5 perl5200delta - what is new for perl v5.20.0
9 This document describes differences between the 5.18.0 release and the
12 If you are upgrading from an earlier release such as 5.16.0, first read
13 L<perl5180delta>, which describes differences between 5.16.0 and 5.18.0.
15 =head1 Core Enhancements
17 =head2 Experimental Subroutine signatures
19 Declarative syntax to unwrap argument list into lexical variables.
20 C<sub foo ($a,$b) {...}> checks the number of arguments and puts the
21 arguments into lexical variables. Signatures are not equivalent to
22 the existing idiom of C<sub foo { my($a,$b) = @_; ... }>. Signatures
23 are only available by enabling a non-default feature, and generate
24 warnings about being experimental. The syntactic clash with
25 prototypes is managed by disabling the short prototype syntax when
26 signatures are enabled.
28 See L<perlsub/Signatures> for details.
30 =head2 C<sub>s now take a C<prototype> attribute
32 When declaring or defining a C<sub>, the prototype can now be specified inside
33 of a C<prototype> attribute instead of in parens following the name.
35 For example, C<sub foo($$){}> could be rewritten as
36 C<sub foo : prototype($$){}>.
38 =head2 More consistent prototype parsing
40 Multiple semicolons in subroutine prototypes have long been tolerated and
41 treated as a single semicolon. There was one case where this did not
42 happen. A subroutine whose prototype begins with "*" or ";*" can affect
43 whether a bareword is considered a method name or sub call. This now
44 applies also to ";;;*".
46 Whitespace has long been allowed inside subroutine prototypes, so
47 C<sub( $ $ )> is equivalent to C<sub($$)>, but until now it was stripped
48 when the subroutine was parsed. Hence, whitespace was I<not> allowed in
49 prototypes set by C<Scalar::Util::set_prototype>. Now it is permitted,
50 and the parser no longer strips whitespace. This means
51 C<prototype &mysub> returns the original prototype, whitespace and all.
53 =head2 C<rand> now uses a consistent random number generator
55 Previously perl would use a platform specific random number generator, varying
56 between the libc rand(), random() or drand48().
58 This meant that the quality of perl's random numbers would vary from platform
59 to platform, from the 15 bits of rand() on Windows to 48-bits on POSIX
60 platforms such as Linux with drand48().
62 Perl now uses its own internal drand48() implementation on all platforms. This
63 does not make perl's C<rand> cryptographically secure. [perl #115928]
65 =head2 New slice syntax
67 The new C<%hash{...}> and C<%array[...]> syntax returns a list of key/value (or
68 index/value) pairs. See L<perldata/"Key/Value Hash Slices">.
70 =head2 Experimental Postfix Dereferencing
72 When the C<postderef> feature is in effect, the following syntactical
73 equivalencies are set up:
75 $sref->$*; # same as ${ $sref } # interpolates
76 $aref->@*; # same as @{ $aref } # interpolates
77 $href->%*; # same as %{ $href }
78 $cref->&*; # same as &{ $cref }
79 $gref->**; # same as *{ $gref }
81 $aref->$#*; # same as $#{ $aref }
83 $gref->*{ $slot }; # same as *{ $gref }{ $slot }
85 $aref->@[ ... ]; # same as @$aref[ ... ] # interpolates
86 $href->@{ ... }; # same as @$href{ ... } # interpolates
87 $aref->%[ ... ]; # same as %$aref[ ... ]
88 $href->%{ ... }; # same as %$href{ ... }
90 Those marked as interpolating only interpolate if the associated
91 C<postderef_qq> feature is also enabled. This feature is B<experimental> and
92 will trigger C<experimental::postderef>-category warnings when used, unless
95 For more information, consult L<the Postfix Dereference Syntax section of
96 perlref|perlref/Postfix Dereference Syntax>.
98 =head2 Unicode 6.3 now supported
100 Perl now supports and is shipped with Unicode 6.3 (though Perl may be
101 recompiled with any previous Unicode release as well). A detailed list of
102 Unicode 6.3 changes is at L<http://www.unicode.org/versions/Unicode6.3.0/>.
104 =head2 New C<\p{Unicode}> regular expression pattern property
106 This is a synonym for C<\p{Any}> and matches the set of Unicode-defined
107 code points 0 - 0x10FFFF.
109 =head2 Better 64-bit support
111 On 64-bit platforms, the internal array functions now use 64-bit offsets,
112 allowing Perl arrays to hold more than 2**31 elements, if you have the memory
115 The regular expression engine now supports strings longer than 2**31
116 characters. [perl #112790, #116907]
118 The functions PerlIO_get_bufsiz, PerlIO_get_cnt, PerlIO_set_cnt and
119 PerlIO_set_ptrcnt now have SSize_t, rather than int, return values and
122 =head2 C<S<use locale>> now works on UTF-8 locales
124 Until this release, only single-byte locales, such as the ISO 8859
125 series were supported. Now, the increasingly common multi-byte UTF-8
126 locales are also supported. A UTF-8 locale is one in which the
127 character set is Unicode and the encoding is UTF-8. The POSIX
128 C<LC_CTYPE> category operations (case changing (like C<lc()>, C<"\U">),
129 and character classification (C<\w>, C<\D>, C<qr/[[:punct:]]/>)) under
130 such a locale work just as if not under locale, but instead as if under
131 C<S<use feature 'unicode_strings'>>, except taint rules are followed.
132 Sorting remains by code point order in this release. [perl #56820].
134 =head2 C<S<use locale>> now compiles on systems without locale ability
136 Previously doing this caused the program to not compile. Within its
137 scope the program behaves as if in the "C" locale. Thus programs
138 written for platforms that support locales can run on locale-less
139 platforms without change. Attempts to change the locale away from the
140 "C" locale will, of course, fail.
142 =head2 More locale initialization fallback options
144 If there was an error with locales during Perl start-up, it immediately
145 gave up and tried to use the C<"C"> locale. Now it first tries using
146 other locales given by the environment variables, as detailed in
147 L<perllocale/ENVIRONMENT>. For example, if C<LC_ALL> and C<LANG> are
148 both set, and using the C<LC_ALL> locale fails, Perl will now try the
149 C<LANG> locale, and only if that fails, will it fall back to C<"C">. On
150 Windows machines, Perl will try, ahead of using C<"C">, the system
151 default locale if all the locales given by environment variables fail.
153 =head2 C<-DL> runtime option now added for tracing locale setting
155 This is designed for Perl core developers to aid in field debugging bugs
158 =head2 B<-F> now implies B<-a> and B<-a> implies B<-n>
160 Previously B<-F> without B<-a> was a no-op, and B<-a> without B<-n> or B<-p>
161 was a no-op, with this change, if you supply B<-F> then both B<-a> and B<-n>
162 are implied and if you supply B<-a> then B<-n> is implied.
164 You can still use B<-p> for its extra behaviour. [perl #116190]
166 =head2 $a and $b warnings exemption
168 The special variables $a and $b, used in C<sort>, are now exempt from "used
169 once" warnings, even where C<sort> is not used. This makes it easier for
170 CPAN modules to provide functions using $a and $b for similar purposes.
175 =head2 Avoid possible read of free()d memory during parsing
177 It was possible that free()d memory could be read during parsing in the unusual
178 circumstance of the Perl program ending with a heredoc and the last line of the
179 file on disk having no terminating newline character. This has now been fixed.
181 =head1 Incompatible Changes
183 =head2 C<do> can no longer be used to call subroutines
185 The C<do SUBROUTINE(LIST)> form has resulted in a deprecation warning
186 since Perl v5.0.0, and is now a syntax error.
188 =head2 Quote-like escape changes
190 The character after C<\c> in a double-quoted string ("..." or qq(...))
191 or regular expression must now be a printable character and may not be
194 A literal C<{> after C<\B> or C<\b> is now fatal.
196 These were deprecated in perl v5.14.0.
198 =head2 Tainting happens under more circumstances; now conforms to documentation
200 This affects regular expression matching and changing the case of a
201 string (C<lc>, C<"\U">, I<etc>.) within the scope of C<use locale>.
202 The result is now tainted based on the operation, no matter what the
203 contents of the string were, as the documentation (L<perlsec>,
204 L<perllocale/SECURITY>) indicates it should. Previously, for the case
205 change operation, if the string contained no characters whose case
206 change could be affected by the locale, the result would not be tainted.
