in contexts like C<foreach>. That allows 3278S<C<for(($#that_array)x2) { ... }>> to work as expected if the loop modifies 3279C<$_>. 3280 3281=item * 3282 3283C<(...) x ...> in scalar context used to corrupt the stack if one operand 3284was an object with "x" overloading, causing erratic behavior. 3285L<[GH #13811]|https://github.com/Perl/perl5/issues/13811>. 3286 3287=item * 3288 3289Assignment to a lexical scalar is often optimised away; for example in 3290C<my $x; $x = $y + $z>, the assign operator is optimised away and the add 3291operator writes its result directly to C<$x>. Various bugs related to 3292this optimisation have been fixed. Certain operators on the right-hand 3293side would sometimes fail to assign the value at all or assign the wrong 3294value, or would call STORE twice or not at all on tied variables. The 3295operators affected were C<$foo++>, C<$foo-->, and C<-$foo> under C<use 3296integer>, C<chomp>, C<chr> and C<setpgrp>. 3297 3298=item * 3299 3300List assignments were sometimes buggy if the same scalar ended up on both 3301sides of the assignment due to use of C<tied>, C<values> or C<each>. The 3302result would be the wrong value getting assigned. 3303 3304=item * 3305 3306C<setpgrp($nonzero)> (with one argument) was accidentally changed in 5.16 3307to mean C<setpgrp(0)>. This has been fixed. 3308 3309=item * 3310 3311C<__SUB__> could return the wrong value or even corrupt memory under the 3312debugger (the C<-d> switch) and in subs containing C<eval $string>. 3313 3314=item * 3315 3316When S<C<sub () { $var }>> becomes inlinable, it now returns a different 3317scalar each time, just as a non-inlinable sub would, though Perl still 3318optimises the copy away in cases where it would make no observable 3319difference. 3320 3321=item * 3322 3323S<C<my sub f () { $var }>> and S<C<sub () : attr { $var }>> are no longer 3324eligible for inlining. The former would crash; the latter would just 3325throw the attributes away. An exception is made for the little-known 3326C<:method> attribute, which does nothing much. 3327 3328=item * 3329 3330Inlining of subs with an empty prototype is now more consistent than 3331before. Previously, a sub with multiple statements, of which all but the last 3332were optimised away, would be inlinable only if it were an anonymous sub 3333containing a string C<eval> or C<state> declaration or closing over an 3334outer lexical variable (or any anonymous sub under the debugger). Now any 3335sub that gets folded to a single constant after statements have been 3336optimised away is eligible for inlining. This applies to things like C<sub 3337() { jabber() if DEBUG; 42 }>. 3338 3339Some subroutines with an explicit C<return> were being made inlinable, 3340contrary to the documentation, Now C<return> always prevents inlining. 3341 3342=item * 3343 3344On some systems, such as VMS, C<crypt> can return a non-ASCII string. If a 3345scalar assigned to had contained a UTF-8 string previously, then C<crypt> 3346would not turn off the UTF-8 flag, thus corrupting the return value. This 3347would happen with S<C<$lexical = crypt ...>>. 3348 3349=item * 3350 3351C<crypt> no longer calls C<FETCH> twice on a tied first argument. 3352 3353=item * 3354 3355An unterminated here-doc on the last line of a quote-like operator 3356(C<qq[${ <<END }]>, C</(?{ <<END })/>) no longer causes a double free. It 3357started doing so in 5.18. 3358 3359=item * 3360 3361C<index()> and C<rindex()> no longer crash when used on strings over 2GB in 3362size. 3363L<[GH #13700]|https://github.com/Perl/perl5/issues/13700>. 3364 3365=item * 3366 3367A small, previously intentional, memory leak in 3368C<PERL_SYS_INIT>/C<PERL_SYS_INIT3> on Win32 builds was fixed. This might 3369affect embedders who repeatedly create and destroy perl engines within 3370the same process. 3371 3372=item * 3373 3374C<POSIX::localeconv()> now returns the data for the program's underlying 3375locale even when called from outside the scope of S<C<use locale>>. 