207 For example, the result of C<uc()> on an empty string or one containing
208 only above-Latin1 code points is now tainted, and wasn't before. This
209 leads to more consistent tainting results. Regular expression patterns
210 taint their non-binary results (like C<$&>, C<$2>) if and only if the
211 pattern contains elements whose matching depends on the current
212 (potentially tainted) locale. Like the case changing functions, the
213 actual contents of the string being matched now do not matter, whereas
214 formerly it did. For example, if the pattern contains a C<\w>, the
215 results will be tainted even if the match did not have to use that
216 portion of the pattern to succeed or fail, because what a C<\w> matches
217 depends on locale. However, for example, a C<.> in a pattern will not
218 enable tainting, because the dot matches any single character, and what
219 the current locale is doesn't change in any way what matches and what
222 =head2 C<\p{}>, C<\P{}> matching has changed for non-Unicode code
225 C<\p{}> and C<\P{}> are defined by Unicode only on Unicode-defined code
226 points (C<U+0000> through C<U+10FFFF>). Their behavior on matching
227 these legal Unicode code points is unchanged, but there are changes for
228 code points C<0x110000> and above. Previously, Perl treated the result
229 of matching C<\p{}> and C<\P{}> against these as C<undef>, which
230 translates into "false". For C<\P{}>, this was then complemented into
231 "true". A warning was supposed to be raised when this happened.
232 However, various optimizations could prevent the warning, and the
233 results were often counter-intuitive, with both a match and its seeming
234 complement being false. Now all non-Unicode code points are treated as
235 typical unassigned Unicode code points. This generally is more
236 Do-What-I-Mean. A warning is raised only if the results are arguably
237 different from a strict Unicode approach, and from what Perl used to do.
238 Code that needs to be strictly Unicode compliant can make this warning
239 fatal, and then Perl always raises the warning.
241 Details are in L<perlunicode/Beyond Unicode code points>.
243 =head2 C<\p{All}> has been expanded to match all possible code points
245 The Perl-defined regular expression pattern element C<\p{All}>, unused
246 on CPAN, used to match just the Unicode code points; now it matches all
247 possible code points; that is, it is equivalent to C<qr/./s>. Thus
248 C<\p{All}> is no longer synonymous with C<\p{Any}>, which continues to
249 match just the Unicode code points, as Unicode says it should.
251 =head2 Data::Dumper's output may change
253 Depending on the data structures dumped and the settings set for
254 Data::Dumper, the dumped output may have changed from previous
257 If you have tests that depend on the exact output of Data::Dumper,
260 To avoid this problem in your code, test against the data structure
261 from evaluating the dumped structure, instead of the dump itself.
263 =head2 Locale decimal point character no longer leaks outside of S<C<use locale>> scope
265 This is actually a bug fix, but some code has come to rely on the bug
266 being present, so this change is listed here. The current locale that
267 the program is running under is not supposed to be visible to Perl code
268 except within the scope of a S<C<use locale>>. However, until now under
269 certain circumstances, the character used for a decimal point (often a
270 comma) leaked outside the scope. If your code is affected by this
271 change, simply add a S<C<use locale>>.
273 =head2 Assignments of Windows sockets error codes to $! now prefer F<errno.h> values over WSAGetLastError() values
275 In previous versions of Perl, Windows sockets error codes as returned by
276 WSAGetLastError() were assigned to $!, and some constants such as ECONNABORTED,
277 not in F<errno.h> in VC++ (or the various Windows ports of gcc) were defined to
278 corresponding WSAE* values to allow $! to be tested against the E* constants
279 exported by L<Errno> and L<POSIX>.
281 This worked well until VC++ 2010 and later, which introduced new E* constants
282 with values E<gt> 100 into F<errno.h>, including some being (re)defined by perl
283 to WSAE* values. That caused problems when linking XS code against other
284 libraries which used the original definitions of F<errno.h> constants.
286 To avoid this incompatibility, perl now maps WSAE* error codes to E* values
287 where possible, and assigns those values to $!. The E* constants exported by
288 L<Errno> and L<POSIX> are updated to match so that testing $! against them,
289 wherever previously possible, will continue to work as expected, and all E*
290 constants found in F<errno.h> are now exported from those modules with their
291 original F<errno.h> values.
293 In order to avoid breakage in existing Perl code which assigns WSAE* values to
294 $!, perl now intercepts the assignment and performs the same mapping to E*
295 values as it uses internally when assigning to $! itself.
297 However, one backwards-incompatibility remains: existing Perl code which
298 compares $! against the numeric values of the WSAE* error codes that were
299 previously assigned to $! will now be broken in those cases where a
300 corresponding E* value has been assigned instead. This is only an issue for
301 those E* values E<lt> 100, which were always exported from L<Errno> and
302 L<POSIX> with their original F<errno.h> values, and therefore could not be used
303 for WSAE* error code tests (e.g. WSAEINVAL is 10022, but the corresponding
304 EINVAL is 22). (E* values E<gt> 100, if present, were redefined to WSAE*
305 values anyway, so compatibility can be achieved by using the E* constants,
306 which will work both before and after this change, albeit using different
307 numeric values under the hood.)
309 =head2 Functions C<PerlIO_vsprintf> and C<PerlIO_sprintf> have been removed
311 These two functions, undocumented, unused in CPAN, and problematic, have been
316 =head2 The C</\C/> character class
318 The C</\C/> regular expression character class is deprecated. From perl
319 5.22 onwards it will generate a warning, and from perl 5.24 onwards it
320 will be a regular expression compiler error. If you need to examine the
321 individual bytes that make up a UTF8-encoded character, then use
322 C<utf8::encode()> on the string (or a copy) first.
324 =head2 Literal control characters in variable names
326 This deprecation affects things like $\cT, where \cT is a literal control (such
327 as a C<NAK> or C<NEGATIVE ACKNOWLEDGE> character) in
328 the source code. Surprisingly, it appears that originally this was intended as
329 the canonical way of accessing variables like $^T, with the caret form only
330 being added as an alternative.
332 The literal control form is being deprecated for two main reasons. It has what
333 are likely unfixable bugs, such as $\cI not working as an alias for $^I, and
334 their usage not being portable to non-ASCII platforms: While $^T will work
335 everywhere, \cT is whitespace in EBCDIC. [perl #119123]
337 =head2 References to non-integers and non-positive integers in C<$/>
339 Setting C<$/> to a reference to zero or a reference to a negative integer is
340 now deprecated, and will behave B<exactly> as though it was set to C<undef>.
341 If you want slurp behavior set C<$/> to C<undef> explicitly.
343 Setting C<$/> to a reference to a non integer is now forbidden and will
344 throw an error. Perl has never documented what would happen in this
345 context and while it used to behave the same as setting C<$/> to
346 the address of the references in future it may behave differently, so we
347 have forbidden this usage.
349 =head2 Character matching routines in POSIX
351 Use of any of these functions in the C<POSIX> module is now deprecated:
352 C<isalnum>, C<isalpha>, C<iscntrl>, C<isdigit>, C<isgraph>, C<islower>,
353 C<isprint>, C<ispunct>, C<isspace>, C<isupper>, and C<isxdigit>. The
354 functions are buggy and don't work on UTF-8 encoded strings. See their
355 entries in L<POSIX> for more information.
357 A warning is raised on the first call to any of them from each place in
358 the code that they are called. (Hence a repeated statement in a loop
359 will raise just the one warning.)
361 =head2 Interpreter-based threads are now I<discouraged>
363 The "interpreter-based threads" provided by Perl are not the fast, lightweight
364 system for multitasking that one might expect or hope for. Threads are
365 implemented in a way that make them easy to misuse. Few people know how to
366 use them correctly or will be able to provide help.
368 The use of interpreter-based threads in perl is officially
369 L<discouraged|perlpolicy/discouraged>.
371 =head2 Module removals
373 The following modules will be removed from the core distribution in a
374 future release, and will at that time need to be installed from CPAN.
375 Distributions on CPAN which require these modules will need to list them as
378 The core versions of these modules will now issue C<"deprecated">-category
379 warnings to alert you to this fact. To silence these deprecation warnings,
380 install the modules in question from CPAN.
382 Note that the planned removal of these modules from core does not reflect a
383 judgement about the quality of the code and should not be taken as a suggestion
384 that their use be halted. Their disinclusion from core primarily hinges on
385 their necessity to bootstrapping a fully functional, CPAN-capable Perl
386 installation, not on concerns over their design.
390 =item L<CGI> and its associated CGI:: packages
394 =item L<Package::Constants>
396 =item L<Module::Build> and its associated Module::Build:: packages
400 =head2 Utility removals
402 The following utilities will be removed from the core distribution in a
403 future release, and will at that time need to be installed from CPAN.
415 =head1 Performance Enhancements
421 Perl has a new copy-on-write mechanism that avoids the need to copy the
422 internal string buffer when assigning from one scalar to another. This
423 makes copying large strings appear much faster. Modifying one of the two
424 (or more) strings after an assignment will force a copy internally. This
425 makes it unnecessary to pass strings by reference for efficiency.