3376 3377=item * 3378 3379C<POSIX::localeconv()> now works properly on platforms which don't have 3380C<LC_NUMERIC> and/or C<LC_MONETARY>, or for which Perl has been compiled 3381to disregard either or both of these locale categories. In such 3382circumstances, there are now no entries for the corresponding values in 3383the hash returned by C<localeconv()>. 3384 3385=item * 3386 3387C<POSIX::localeconv()> now marks appropriately the values it returns as 3388UTF-8 or not. Previously they were always returned as bytes, even if 3389they were supposed to be encoded as UTF-8. 3390 3391=item * 3392 3393On Microsoft Windows, within the scope of C<S<use locale>>, the following 3394POSIX character classes gave results for many locales that did not 3395conform to the POSIX standard: 3396C<[[:alnum:]]>, 3397C<[[:alpha:]]>, 3398C<[[:blank:]]>, 3399C<[[:digit:]]>, 3400C<[[:graph:]]>, 3401C<[[:lower:]]>, 3402C<[[:print:]]>, 3403C<[[:punct:]]>, 3404C<[[:upper:]]>, 3405C<[[:word:]]>, 3406and 3407C<[[:xdigit:]]>. 3408This was because the underlying Microsoft implementation does not 3409follow the standard. Perl now takes special precautions to correct for 3410this. 3411 3412=item * 3413 3414Many issues have been detected by L<Coverity|http://www.coverity.com/> and 3415fixed. 3416 3417=item * 3418 3419C<system()> and friends should now work properly on more Android builds. 3420 3421Due to an oversight, the value specified through C<-Dtargetsh> to F<Configure> 3422would end up being ignored by some of the build process. This caused perls 3423cross-compiled for Android to end up with defective versions of C<system()>, 3424C<exec()> and backticks: the commands would end up looking for C</bin/sh> 3425instead of C</system/bin/sh>, and so would fail for the vast majority 3426of devices, leaving C<$!> as C<ENOENT>. 3427 3428=item * 3429 3430C<qr(...\(...\)...)>, 3431C<qr[...\[...\]...]>, 3432and 3433C<qr{...\{...\}...}> 3434now work. Previously it was impossible to escape these three 3435left-characters with a backslash within a regular expression pattern 3436where otherwise they would be considered metacharacters, and the pattern 3437opening delimiter was the character, and the closing delimiter was its 3438mirror character. 3439 3440=item * 3441 3442C<< s///e >> on tainted UTF-8 strings corrupted C<< pos() >>. This bug, 3443introduced in 5.20, is now fixed. 3444L<[GH #13948]|https://github.com/Perl/perl5/issues/13948>. 3445 3446=item * 3447 3448A non-word boundary in a regular expression (C<< \B >>) did not always 3449match the end of the string; in particular C<< q{} =~ /\B/ >> did not 3450match. This bug, introduced in perl 5.14, is now fixed. 3451L<[GH #13917]|https://github.com/Perl/perl5/issues/13917>. 3452 3453=item * 3454 3455C<< " P" =~ /(?=.*P)P/ >> should match, but did not. This is now fixed. 3456L<[GH #13954]|https://github.com/Perl/perl5/issues/13954>. 3457 3458=item * 3459 3460Failing to compile C<use Foo> in an C<eval> could leave a spurious 3461C<BEGIN> subroutine definition, which would produce a "Subroutine 3462BEGIN redefined" warning on the next use of C<use>, or other C<BEGIN> 3463block. 3464L<[GH #13926]|https://github.com/Perl/perl5/issues/13926>. 3465 3466=item * 3467 3468C<method { BLOCK } ARGS> syntax now correctly parses the arguments if they 3469begin with an opening brace. 3470L<[GH #9085]|https://github.com/Perl/perl5/issues/9085>. 3471 3472=item * 3473 3474External libraries and Perl may have different ideas of what the locale is. 3475This is problematic when parsing version strings if the locale's numeric 3476separator has been changed. Version parsing has been patched to ensure 3477it handles the locales correctly. 3478L<[GH #13863]|https://github.com/Perl/perl5/issues/13863>. 3479 3480=item * 3481 3482A bug has been fixed where zero-length assertions and code blocks inside of a 3483regex could cause C<pos> to see an incorrect value. 