427 This feature was already available in 5.18.0, but wasn't enabled by
428 default. It is the default now, and so you no longer need build perl with
429 the F<Configure> argument:
431 -Accflags=-DPERL_NEW_COPY_ON_WRITE
433 It can be disabled (for now) in a perl build with:
435 -Accflags=-DPERL_NO_COW
437 On some operating systems Perl can be compiled in such a way that any
438 attempt to modify string buffers shared by multiple SVs will crash. This
439 way XS authors can test that their modules handle copy-on-write scalars
440 correctly. See L<perlguts/"Copy on Write"> for detail.
444 Perl has an optimizer for regular expression patterns. It analyzes the pattern
445 to find things such as the minimum length a string has to be to match, etc. It
446 now better handles code points that are above the Latin1 range.
450 Executing a regex that contains the C<^> anchor (or its variant under the
451 C</m> flag) has been made much faster in several situations.
455 Precomputed hash values are now used in more places during method lookup.
459 Constant hash key lookups (C<$hash{key}> as opposed to C<$hash{$key}>) have
460 long had the internal hash value computed at compile time, to speed up
461 lookup. This optimisation has only now been applied to hash slices as
466 Combined C<and> and C<or> operators in void context, like those
467 generated for C<< unless ($a && $b) >> and C<< if ($a || b) >> now
468 short circuit directly to the end of the statement. [perl #120128]
472 In certain situations, when C<return> is the last statement in a subroutine's
473 main scope, it will be optimized out. This means code like:
475 sub baz { return $cat; }
477 will now behave like:
481 which is notably faster.
496 In combination with the L<padrange optimization introduced in
497 v5.18.0|perl5180delta/Internal Changes>, this means longer uninitialized my
498 variable statements are also optimized, so:
510 The creation of certain sorts of lists, including array and hash slices, is now
515 The optimisation for arrays indexed with a small constant integer is now
516 applied for integers in the range -128..127, rather than 0..255. This should
517 speed up Perl code using expressions like C<$x[-1]>, at the expense of
518 (presumably much rarer) code using expressions like C<$x[200]>.
522 The first iteration over a large hash (using C<keys> or C<each>) is now
523 faster. This is achieved by preallocating the hash's internal iterator
524 state, rather than lazily creating it when the hash is first iterated. (For
525 small hashes, the iterator is still created only when first needed. The
526 assumption is that small hashes are more likely to be used as objects, and
527 therefore never allocated. For large hashes, that's less likely to be true,
528 and the cost of allocating the iterator is swamped by the cost of allocating
529 space for the hash itself.)
533 When doing a global regex match on a string that came from the C<readline>
534 or C<E<lt>E<gt>> operator, the data is no longer copied unnecessarily.
539 Dereferencing (as in C<$obj-E<gt>[0]> or C<$obj-E<gt>{k}>) is now faster
540 when C<$obj> is an instance of a class that has overloaded methods, but
541 doesn't overload any of the dereferencing methods C<@{}>, C<%{}>, and so on.
545 Perl's optimiser no longer skips optimising code that follows certain
546 C<eval {}> expressions (including those with an apparent infinite loop).
550 The implementation now does a better job of avoiding meaningless work at
551 runtime. Internal effect-free "null" operations (created as a side-effect of
552 parsing Perl programs) are normally deleted during compilation. That
553 deletion is now applied in some situations that weren't previously handled.
557 Perl now does less disk I/O when dealing with Unicode properties that cover
558 up to three ranges of consecutive code points.
562 =head1 Modules and Pragmata
564 =head2 New Modules and Pragmata
570 L<experimental> 0.007 has been added to the Perl core.
574 L<IO::Socket::IP> 0.29 has been added to the Perl core.
578 =head2 Updated Modules and Pragmata
584 L<Archive::Tar> has been upgraded from version 1.90 to 1.96.
588 L<arybase> has been upgraded from version 0.06 to 0.07.
592 L<Attribute::Handlers> has been upgraded from version 0.94 to 0.96.
596 L<attributes> has been upgraded from version 0.21 to 0.22.
600 L<autodie> has been upgraded from version 2.13 to 2.23.
604 L<AutoLoader> has been upgraded from version 5.73 to 5.74.
608 L<autouse> has been upgraded from version 1.07 to 1.08.
612 L<B> has been upgraded from version 1.42 to 1.48.
616 L<B::Concise> has been upgraded from version 0.95 to 0.992.
620 L<B::Debug> has been upgraded from version 1.18 to 1.19.
624 L<B::Deparse> has been upgraded from version 1.20 to 1.26.
628 L<base> has been upgraded from version 2.18 to 2.22.
632 L<Benchmark> has been upgraded from version 1.15 to 1.18.
636 L<bignum> has been upgraded from version 0.33 to 0.37.
640 L<Carp> has been upgraded from version 1.29 to 1.3301.
644 L<CGI> has been upgraded from version 3.63 to 3.65.
645 NOTE: L<CGI> is deprecated and may be removed from a future version of Perl.
649 L<charnames> has been upgraded from version 1.36 to 1.40.
653 L<Class::Struct> has been upgraded from version 0.64 to 0.65.
657 L<Compress::Raw::Bzip2> has been upgraded from version 2.060 to 2.064.
661 L<Compress::Raw::Zlib> has been upgraded from version 2.060 to 2.065.
665 L<Config::Perl::V> has been upgraded from version 0.17 to 0.20.
669 L<constant> has been upgraded from version 1.27 to 1.31.
673 L<CPAN> has been upgraded from version 2.00 to 2.05.
677 L<CPAN::Meta> has been upgraded from version 2.120921 to 2.140640.
681 L<CPAN::Meta::Requirements> has been upgraded from version 2.122 to 2.125.
685 L<CPAN::Meta::YAML> has been upgraded from version 0.008 to 0.012.
689 L<Data::Dumper> has been upgraded from version 2.145 to 2.151.
693 L<DB> has been upgraded from version 1.04 to 1.07.
697 L<DB_File> has been upgraded from version 1.827 to 1.831.
701 L<DBM_Filter> has been upgraded from version 0.05 to 0.06.
705 L<deprecate> has been upgraded from version 0.02 to 0.03.
709 L<Devel::Peek> has been upgraded from version 1.11 to 1.16.
713 L<Devel::PPPort> has been upgraded from version 3.20 to 3.21.
717 L<diagnostics> has been upgraded from version 1.31 to 1.34.
721 L<Digest::MD5> has been upgraded from version 2.52 to 2.53.
725 L<Digest::SHA> has been upgraded from version 5.84 to 5.88.
729 L<DynaLoader> has been upgraded from version 1.18 to 1.25.
733 L<Encode> has been upgraded from version 2.49 to 2.60.
737 L<encoding> has been upgraded from version 2.6_01 to 2.12.
741 L<English> has been upgraded from version 1.06 to 1.09.
745 L<Errno> has been upgraded from version 1.18 to 1.20_03.
749 L<Exporter> has been upgraded from version 5.68 to 5.70.
753 L<ExtUtils::CBuilder> has been upgraded from version 0.280210 to 0.280216.
757 L<ExtUtils::Command> has been upgraded from version 1.17 to 1.18.
761 L<ExtUtils::Embed> has been upgraded from version 1.30 to 1.32.
765 L<ExtUtils::Install> has been upgraded from version 1.59 to 1.67.
769 L<ExtUtils::MakeMaker> has been upgraded from version 6.66 to 6.98.
773 L<ExtUtils::Miniperl> has been upgraded from version to 1.01.
777 L<ExtUtils::ParseXS> has been upgraded from version 3.18 to 3.24.
781 L<ExtUtils::Typemaps> has been upgraded from version 3.19 to 3.24.
785 L<ExtUtils::XSSymSet> has been upgraded from version 1.2 to 1.3.
789 L<feature> has been upgraded from version 1.32 to 1.36.
793 L<fields> has been upgraded from version 2.16 to 2.17.
797 L<File::Basename> has been upgraded from version 2.84 to 2.85.
801 L<File::Copy> has been upgraded from version 2.26 to 2.29.
805 L<File::DosGlob> has been upgraded from version 1.10 to 1.12.
809 L<File::Fetch> has been upgraded from version 0.38 to 0.48.
813 L<File::Find> has been upgraded from version 1.23 to 1.27.
817 L<File::Glob> has been upgraded from version 1.20 to 1.23.
821 L<File::Spec> has been upgraded from version 3.40 to 3.47.
825 L<File::Temp> has been upgraded from version 0.23 to 0.2304.
829 L<FileCache> has been upgraded from version 1.08 to 1.09.
833 L<Filter::Simple> has been upgraded from version 0.89 to 0.91.
837 L<Filter::Util::Call> has been upgraded from version 1.45 to 1.49.
841 L<Getopt::Long> has been upgraded from version 2.39 to 2.42.
845 L<Getopt::Std> has been upgraded from version 1.07 to 1.10.
849 L<Hash::Util::FieldHash> has been upgraded from version 1.10 to 1.15.
853 L<HTTP::Tiny> has been upgraded from version 0.025 to 0.043.
857 L<I18N::Langinfo> has been upgraded from version 0.10 to 0.11.
861 L<I18N::LangTags> has been upgraded from version 0.39 to 0.40.
865 L<if> has been upgraded from version 0.0602 to 0.0603.
869 L<inc::latest> has been upgraded from version 0.4003 to 0.4205.
870 NOTE: L<inc::latest> is deprecated and may be removed from a future version of Perl.
874 L<integer> has been upgraded from version 1.00 to 1.01.
878 L<IO> has been upgraded from version 1.28 to 1.31.
882 L<IO::Compress::Gzip> and friends have been upgraded from version 2.060 to
887 L<IPC::Cmd> has been upgraded from version 0.80 to 0.92.
891 L<IPC::Open3> has been upgraded from version 1.13 to 1.16.
895 L<IPC::SysV> has been upgraded from version 2.03 to 2.04.
899 L<JSON::PP> has been upgraded from version 2.27202 to 2.27203.
903 L<List::Util> has been upgraded from version 1.27 to 1.38.
907 L<locale> has been upgraded from version 1.02 to 1.03.
911 L<Locale::Codes> has been upgraded from version 3.25 to 3.30.
915 L<Locale::Maketext> has been upgraded from version 1.23 to 1.25.
919 L<Math::BigInt> has been upgraded from version 1.9991 to 1.9993.
923 L<Math::BigInt::FastCalc> has been upgraded from version 0.30 to 0.31.
927 L<Math::BigRat> has been upgraded from version 0.2604 to 0.2606.
931 L<MIME::Base64> has been upgraded from version 3.13 to 3.14.
935 L<Module::Build> has been upgraded from version 0.4003 to 0.4205.
936 NOTE: L<Module::Build> is deprecated and may be removed from a future version of Perl.
940 L<Module::CoreList> has been upgraded from version 2.89 to 3.10.
944 L<Module::Load> has been upgraded from version 0.24 to 0.32.
948 L<Module::Load::Conditional> has been upgraded from version 0.54 to 0.62.
952 L<Module::Metadata> has been upgraded from version 1.000011 to 1.000019.
956 L<mro> has been upgraded from version 1.11 to 1.16.
960 L<Net::Ping> has been upgraded from version 2.41 to 2.43.
964 L<Opcode> has been upgraded from version 1.25 to 1.27.
968 L<Package::Constants> has been upgraded from version 0.02 to 0.04.
969 NOTE: L<Package::Constants> is deprecated and may be removed from a future version of Perl.
973 L<Params::Check> has been upgraded from version 0.36 to 0.38.
977 L<parent> has been upgraded from version 0.225 to 0.228.
981 L<Parse::CPAN::Meta> has been upgraded from version 1.4404 to 1.4414.
985 L<Perl::OSType> has been upgraded from version 1.003 to 1.007.
989 L<perlfaq> has been upgraded from version 5.0150042 to 5.0150044.
993 L<PerlIO> has been upgraded from version 1.07 to 1.09.
997 L<PerlIO::encoding> has been upgraded from version 0.16 to 0.18.
1001 L<PerlIO::scalar> has been upgraded from version 0.16 to 0.18.
1005 L<PerlIO::via> has been upgraded from version 0.12 to 0.14.
1009 L<Pod::Escapes> has been upgraded from version 1.04 to 1.06.
1013 L<Pod::Functions> has been upgraded from version 1.06 to 1.08.
1017 L<Pod::Html> has been upgraded from version 1.18 to 1.21.
1021 L<Pod::Parser> has been upgraded from version 1.60 to 1.62.
1025 L<Pod::Perldoc> has been upgraded from version 3.19 to 3.23.
1029 L<Pod::Usage> has been upgraded from version 1.61 to 1.63.
1033 L<POSIX> has been upgraded from version 1.32 to 1.38_03.
1037 L<re> has been upgraded from version 0.23 to 0.26.
1041 L<Safe> has been upgraded from version 2.35 to 2.37.
1045 L<Scalar::Util> has been upgraded from version 1.27 to 1.38.
1049 L<SDBM_File> has been upgraded from version 1.09 to 1.11.
1053 L<Socket> has been upgraded from version 2.009 to 2.013.
1057 L<Storable> has been upgraded from version 2.41 to 2.49.
1061 L<strict> has been upgraded from version 1.07 to 1.08.
1065 L<subs> has been upgraded from version 1.01 to 1.02.
1069 L<Sys::Hostname> has been upgraded from version 1.17 to 1.18.
1073 L<Sys::Syslog> has been upgraded from version 0.32 to 0.33.
1077 L<Term::Cap> has been upgraded from version 1.13 to 1.15.
1081 L<Term::ReadLine> has been upgraded from version 1.12 to 1.14.
1085 L<Test::Harness> has been upgraded from version 3.26 to 3.30.
1089 L<Test::Simple> has been upgraded from version 0.98 to 1.001002.
1093 L<Text::ParseWords> has been upgraded from version 3.28 to 3.29.
1097 L<Text::Tabs> has been upgraded from version 2012.0818 to 2013.0523.
1101 L<Text::Wrap> has been upgraded from version 2012.0818 to 2013.0523.
1105 L<Thread> has been upgraded from version 3.02 to 3.04.
1109 L<Thread::Queue> has been upgraded from version 3.02 to 3.05.
1113 L<threads> has been upgraded from version 1.86 to 1.93.
1117 L<threads::shared> has been upgraded from version 1.43 to 1.46.
1121 L<Tie::Array> has been upgraded from version 1.05 to 1.06.
1125 L<Tie::File> has been upgraded from version 0.99 to 1.00.
1129 L<Tie::Hash> has been upgraded from version 1.04 to 1.05.
1133 L<Tie::Scalar> has been upgraded from version 1.02 to 1.03.
1137 L<Tie::StdHandle> has been upgraded from version 4.3 to 4.4.
1141 L<Time::HiRes> has been upgraded from version 1.9725 to 1.9726.
1145 L<Time::Piece> has been upgraded from version 1.20_01 to 1.27.
1149 L<Unicode::Collate> has been upgraded from version 0.97 to 1.04.
1153 L<Unicode::Normalize> has been upgraded from version 1.16 to 1.17.
1157 L<Unicode::UCD> has been upgraded from version 0.51 to 0.57.
1161 L<utf8> has been upgraded from version 1.10 to 1.13.
1165 L<version> has been upgraded from version 0.9902 to 0.9908.
1169 L<vmsish> has been upgraded from version 1.03 to 1.04.
1173 L<warnings> has been upgraded from version 1.18 to 1.23.
1177 L<Win32> has been upgraded from version 0.47 to 0.49.
1181 L<XS::Typemap> has been upgraded from version 0.10 to 0.13.
1185 L<XSLoader> has been upgraded from version 0.16 to 0.17.
1189 =head1 Documentation
1191 =head2 New Documentation
1193 =head3 L<perlrepository>
1195 This document was removed (actually, renamed L<perlgit> and given a major
1196 overhaul) in Perl v5.14, causing Perl documentation websites to show the now
1197 out of date version in Perl v5.12 as the latest version. It has now been
1198 restored in stub form, directing readers to current information.
1200 =head2 Changes to Existing Documentation
1208 New sections have been added to document the new index/value array slice and
1209 key/value hash slice syntax.
1213 =head3 L<perldebguts>
1219 The C<DB::goto> and C<DB::lsub> debugger subroutines are now documented. [perl
1224 =head3 L<perlexperiment>
1230 C<\s> matching C<\cK> is marked experimental.
1234 ithreads were accepted in v5.8.0 (but are discouraged as of v5.20.0).
1238 Long doubles are not considered experimental.
1242 Code in regular expressions, regular expression backtracking verbs,
1243 and lvalue subroutines are no longer listed as experimental. (This
1244 also affects L<perlre> and L<perlsub>.)
1254 C<chop> and C<chomp> now note that they can reset the hash iterator.
1258 C<exec>'s handling of arguments is now more clearly documented.
1262 C<eval EXPR> now has caveats about expanding floating point numbers in some
1267 C<goto EXPR> is now documented to handle an expression that evalutes to a
1268 code reference as if it was C<goto &$coderef>. This behavior is at least ten
1273 Since Perl v5.10, it has been possible for subroutines in C<@INC> to return
1274 a reference to a scalar holding initial source code to prepend to the file.
1275 This is now documented.
1279 The documentation of C<ref> has been updated to recommend the use of
1280 C<blessed>, C<isa> and C<reftype> when dealing with references to blessed
1291 Numerous minor changes have been made to reflect changes made to the perl
1292 internals in this release.