3484L<[GH #14016]|https://github.com/Perl/perl5/issues/14016>. 3485 3486=item * 3487 3488Dereferencing of constants now works correctly for typeglob constants. Previously 3489the glob was stringified and its name looked up. Now the glob itself is used. 3490L<[GH #9891]|https://github.com/Perl/perl5/issues/9891> 3491 3492=item * 3493 3494When parsing a sigil (C<$> C<@> C<%> C<&)> followed by braces, 3495the parser no 3496longer tries to guess whether it is a block or a hash constructor (causing a 3497syntax error when it guesses the latter), since it can only be a block. 3498 3499=item * 3500 3501S<C<undef $reference>> now frees the referent immediately, instead of hanging on 3502to it until the next statement. 3503L<[GH #14032]|https://github.com/Perl/perl5/issues/14032> 3504 3505=item * 3506 3507Various cases where the name of a sub is used (autoload, overloading, error 3508messages) used to crash for lexical subs, but have been fixed. 3509 3510=item * 3511 3512Bareword lookup now tries to avoid vivifying packages if it turns out the 3513bareword is not going to be a subroutine name. 3514 3515=item * 3516 3517Compilation of anonymous constants (I<e.g.>, C<sub () { 3 }>) no longer deletes 3518any subroutine named C<__ANON__> in the current package. Not only was 3519C<*__ANON__{CODE}> cleared, but there was a memory leak, too. This bug goes 3520back to Perl 5.8.0. 3521 3522=item * 3523 3524Stub declarations like C<sub f;> and C<sub f ();> no longer wipe out constants 3525of the same name declared by C<use constant>. This bug was introduced in Perl 35265.10.0. 3527 3528=item * 3529 3530C<qr/[\N{named sequence}]/> now works properly in many instances. 3531 3532Some names 3533known to C<\N{...}> refer to a sequence of multiple characters, instead of the 3534usual single character. Bracketed character classes generally only match 3535single characters, but now special handling has been added so that they can 3536match named sequences, but not if the class is inverted or the sequence is 3537specified as the beginning or end of a range. In these cases, the only 3538behavior change from before is a slight rewording of the fatal error message 3539given when this class is part of a C<?[...])> construct. When the C<[...]> 3540stands alone, the same non-fatal warning as before is raised, and only the 3541first character in the sequence is used, again just as before. 3542 3543=item * 3544 3545Tainted constants evaluated at compile time no longer cause unrelated 3546statements to become tainted. 3547L<[GH #14059]|https://github.com/Perl/perl5/issues/14059> 3548 3549=item * 3550 3551S<C<open $$fh, ...>>, which vivifies a handle with a name like 3552C<"main::_GEN_0">, was not giving the handle the right reference count, so 3553a double free could happen. 3554 3555=item * 3556 3557When deciding that a bareword was a method name, the parser would get confused 3558if an C<our> sub with the same name existed, and look up the method in the 3559package of the C<our> sub, instead of the package of the invocant. 3560 3561=item * 3562 3563The parser no longer gets confused by C<\U=> within a double-quoted string. It 3564used to produce a syntax error, but now compiles it correctly. 3565L<[GH #10882]|https://github.com/Perl/perl5/issues/10882> 3566 3567=item * 3568 3569It has always been the intention for the C<-B> and C<-T> file test operators to 3570treat UTF-8 encoded files as text. (L<perlfunc|perlfunc/-X FILEHANDLE> has 3571been updated to say this.) Previously, it was possible for some files to be 3572considered UTF-8 that actually weren't valid UTF-8. This is now fixed. The 3573operators now work on EBCDIC platforms as well. 3574 3575=item * 3576 3577Under some conditions warning messages raised during regular expression pattern 3578compilation were being output more than once. This has now been fixed. 3579 3580=item * 3581 3582Perl 5.20.0 introduced a regression in which a UTF-8 encoded regular 3583expression pattern that contains a single ASCII lowercase letter did not 3584match its uppercase counterpart. That has been fixed in both 5.20.1 and 35855.22.0. 3586L<[GH #14051]|https://github.com/Perl/perl5/issues/14051> 3587 3588=item * 3589 3590Constant folding could incorrectly suppress warnings if lexical warnings 3591(C<use warnings> or C<no warnings>) were not in effect and C<$^W> were 3592false at compile time and true at run time. 3593 3594=item * 3595 3596Loading Unicode tables during a regular expression match could cause assertion 3597failures under debugging builds if the previous match used the very same 3598regular expression. 3599L<[GH #14081]|https://github.com/Perl/perl5/issues/14081> 3600 3601=item * 3602 3603Thread cloning used to work incorrectly for lexical subs, possibly causing 3604crashes or double frees on exit. 3605 3606=item * 3607 3608Since Perl 5.14.0, deleting C<$SomePackage::{__ANON__}> and then undefining an 3609anonymous subroutine could corrupt things internally, resulting in 3610L<Devel::Peek> crashing or L<B.pm|B> giving nonsensical data. This has been 3611fixed. 3612 3613=item * 3614 3615S<C<(caller $n)[3]>> now reports names of lexical subs, instead of 3616treating them as C<"(unknown)">. 3617 3618=item * 3619 3620C<sort subname LIST> now supports using a lexical sub as the comparison 3621routine. 3622 3623=item * 3624 3625Aliasing (I<e.g.>, via S<C<*x = *y>>) could confuse list assignments that mention the 3626two names for the same variable on either side, causing wrong values to be 3627assigned. 3628L<[GH #5788]|https://github.com/Perl/perl5/issues/5788> 3629 3630=item * 3631 3632Long here-doc terminators could cause a bad read on short lines of input. This 3633has been fixed. It is doubtful that any crash could have occurred. This bug 3634goes back to when here-docs were introduced in Perl 3.000 twenty-five years 3635ago. 3636 3637=item * 3638 3639An optimization in C<split> to treat S<C<split /^/>> like S<C<split /^/m>> had the 3640unfortunate side-effect of also treating S<C<split /\A/>> like S<C<split /^/m>>, 3641which it should not. This has been fixed. (Note, however, that S<C<split /^x/>> 3642does not behave like S<C<split /^x/m>>, which is also considered to be a bug and 3643will be fixed in a future version.) 3644L<[GH #14086]|https://github.com/Perl/perl5/issues/14086> 3645 3646=item * 3647 3648The little-known S<C<my Class $var>> syntax (see L<fields> and L<attributes>) 3649could get confused in the scope of C<use utf8> if C<Class> were a constant 3650whose value contained Latin-1 characters. 3651 3652=item * 3653 3654Locking and unlocking values via L<Hash::Util> or C<Internals::SvREADONLY> 3655no longer has any effect on values that were read-only to begin with. 3656Previously, unlocking such values could result in crashes, hangs or 3657other erratic behavior. 3658 3659=item * 3660 3661Some unterminated C<(?(...)...)> constructs in regular expressions would 3662either crash or give erroneous error messages. C</(?(1)/> is one such 3663example. 3664 3665=item * 3666 3667S<C<pack "w", $tied>> no longer calls FETCH twice. 3668 3669=item * 3670 3671List assignments like S<C<($x, $z) = (1, $y)>> now work correctly if C<$x> and 3672C<$y> have been aliased by C<foreach>. 3673 3674=item * 3675 3676Some patterns including code blocks with syntax errors, such as 3677S<C</ (?{(^{})/>>, would hang or fail assertions on debugging builds. Now 3678they produce errors. 3679 3680=item * 3681 3682An assertion failure when parsing C<sort> with debugging enabled has been 3683fixed. 3684L<[GH #14087]|https://github.com/Perl/perl5/issues/14087>. 3685 3686=item * 3687 3688S<C<*a = *b; @a = split //, $b[1]>> could do a bad read and produce junk 3689results. 3690 3691=item * 3692 3693In S<C<() = @array = split>>, the S<C<() =>> at the beginning no longer confuses 3694the optimizer into assuming a limit of 1. 3695 3696=item * 3697 3698Fatal warnings no longer prevent the output of syntax errors. 