1296 New sections on L<Read-Only Values|perlguts/"Read-Only Values"> and
1297 L<Copy on Write|perlguts/"Copy on Write"> have been added.
1307 The L<Super Quick Patch Guide|perlhack/SUPER QUICK PATCH GUIDE> section has
1312 =head3 L<perlhacktips>
1318 The documentation has been updated to include some more examples of C<gdb>
1323 =head3 L<perllexwarn>
1329 The L<perllexwarn> documentation used to describe the hierarchy of warning
1330 categories understood by the L<warnings> pragma. That description has now
1331 been moved to the L<warnings> documentation itself, leaving L<perllexwarn>
1332 as a stub that points to it. This change consolidates all documentation for
1333 lexical warnings in a single place.
1337 =head3 L<perllocale>
1343 The documentation now mentions F<fc()> and C<\F>, and includes many
1344 clarifications and corrections in general.
1354 The language design of Perl has always called for monomorphic operators.
1355 This is now mentioned explicitly.
1359 =head3 L<perlopentut>
1365 The C<open> tutorial has been completely rewritten by Tom Christiansen, and now
1366 focuses on covering only the basics, rather than providing a comprehensive
1367 reference to all things openable. This rewrite came as the result of a
1368 vigorous discussion on perl5-porters kicked off by a set of improvements
1369 written by Alexander Hartmaier to the existing L<perlopentut>. A "more than
1370 you ever wanted to know about C<open>" document may follow in subsequent
1381 The fact that the regexp engine makes no effort to call (?{}) and (??{})
1382 constructs any specified number of times (although it will basically DWIM
1383 in case of a successful match) has been documented.
1387 The C</r> modifier (for non-destructive substitution) is now documented. [perl
1392 The documentation for C</x> and C<(?# comment)> has been expanded and clarified.
1396 =head3 L<perlreguts>
1402 The documentation has been updated in the light of recent changes to
1413 The need to predeclare recursive functions with prototypes in order for the
1414 prototype to be honoured in the recursive call is now documented. [perl #2726]
1418 A list of subroutine names used by the perl implementation is now included.
1429 There is now a L<JavaScript|perltrap/JavaScript Traps> section.
1433 =head3 L<perlunicode>
1439 The documentation has been updated to reflect C<Bidi_Class> changes in
1450 A new section explaining the performance issues of $`, $& and $', including
1451 workarounds and changes in different versions of Perl, has been added.
1455 Three L<English> variable names which have long been documented but do not
1456 actually exist have been removed from the documentation. These were
1457 C<$OLD_PERL_VERSION>, C<$OFMT>, and C<$ARRAY_BASE>.
1467 Several problems in the C<MY_CXT> example have been fixed.
1473 The following additions or changes have been made to diagnostic output,
1474 including warnings and fatal error messages. For the complete list of
1475 diagnostic messages, see L<perldiag>.
1477 =head2 New Diagnostics
1485 L<delete argument is indexE<sol>value array slice, use array slice|perldiag/"delete argument is index/value array slice, use array slice">
1487 (F) You used index/value array slice syntax (C<%array[...]>) as the argument to
1488 C<delete>. You probably meant C<@array[...]> with an @ symbol instead.
1492 L<delete argument is keyE<sol>value hash slice, use hash slice|perldiag/"delete argument is key/value hash slice, use hash slice">
1494 (F) You used key/value hash slice syntax (C<%hash{...}>) as the argument to
1495 C<delete>. You probably meant C<@hash{...}> with an @ symbol instead.
1499 L<Magical list constants are not supported|perldiag/"Magical list constants are
1502 (F) You assigned a magical array to a stash element, and then tried to use the
1503 subroutine from the same slot. You are asking Perl to do something it cannot
1504 do, details subject to change between Perl versions.
1508 Added L<Setting $E<sol> to a %s reference is forbidden|perldiag/"Setting $E<sol> to %s reference is forbidden">
1518 L<%s on reference is experimental|perldiag/"push on reference is experimental">:
1520 The "auto-deref" feature is experimental.
1522 Starting in v5.14.0, it was possible to use push, pop, keys, and other
1523 built-in functions not only on aggregate types, but on references to
1524 them. The feature was not deployed to its original intended
1525 specification, and now may become redundant to postfix dereferencing.
1526 It has always been categorized as an experimental feature, and in
1527 v5.20.0 is carries a warning as such.
1529 Warnings will now be issued at compile time when these operations are
1532 no if $] >= 5.01908, warnings => "experimental::autoderef";
1534 Consider, though, replacing the use of these features, as they may
1535 change behavior again before becoming stable.
1539 L<A sequence of multiple spaces in a charnames alias definition is deprecated|perldiag/"A sequence of multiple spaces in a charnames alias definition is deprecated">
1541 L<Trailing white-space in a charnames alias definition is deprecated|perldiag/"Trailing white-space in a charnames alias definition is deprecated">
1543 These two deprecation warnings involving C<\N{...}> were incorrectly
1544 implemented. They did not warn by default (now they do) and could not be
1545 made fatal via C<< use warnings FATAL => 'deprecated' >> (now they can).
1549 L<Attribute prototype(%s) discards earlier prototype attribute in same sub|perldiag/"Attribute prototype(%s) discards earlier prototype attribute in same sub">
1551 (W misc) A sub was declared as C<sub foo : prototype(A) : prototype(B) {}>, for
1552 example. Since each sub can only have one prototype, the earlier
1553 declaration(s) are discarded while the last one is applied.
1557 L<Invalid \0 character in %s for %s: %s\0%s|perldiag/"Invalid \0 character in %s for %s: %s\0%s">
1559 (W syscalls) Embedded \0 characters in pathnames or other system call arguments
1560 produce a warning as of 5.20. The parts after the \0 were formerly ignored by
1565 L<Matched non-Unicode code point 0x%X against Unicode property; may not be portable|perldiag/"Matched non-Unicode code point 0x%X against Unicode property; may not be portable">.
1567 This replaces the message "Code point 0x%X is not Unicode, all \p{} matches
1568 fail; all \P{} matches succeed".
1572 L<Missing ']' in prototype for %s : %s|perldiag/"Missing ']' in prototype for %s : %s">
1574 (W illegalproto) A grouping was started with C<[> but never closed with C<]>.
1578 L<Possible precedence issue with control flow operator|perldiag/"Possible precedence issue with control flow operator">
1580 (W syntax) There is a possible problem with the mixing of a control flow
1581 operator (e.g. C<return>) and a low-precedence operator like C<or>. Consider:
1583 sub { return $a or $b; }
1587 sub { (return $a) or $b; }
1589 Which is effectively just:
1593 Either use parentheses or the high-precedence variant of the operator.
1595 Note this may be also triggered for constructs like:
1601 L<Postfix dereference is experimental|perldiag/"Postfix dereference is experimental">
1603 (S experimental::postderef) This warning is emitted if you use the experimental
1604 postfix dereference syntax. Simply suppress the warning if you want to use the
1605 feature, but know that in doing so you are taking the risk of using an
1606 experimental feature which may change or be removed in a future Perl version:
1608 no warnings "experimental::postderef";
1609 use feature "postderef", "postderef_qq";
1617 L<Prototype '%s' overridden by attribute 'prototype(%s)' in %s|perldiag/"Prototype '%s' overridden by attribute 'prototype(%s)' in %s">
1619 (W prototype) A prototype was declared in both the parentheses after the sub
1620 name and via the prototype attribute. The prototype in parentheses is useless,
1621 since it will be replaced by the prototype from the attribute before it's ever
1626 L<Scalar value @%s[%s] better written as $%s[%s]|perldiag/"Scalar value @%s[%s] better written as $%s[%s]">
1628 (W syntax) In scalar context, you've used an array index/value slice (indicated
1629 by %) to select a single element of an array. Generally it's better to ask for
1630 a scalar value (indicated by $). The difference is that C<$foo[&bar]> always
1631 behaves like a scalar, both in the value it returns and when evaluating its
1632 argument, while C<%foo[&bar]> provides a list context to its subscript, which
1633 can do weird things if you're expecting only one subscript. When called in
1634 list context, it also returns the index (what C<&bar> returns) in addition to
1639 L<Scalar value @%s{%s} better written as $%s{%s}|perldiag/"Scalar value @%s{%s} better written as $%s{%s}">
1641 (W syntax) In scalar context, you've used a hash key/value slice (indicated by
1642 %) to select a single element of a hash. Generally it's better to ask for a
1643 scalar value (indicated by $). The difference is that C<$foo{&bar}> always
1644 behaves like a scalar, both in the value it returns and when evaluating its
1645 argument, while C<@foo{&bar}> and provides a list context to its subscript,
1646 which can do weird things if you're expecting only one subscript. When called
1647 in list context, it also returns the key in addition to the value.
1651 L<Setting $E<sol> to a reference to %s as a form of slurp is deprecated, treating as undef|perldiag/"Setting $E<sol> to a reference to %s as a form of slurp is deprecated, treating as undef">
1655 L<Unexpected exit %u|perldiag/"Unexpected exit %u">
1657 (S) exit() was called or the script otherwise finished gracefully when
1658 C<PERL_EXIT_WARN> was set in C<PL_exit_flags>.
1662 L<Unexpected exit failure %d|perldiag/"Unexpected exit failure %d">
1664 (S) An uncaught die() was called when C<PERL_EXIT_WARN> was set in
1669 L<Use of literal control characters in variable names is deprecated|perldiag/"Use of literal control characters in variable names is deprecated">
1671 (D deprecated) Using literal control characters in the source to refer to the
1672 ^FOO variables, like $^X and ${^GLOBAL_PHASE} is now deprecated. This only
1673 affects code like $\cT, where \cT is a control (like a C<SOH>) in the
1674 source code: ${"\cT"} and $^T remain valid.
1678 L<Useless use of greediness modifier|perldiag/"Useless use of greediness modifier '%c' in regex; marked by <-- HERE in m/%s/">
1680 This fixes [Perl #42957].
1684 =head2 Changes to Existing Diagnostics
1690 Warnings and errors from the regexp engine are now UTF-8 clean.
1694 The "Unknown switch condition" error message has some slight changes. This
1695 error triggers when there is an unknown condition in a C<(?(foo))> conditional.
1696 The error message used to read:
1698 Unknown switch condition (?(%s in regex;
1700 But what %s could be was mostly up to luck. For C<(?(foobar))>, you might have
1701 seen "fo" or "f". For Unicode characters, you would generally get a corrupted
1702 string. The message has been changed to read:
1704 Unknown switch condition (?(...)) in regex;
1706 Additionally, the C<'E<lt>-- HERE'> marker in the error will now point to the
1707 correct spot in the regex.
1711 The "%s "\x%X" does not map to Unicode" warning is now correctly listed as a
1712 severe warning rather than as a fatal error.
1716 Under rare circumstances, one could get a "Can't coerce readonly REF to
1717 string" instead of the customary "Modification of a read-only value". This
1718 alternate error message has been removed.
1722 "Ambiguous use of * resolved as operator *": This and similar warnings
1723 about "%" and "&" used to occur in some circumstances where there was no
1724 operator of the type cited, so the warning was completely wrong. This has
1725 been fixed [perl #117535, #76910].
1729 Warnings about malformed subroutine prototypes are now more consistent in
1730 how the prototypes are rendered. Some of these warnings would truncate
1731 prototypes containing nulls. In other cases one warning would suppress
1732 another. The warning about illegal characters in prototypes no longer says
1733 "after '_'" if the bad character came before the underscore.
1737 L<Perl folding rules are not up-to-date for 0x%X; please use the perlbug
1738 utility to report; in regex; marked by <-- HERE in
1739 mE<sol>%sE<sol>|perldiag/"Perl folding rules are not up-to-date for 0x%X;
1740 please use the perlbug utility to report; in regex; marked by <-- HERE in
1743 This message is now only in the regexp category, and not in the deprecated
1744 category. It is still a default (i.e., severe) warning [perl #89648].
1748 L<%%s[%s] in scalar context better written as $%s[%s]|perldiag/"%%s[%s] in scalar context better written as $%s[%s]">
1750 This warning now occurs for any C<%array[$index]> or C<%hash{key}> known to
1751 be in scalar context at compile time. Previously it was worded "Scalar
1752 value %%s[%s] better written as $%s[%s]".
1756 L<Switch condition not recognized in regex; marked by <-- HERE in mE<sol>%sE<sol>|perldiag/"Switch condition not recognized in regex; marked by <-- HERE in m/%s/">:
1758 The description for this diagnostic has been extended to cover all cases where the warning may occur.
1759 Issues with the positioning of the arrow indicator have also been resolved.
1763 The error messages for C<my($a?$b$c)> and C<my(do{})> now mention "conditional
1764 expression" and "do block", respectively, instead of reading 'Can't declare
1765 null operation in "my"'.
1769 When C<use re "debug"> executes a regex containing a backreference, the
1770 debugging output now shows what string is being matched.
1774 The now fatal error message C<Character following "\c" must be ASCII> has been
1775 reworded as C<Character following "\c" must be printable ASCII> to emphasize
1776 that in C<\cI<X>>, I<X> must be a I<printable (non-control)> ASCII character.
1780 =head1 Utility Changes
1788 A possible crash from an off-by-one error when trying to access before the
1789 beginning of a buffer has been fixed. [perl #120244]
1795 The git bisection tool F<Porting/bisect.pl> has had many enhancements.
1797 It is provided as part of the source distribution but not installed because
1798 it is not self-contained as it relies on being run from within a git
1799 checkout. Note also that it makes no attempt to fix tests, correct runtime
1800 bugs or make something useful to install - its purpose is to make minimal
1801 changes to get any historical revision of interest to build and run as close
1802 as possible to "as-was", and thereby make C<git bisect> easy to use.
1808 Can optionally run the test case with a timeout.
1812 Can now run in-place in a clean git checkout.
1816 Can run the test case under C<valgrind>.
1820 Can apply user supplied patches and fixes to the source checkout before
1825 Now has fixups to enable building several more historical ranges of bleadperl,
1826 which can be useful for pinpointing the origins of bugs or behaviour changes.
1836 L<find2perl> now handles C<?> wildcards correctly. [perl #113054]
1846 F<perlbug> now has a C<-p> option for attaching patches with a bug report.
1850 L<perlbug> has been modified to supply the report template with CRLF line
1852 [L<perl #121277|https://rt.perl.org/Public/Bug/Display.html?id=121277>]
1856 L<perlbug> now makes as few assumptions as possible about the encoding of the
1857 report. This will likely change in the future to assume UTF-8 by default but
1858 allow a user override.
1862 =head1 Configuration and Compilation
1868 The F<Makefile.PL> for L<SDBM_File> now generates a better F<Makefile>, which
1869 avoids a race condition during parallel makes, which could cause the build to
1870 fail. This is the last known parallel make problem (on *nix platforms), and
1871 therefore we believe that a parallel make should now always be error free.
1875 F<installperl> and F<installman>'s option handling has been refactored to use
1876 L<Getopt::Long>. Both are used by the F<Makefile> C<install> targets, and
1877 are not installed, so these changes are only likely to affect custom
1878 installation scripts.
1884 Single letter options now also have long names.
1888 Invalid options are now rejected.
1892 Command line arguments that are not options are now rejected.
1896 Each now has a C<--help> option to display the usage message.
1900 The behaviour for all valid documented invocations is unchanged.
1904 Where possible, the build now avoids recursive invocations of F<make> when
1905 building pure-Perl extensions, without removing any parallelism from the
1906 build. Currently around 80 extensions can be processed directly by the
1907 F<make_ext.pl> tool, meaning that 80 invocations of F<make> and 160
1908 invocations of F<miniperl> are no longer made.
1912 The build system now works correctly when compiling under GCC or Clang with
1913 link-time optimization enabled (the C<-flto> option). [perl #113022]
1917 Distinct library basenames with C<d_libname_unique>.
1919 When compiling perl with this option, the library files for XS modules are
1920 named something "unique" -- for example, Hash/Util/Util.so becomes
1921 Hash/Util/PL_Hash__Util.so. This behavior is similar to what currently
1922 happens on VMS, and serves as groundwork for the Android port.
1926 C<sysroot> option to indicate the logical root directory under gcc and clang.
1928 When building with this option set, both Configure and the compilers search
1929 for all headers and libraries under this new sysroot, instead of /.
1931 This is a huge time saver if cross-compiling, but can also help
1932 on native builds if your toolchain's files have non-standard locations.
1936 The cross-compilation model has been renovated.
1937 There's several new options, and some backwards-incompatible changes:
1939 We now build binaries for miniperl and generate_uudmap to be used on the host,
1940 rather than running every miniperl call on the target; this means that, short
1941 of 'make test', we no longer need access to the target system once Configure is
1942 done. You can provide already-built binaries through the C<hostperl> and
1943 C<hostgenerate> options to Configure.
1945 Additionally, if targeting an EBCDIC platform from an ASCII host,
1946 or viceversa, you'll need to run Configure with C<-Uhostgenerate>, to
1947 indicate that generate_uudmap should be run on the target.
1949 Finally, there's also a way of having Configure end early, right after
1950 building the host binaries, by cross-compiling without specifying a
1953 The incompatible changes include no longer using xconfig.h, xlib, or
1954 Cross.pm, so canned config files and Makefiles will have to be updated.
1958 Related to the above, there is now a way of specifying the location of sh
1959 (or equivalent) on the target system: C<targetsh>.