3699L<[GH #14155]|https://github.com/Perl/perl5/issues/14155>. 3700 3701=item * 3702 3703Fixed a NaN double-to-long-double conversion error on VMS. For quiet NaNs 3704(and only on Itanium, not Alpha) negative infinity instead of NaN was 3705produced. 3706 3707=item * 3708 3709Fixed the issue that caused C<< make distclean >> to incorrectly leave some 3710files behind. 3711L<[GH #14108]|https://github.com/Perl/perl5/issues/14108>. 3712 3713=item * 3714 3715AIX now sets the length in C<< getsockopt >> correctly. 3716L<[GH #13484]|https://github.com/Perl/perl5/issues/13484>. 3717L<[cpan #91183]|https://rt.cpan.org/Ticket/Display.html?id=91183>. 3718L<[cpan #85570]|https://rt.cpan.org/Ticket/Display.html?id=85570>. 3719 3720=item * 3721 3722The optimization phase of a regexp compilation could run "forever" and 3723exhaust all memory under certain circumstances; now fixed. 3724L<[GH #13984]|https://github.com/Perl/perl5/issues/13984>. 3725 3726=item * 3727 3728The test script F<< t/op/crypt.t >> now uses the SHA-256 algorithm if the 3729default one is disabled, rather than giving failures. 3730L<[GH #13715]|https://github.com/Perl/perl5/issues/13715>. 3731 3732=item * 3733 3734Fixed an off-by-one error when setting the size of a shared array. 3735L<[GH #14151]|https://github.com/Perl/perl5/issues/14151>. 3736 3737=item * 3738 3739Fixed a bug that could cause perl to enter an infinite loop during 3740compilation. In particular, a C<while(1)> within a sublist, I<e.g.> 3741 3742 sub foo { () = ($a, my $b, ($c, do { while(1) {} })) } 3743 3744The bug was introduced in 5.20.0 3745L<[GH #14165]|https://github.com/Perl/perl5/issues/14165>. 3746 3747=item * 3748 3749On Win32, if a variable was C<local>-ized in a pseudo-process that later 3750forked, restoring the original value in the child pseudo-process caused 3751memory corruption and a crash in the child pseudo-process (and therefore the 3752OS process). 3753L<[GH #8641]|https://github.com/Perl/perl5/issues/8641>. 3754 3755=item * 3756 3757Calling C<write> on a format with a C<^**> field could produce a panic 3758in C<sv_chop()> if there were insufficient arguments or if the variable 3759used to fill the field was empty. 3760L<[GH #14255]|https://github.com/Perl/perl5/issues/14255>. 3761 3762=item * 3763 3764Non-ASCII lexical sub names now appear without trailing junk when they 3765appear in error messages. 3766 3767=item * 3768 3769The C<\@> subroutine prototype no longer flattens parenthesized arrays 3770(taking a reference to each element), but takes a reference to the array 3771itself. 3772L<[GH #9111]|https://github.com/Perl/perl5/issues/9111>. 3773 3774=item * 3775 3776A block containing nothing except a C-style C<for> loop could corrupt the 3777stack, causing lists outside the block to lose elements or have elements 3778overwritten. This could happen with C<map { for(...){...} } ...> and with 3779lists containing C<do { for(...){...} }>. 3780L<[GH #14269]|https://github.com/Perl/perl5/issues/14269>. 3781 3782=item * 3783 3784C<scalar()> now propagates lvalue context, so that 3785S<C<for(scalar($#foo)) { ... }>> can modify C<$#foo> through C<$_>. 3786 3787=item * 3788 3789C<qr/@array(?{block})/> no longer dies with "Bizarre copy of ARRAY". 3790L<[GH #14292]|https://github.com/Perl/perl5/issues/14292>. 3791 3792=item * 3793 3794S<C<eval '$variable'>> in nested named subroutines would sometimes look up a 3795global variable even with a lexical variable in scope. 3796 3797=item * 3798 3799In perl 5.20.0, C<sort CORE::fake> where 'fake' is anything other than a 3800keyword, started chopping off the last 6 characters and treating the result 3801as a sort sub name. The previous behavior of treating C<CORE::fake> as a 3802sort sub name has been restored. 3803L<[GH #14323]|https://github.com/Perl/perl5/issues/14323>. 3804 3805=item * 3806 3807Outside of C<use utf8>, a single-character Latin-1 lexical variable is 3808disallowed. The error message for it, "Can't use global C<$foo>...", was 3809giving garbage instead of the variable name. 3810 3811=item * 3812 3813C<readline> on a nonexistent handle was causing C<${^LAST_FH}> to produce a 3814reference to an undefined scalar (or fail an assertion). Now 3815C<${^LAST_FH}> ends up undefined. 3816 3817=item * 3818 3819C<(...) x ...> in void context now applies scalar context to the left-hand 3820argument, instead of the context the current sub was called in. 3821L<[GH #14174]|https://github.com/Perl/perl5/issues/14174>. 3822 3823=back 3824 3825=head1 Known Problems 3826 3827=over 4 3828 3829=item * 3830 3831C<pack>-ing a NaN on a perl compiled with Visual C 6 does not behave properly, 3832leading to a test failure in F<t/op/infnan.t>. 3833L<[GH #14705]|https://github.com/Perl/perl5/issues/14705> 3834 3835=item * 3836 3837A goal is for Perl to be able to be recompiled to work reasonably well on any 3838Unicode version. In Perl 5.22, though, the earliest such version is Unicode 38395.1 (current is 7.0). 3840 3841=item * 3842 3843EBCDIC platforms 3844 3845=over 4 3846 3847=item * 3848 3849The C<cmp> (and hence C<sort>) operators do not necessarily give the 3850correct results when both operands are UTF-EBCDIC encoded strings and 3851there is a mixture of ASCII and/or control characters, along with other 3852characters. 3853 3854=item * 3855 3856Ranges containing C<\N{...}> in the C<tr///> (and C<y///>) 3857transliteration operators are treated differently than the equivalent 3858ranges in regular expression patterns. They should, but don't, cause 3859the values in the ranges to all be treated as Unicode code points, and 3860not native ones. (L<perlre/Version 8 Regular Expressions> gives 3861details as to how it should work.) 3862 3863=item * 3864 3865Encode and encoding are mostly broken. 3866 3867=item * 3868 3869Many CPAN modules that are shipped with core show failing tests. 3870 3871=item * 3872 3873C<pack>/C<unpack> with C<"U0"> format may not work properly. 3874 3875=back 3876 3877=item * 3878 3879The following modules are known to have test failures with this version of 3880Perl. In many cases, patches have been submitted, so there will hopefully be 3881new releases soon: 3882 3883=over 3884 3885=item * 3886 3887L<B::Generate> version 1.50 3888 3889=item * 3890 3891L<B::Utils> version 0.25 3892 3893=item * 3894 3895L<Coro> version 6.42 3896 3897=item * 3898 3899L<Dancer> version 1.3130 3900 3901=item * 3902 3903L<Data::Alias> version 1.18 3904 3905=item * 3906 3907L<Data::Dump::Streamer> version 2.38 3908 3909=item * 3910 3911L<Data::Util> version 0.63 3912 3913=item * 3914 3915L<Devel::Spy> version 0.07 3916 3917=item * 3918 3919L<invoker> version 0.34 3920 3921=item * 3922 3923L<Lexical::Var> version 0.009 3924 3925=item * 3926 3927L<LWP::ConsoleLogger> version 0.000018 3928 3929=item * 3930 3931L<Mason> version 2.22 3932 3933=item * 3934 3935L<NgxQueue> version 0.02 3936 3937=item * 3938 3939L<Padre> version 1.00 3940 3941=item * 3942 3943L<Parse::Keyword> 0.08 3944 3945=back 3946 3947=back 3948 3949=head1 Obituary 3950 3951Brian McCauley died on May 8, 2015. He was a frequent poster to Usenet, Perl 3952Monks, and other Perl forums, and made several CPAN contributions under the 3953nick NOBULL, including to the Perl FAQ. He attended almost every 3954YAPC::Europe, and indeed, helped organise YAPC::Europe 2006 and the QA 3955Hackathon 2009. His wit and his delight in intricate systems were 3956particularly apparent in his love of board games; many Perl mongers will 3957have fond memories of playing Fluxx and other games with Brian. He will be 3958missed. 3959 3960=head1 Acknowledgements 3961 3962Perl 5.22.0 represents approximately 12 months of development since Perl 5.20.0 3963and contains approximately 590,000 lines of changes across 2,400 files from 94 3964authors. 3965 3966Excluding auto-generated files, documentation and release tools, there were 3967approximately 370,000 lines of changes to 1,500 .pm, .t, .c and .h files. 