1961 For example, Android has its sh in /system/bin/sh, so if cross-compiling
1962 from a more normal Unixy system with sh in /bin/sh, "targetsh" would end
1963 up as /system/bin/sh, and "sh" as /bin/sh.
1967 By default, B<gcc> 4.9 does some optimizations that break perl. The B<-fwrapv>
1968 option disables those optimizations (and probably others), so for B<gcc> 4.3
1969 and later (since the there might be similar problems lurking on older versions
1970 too, but B<-fwrapv> was broken before 4.3, and the optimizations probably won't
1971 go away), F<Configure> now adds B<-fwrapv> unless the user requests
1972 B<-fno-wrapv>, which disables B<-fwrapv>, or B<-fsanitize=undefined>, which
1973 turns the overflows B<-fwrapv> ignores into runtime errors.
1974 [L<perl #121505|https://rt.perl.org/Public/Bug/Display.html?id=121505>]
1984 The C<test.valgrind> make target now allows tests to be run in parallel.
1985 This target allows Perl's test suite to be run under Valgrind, which detects
1986 certain sorts of C programming errors, though at significant cost in running
1987 time. On suitable hardware, allowing parallel execution claws back a lot of
1988 that additional cost. [perl #121431]
1992 Various tests in F<t/porting/> are no longer skipped when the perl
1993 F<.git> directory is outside the perl tree and pointed to by
1994 C<$GIT_DIR>. [perl #120505]
1998 The test suite no longer fails when the user's interactive shell maintains a
1999 C<$PWD> environment variable, but the F</bin/sh> used for running tests
2004 =head1 Platform Support
2006 =head2 New Platforms
2012 Perl can now be built for Android, either natively or through
2013 cross-compilation, for all three currently available architectures (ARM,
2014 MIPS, and x86), on a wide range of versions.
2018 Compile support has been added for Bitrig, a fork of OpenBSD.
2022 Support has been added for FreeMiNT, a free open-source OS for the Atari ST
2023 system and its successors, based on the original MiNT that was officially
2028 Synology ships its NAS boxes with a lean Linux distribution (DSM) on relative
2029 cheap CPU's (like the Marvell Kirkwood mv6282 - ARMv5tel or Freescale QorIQ
2030 P1022 ppc - e500v2) not meant for workstations or development. These boxes
2031 should build now. The basic problems are the non-standard location for tools.
2035 =head2 Discontinued Platforms
2041 Code related to supporting the C<sfio> I/O system has been removed.
2043 Perl 5.004 added support to use the native API of C<sfio>, AT&T's Safe/Fast
2044 I/O library. This code still built with v5.8.0, albeit with many regression
2045 tests failing, but was inadvertently broken before the v5.8.1 release,
2046 meaning that it has not worked on any version of Perl released since then.
2047 In over a decade we have received no bug reports about this, hence it is clear
2048 that no-one is using this functionality on any version of Perl that is still
2049 supported to any degree.
2053 Configure support for the 3b1, also known as the AT&T Unix PC (and the similar
2054 AT&T 7300), has been removed.
2058 DG/UX was a Unix sold by Data General. The last release was in April 2001.
2059 It only runs on Data General's own hardware.
2063 In the absence of a regular source of smoke reports, code intended to support
2064 native EBCDIC platforms will be removed from perl before 5.22.0.
2068 =head2 Platform-Specific Notes
2078 recv() on a connected handle would populate the returned sender
2079 address with whatever happened to be in the working buffer. recv()
2080 now uses a workaround similar to the Win32 recv() wrapper and returns
2081 an empty string when recvfrom(2) doesn't modify the supplied address
2082 length. [perl #118843]
2086 Fixed a build error in cygwin.c on Cygwin 1.7.28.
2088 Tests now handle the errors that occur when C<cygserver> isn't
2095 The BSD compatibility library C<libbsd> is no longer required for builds.
2099 The hints file now looks for C<libgdbm_compat> only if C<libgdbm> itself is
2100 also wanted. The former is never useful without the latter, and in some
2101 circumstances, including it could actually prevent building.
2105 The build system now honors an C<ld> setting supplied by the user running
2110 C<objformat> was removed from version 0.4-RELEASE of MidnightBSD and had been
2111 deprecated on earlier versions. This caused the build environment to be
2112 erroneously configured for C<a.out> rather than C<elf>. This has been now
2115 =item Mixed-endian platforms
2117 The code supporting C<pack> and C<unpack> operations on mixed endian
2118 platforms has been removed. We believe that Perl has long been unable to
2119 build on mixed endian architectures (such as PDP-11s), so we don't think
2120 that this change will affect any platforms which were able to build v5.18.0.
2128 The C<PERL_ENV_TABLES> feature to control the population of %ENV at perl
2129 start-up was broken in Perl 5.16.0 but has now been fixed.
2133 Skip access checks on remotes in opendir(). [perl #121002]
2137 A check for glob metacharacters in a path returned by the
2138 L<C<glob()>|perlfunc/glob> operator has been replaced with a check for VMS
2139 wildcard characters. This saves a significant number of unnecessary
2140 L<C<lstat()>|perlfunc/lstat> calls such that some simple glob operations become
2151 C<rename> and C<link> on Win32 now set $! to ENOSPC and EDQUOT when
2152 appropriate. [perl #119857]
2156 The BUILD_STATIC and ALL_STATIC makefile options for linking some or (nearly)
2157 all extensions statically (into perl520.dll, and into a separate
2158 perl-static.exe too) were broken for MinGW builds. This has now been fixed.
2160 The ALL_STATIC option has also been improved to include the Encode and Win32
2161 extensions (for both VC++ and MinGW builds).
2165 Support for building with Visual C++ 2013 has been added. There are currently
2166 two possible test failures (see L<perlwin32/"Testing Perl on Windows">) which
2167 will hopefully be resolved soon.
2171 Experimental support for building with Intel C++ Compiler has been added. The
2172 nmake makefile (win32/Makefile) and the dmake makefile (win32/makefile.mk) can
2173 be used. A "nmake test" will not pass at this time due to F<cpan/CGI/t/url.t>.
2177 Killing a process tree with L<perlfunc/kill> and a negative signal, was broken
2178 starting in 5.18.0. In this bug, C<kill> always returned 0 for a negative
2179 signal even for valid PIDs, and no processes were terminated. This has been
2180 fixed [perl #121230].
2184 The time taken to build perl on Windows has been reduced quite significantly
2185 (time savings in the region of 30-40% are typically seen) by reducing the
2186 number of, usually failing, I/O calls for each L<C<require()>|perlfunc/require>
2187 (for B<miniperl.exe> only).
2188 [L<perl #121119|https://rt.perl.org/Public/Bug/Display.html?id=121119>]
2192 About 15 minutes of idle sleeping was removed from running C<make test> due to
2193 a bug in which the timeout monitor used for tests could not be cancelled once
2194 the test completes, and the full timeout period elapsed before running the next
2196 [L<perl #121395|https://rt.perl.org/Public/Bug/Display.html?id=121395>]
2200 On a perl built without pseudo-fork (pseudo-fork builds were not affected by
2201 this bug), killing a process tree with L<C<kill()>|perlfunc/kill> and a negative
2202 signal resulted in C<kill()> inverting the returned value. For example, if
2203 C<kill()> killed 1 process tree PID then it returned 0 instead of 1, and if
2204 C<kill()> was passed 2 invalid PIDs then it returned 2 instead of 0. This has
2205 probably been the case since the process tree kill feature was implemented on
2206 Win32. It has now been corrected to follow the documented behaviour.
2207 [L<perl #121230|https://rt.perl.org/Public/Bug/Display.html?id=121230>]
2211 When building a 64-bit perl, an uninitialized memory read in B<miniperl.exe>,
2212 used during the build process, could lead to a 4GB B<wperl.exe> being created.
2213 This has now been fixed. (Note that B<perl.exe> itself was unaffected, but
2214 obviously B<wperl.exe> would have been completely broken.)
2215 [L<perl #121471|https://rt.perl.org/Public/Bug/Display.html?id=121471>]
2219 Perl can now be built with B<gcc> version 4.8.1 from L<http://www.mingw.org>.
2220 This was previously broken due to an incorrect definition of DllMain() in one
2221 of perl's source files. Earlier B<gcc> versions were also affected when using
2222 version 4 of the w32api package. Versions of B<gcc> available from
2223 L<http://mingw-w64.sourceforge.net/> were not affected.
2224 [L<perl #121643|https://rt.perl.org/Public/Bug/Display.html?id=121643>]
2228 The test harness now has no failures when perl is built on a FAT drive with the
2229 Windows OS on an NTFS drive.
2230 [L<perl #21442|https://rt.perl.org/Public/Bug/Display.html?id=21442>]
2234 When cloning the context stack in fork() emulation, Perl_cx_dup()
2235 would crash accessing parameter information for context stack entries
2236 that included no parameters, as with C<&foo;>.