3968 3969Perl continues to flourish into its third decade thanks to a vibrant community 3970of users and developers. The following people are known to have contributed the 3971improvements that became Perl 5.22.0: 3972 3973Aaron Crane, Abhijit Menon-Sen, Abigail, Alberto Simões, Alex Solovey, Alex 3974Vandiver, Alexandr Ciornii, Alexandre (Midnite) Jousset, Andreas König, 3975Andreas Voegele, Andrew Fresh, Andy Dougherty, Anthony Heading, Aristotle 3976Pagaltzis, brian d foy, Brian Fraser, Chad Granum, Chris 'BinGOs' Williams, 3977Craig A. Berry, Dagfinn Ilmari Mannsåker, Daniel Dragan, Darin McBride, Dave 3978Rolsky, David Golden, David Mitchell, David Wheeler, Dmitri Tikhonov, Doug 3979Bell, E. Choroba, Ed J, Eric Herman, Father Chrysostomos, George Greer, Glenn 3980D. Golden, Graham Knop, H.Merijn Brand, Herbert Breunung, Hugo van der Sanden, 3981James E Keenan, James McCoy, James Raspass, Jan Dubois, Jarkko Hietaniemi, 3982Jasmine Ngan, Jerry D. Hedden, Jim Cromie, John Goodyear, kafka, Karen 3983Etheridge, Karl Williamson, Kent Fredric, kmx, Lajos Veres, Leon Timmermans, 3984Lukas Mai, Mathieu Arnold, Matthew Horsfall, Max Maischein, Michael Bunk, 3985Nicholas Clark, Niels Thykier, Niko Tyni, Norman Koch, Olivier Mengué, Peter 3986John Acklam, Peter Martini, Petr Písař, Philippe Bruhat (BooK), Pierre 3987Bogossian, Rafael Garcia-Suarez, Randy Stauner, Reini Urban, Ricardo Signes, 3988Rob Hoelz, Rostislav Skudnov, Sawyer X, Shirakata Kentaro, Shlomi Fish, 3989Sisyphus, Slaven Rezic, Smylers, Steffen Müller, Steve Hay, Sullivan Beck, 3990syber, Tadeusz Sośnierz, Thomas Sibley, Todd Rinaldo, Tony Cook, Vincent Pit, 3991Vladimir Marek, Yaroslav Kuzmin, Yves Orton, Ævar Arnfjörð Bjarmason. 3992 3993The list above is almost certainly incomplete as it is automatically generated 3994from version control history. In particular, it does not include the names of 3995the (very much appreciated) contributors who reported issues to the Perl bug 3996tracker. 3997 3998Many of the changes included in this version originated in the CPAN modules 3999included in Perl's core. We're grateful to the entire CPAN community for 4000helping Perl to flourish. 4001 4002For a more complete list of all of Perl's historical contributors, please see 4003the F<AUTHORS> file in the Perl source distribution. 4004 4005=head1 Reporting Bugs 4006 4007If you find what you think is a bug, you might check the articles recently 4008posted to the comp.lang.perl.misc newsgroup and the perl bug database at 4009L<https://rt.perl.org/>. There may also be information at 4010L<http://www.perl.org/>, the Perl Home Page. 4011 4012If you believe you have an unreported bug, please run the L<perlbug> program 4013included with your release. Be sure to trim your bug down to a tiny but 4014sufficient test case. Your bug report, along with the output of C<perl -V>, 4015will be sent off to perlbug@perl.org to be analysed by the Perl porting team. 4016 4017If the bug you are reporting has security implications, which make it 4018inappropriate to send to a publicly archived mailing list, then please send it 4019to perl5-security-report@perl.org. This points to a closed subscription 4020unarchived mailing list, which includes all the core committers, who will be 4021able to help assess the impact of issues, figure out a resolution, and help 4022co-ordinate the release of patches to mitigate or fix the problem across all 4023platforms on which Perl is supported. Please only use this address for 4024security issues in the Perl core, not for modules independently distributed on 4025CPAN. 4026 4027=head1 SEE ALSO 4028 4029The F<Changes> file for an explanation of how to view exhaustive details on 4030what changed. 4031 4032The F<INSTALL> file for how to build Perl. 4033 4034The F<README> file for general stuff. 4035 4036The F<Artistic> and F<Copying> files for copyright information. 4037 4038=cut 4039