2237 [L<perl #121721|https://rt.perl.org/Public/Bug/Display.html?id=121721>]
2242 L<perl #113536|https://rt.perl.org/Public/Bug/Display.html?id=113536>, a memory
2243 leak on every call to C<system> and backticks (C< `` >), on most Win32 Perls
2244 starting from 5.18.0 has been fixed. The memory leak only occurred if you
2245 enabled psuedo-fork in your build of Win32 Perl, and were running that build on
2246 Server 2003 R2 or newer OS. The leak does not appear on WinXP SP3.
2247 [L<perl #121676|https://rt.perl.org/Public/Bug/Display.html?id=121676>]
2257 The building of XS modules has largely been restored. Several still cannot
2258 (yet) be built but it is now possible to build Perl on WinCE with only a couple
2259 of further patches (to L<Socket> and L<ExtUtils::MakeMaker>), hopefully to be
2264 Perl can now be built in one shot with no user intervention on WinCE by running
2265 C<nmake -f Makefile.ce all>.
2267 Support for building with EVC (Embedded Visual C++) 4 has been restored. Perl
2268 can also be built using Smart Devices for Visual C++ 2005 or 2008.
2274 =head1 Internal Changes
2280 The internal representation has changed for the match variables $1, $2 etc.,
2281 $`, $&, $', ${^PREMATCH}, ${^MATCH} and ${^POSTMATCH}. It uses slightly less
2282 memory, avoids string comparisons and numeric conversions during lookup, and
2283 uses 23 fewer lines of C. This change should not affect any external code.
2287 Arrays now use NULL internally to represent unused slots, instead of
2288 &PL_sv_undef. &PL_sv_undef is no longer treated as a special value, so
2289 av_store(av, 0, &PL_sv_undef) will cause element 0 of that array to hold a
2290 read-only undefined scalar. C<$array[0] = anything> will croak and
2291 C<\$array[0]> will compare equal to C<\undef>.
2295 The SV returned by HeSVKEY_force() now correctly reflects the UTF8ness of the
2296 underlying hash key when that key is not stored as a SV. [perl #79074]
2300 Certain rarely used functions and macros available to XS code are now
2301 deprecated. These are:
2302 C<utf8_to_uvuni_buf> (use C<utf8_to_uvchr_buf> instead),
2303 C<valid_utf8_to_uvuni> (use C<utf8_to_uvchr_buf> instead),
2304 C<NATIVE_TO_NEED> (this did not work properly anyway),
2305 and C<ASCII_TO_NEED> (this did not work properly anyway).
2307 Starting in this release, almost never does application code need to
2308 distinguish between the platform's character set and Latin1, on which the
2309 lowest 256 characters of Unicode are based. New code should not use
2310 C<utf8n_to_uvuni> (use C<utf8_to_uvchr_buf> instead),
2312 C<uvuni_to_utf8> (use C<uvchr_to_utf8> instead),
2316 The Makefile shortcut targets for many rarely (or never) used testing and
2317 profiling targets have been removed, or merged into the only other Makefile
2318 target that uses them. Specifically, these targets are gone, along with
2319 documentation that referenced them or explained how to use them:
2321 check.third check.utf16 check.utf8 coretest minitest.prep
2322 minitest.utf16 perl.config.dashg perl.config.dashpg
2323 perl.config.gcov perl.gcov perl.gprof perl.gprof.config
2324 perl.pixie perl.pixie.atom perl.pixie.config perl.pixie.irix
2325 perl.third perl.third.config perl.valgrind.config purecovperl
2326 pureperl quantperl test.deparse test.taintwarn test.third
2327 test.torture test.utf16 test.utf8 test_notty.deparse
2328 test_notty.third test_notty.valgrind test_prep.third
2329 test_prep.valgrind torturetest ucheck ucheck.third ucheck.utf16
2330 ucheck.valgrind utest utest.third utest.utf16 utest.valgrind
2332 It's still possible to run the relevant commands by "hand" - no underlying
2333 functionality has been removed.
2337 It is now possible to keep Perl from initializing locale handling.
2338 For the most part, Perl doesn't pay attention to locale. (See
2339 L<perllocale>.) Nonetheless, until now, on startup, it has always
2340 initialized locale handling to the system default, just in case the
2341 program being executed ends up using locales. (This is one of the first
2342 things a locale-aware program should do, long before Perl knows if it
2343 will actually be needed or not.) This works well except when Perl is
2344 embedded in another application which wants a locale that isn't the
2345 system default. Now, if the environment variable
2346 C<PERL_SKIP_LOCALE_INIT> is set at the time Perl is started, this
2347 initialization step is skipped. Prior to this, on Windows platforms,
2348 the only workaround for this deficiency was to use a hacked-up copy of
2349 internal Perl code. Applications that need to use older Perls can
2350 discover if the embedded Perl they are using needs the workaround by
2351 testing that the C preprocessor symbol C<HAS_SKIP_LOCALE_INIT> is not
2352 defined. [RT #38193]
2356 C<BmRARE> and C<BmPREVIOUS> have been removed. They were not used anywhere
2357 and are not part of the API. For XS modules, they are now #defined as 0.
2361 C<sv_force_normal>, which usually croaks on read-only values, used to allow
2362 read-only values to be modified at compile time. This has been changed to
2363 croak on read-only values regardless. This change uncovered several core
2368 Perl's new copy-on-write mechanism (which is now enabled by default),
2369 allows any C<SvPOK> scalar to be automatically upgraded to a copy-on-write
2370 scalar when copied. A reference count on the string buffer is stored in
2371 the string buffer itself.
2375 $ perl -MDevel::Peek -e'$a="abc"; $b = $a; Dump $a; Dump $b'
2376 SV = PV(0x260cd80) at 0x2620ad8
2378 FLAGS = (POK,IsCOW,pPOK)
2379 PV = 0x2619bc0 "abc"\0
2383 SV = PV(0x260ce30) at 0x2620b20
2385 FLAGS = (POK,IsCOW,pPOK)
2386 PV = 0x2619bc0 "abc"\0
2391 Note that both scalars share the same PV buffer and have a COW_REFCNT
2394 This means that XS code which wishes to modify the C<SvPVX()> buffer of an
2395 SV should call C<SvPV_force()> or similar first, to ensure a valid (and
2396 unshared) buffer, and to call C<SvSETMAGIC()> afterwards. This in fact has
2397 always been the case (for example hash keys were already copy-on-write);
2398 this change just spreads the COW behaviour to a wider variety of SVs.
2400 One important difference is that before 5.18.0, shared hash-key scalars
2401 used to have the C<SvREADONLY> flag set; this is no longer the case.
2403 This new behaviour can still be disabled by running F<Configure> with
2404 B<-Accflags=-DPERL_NO_COW>. This option will probably be removed in Perl
2409 C<PL_sawampersand> is now a constant. The switch this variable provided
2410 (to enable/disable the pre-match copy depending on whether C<$&> had been
2411 seen) has been removed and replaced with copy-on-write, eliminating a few
2414 The previous behaviour can still be enabled by running F<Configure> with
2415 B<-Accflags=-DPERL_SAWAMPERSAND>.
2419 The functions C<my_swap>, C<my_htonl> and C<my_ntohl> have been removed.
2420 It is unclear why these functions were ever marked as I<A>, part of the
2421 API. XS code can't call them directly, as it can't rely on them being
2422 compiled. Unsurprisingly, no code on CPAN references them.
2426 The signature of the C<Perl_re_intuit_start()> regex function has changed;
2427 the function pointer C<intuit> in the regex engine plugin structure
2428 has also changed accordingly. A new parameter, C<strbeg> has been added;
2429 this has the same meaning as the same-named parameter in
2430 C<Perl_regexec_flags>. Previously intuit would try to guess the start of
2431 the string from the passed SV (if any), and would sometimes get it wrong
2432 (e.g. with an overloaded SV).
2436 The signature of the C<Perl_regexec_flags()> regex function has
2437 changed; the function pointer C<exec> in the regex engine plugin
2438 structure has also changed to match. The C<minend> parameter now has
2439 type C<SSize_t> to better support 64-bit systems.
2443 XS code may use various macros to change the case of a character or code
2444 point (for example C<toLOWER_utf8()>). Only a couple of these were
2445 documented until now;
2446 and now they should be used in preference to calling the underlying
2447 functions. See L<perlapi/Character case changing>.
2451 The code dealt rather inconsistently with uids and gids. Some
2452 places assumed that they could be safely stored in UVs, others
2453 in IVs, others in ints. Four new macros are introduced:
2454 SvUID(), sv_setuid(), SvGID(), and sv_setgid()
2458 C<sv_pos_b2u_flags> has been added to the API. It is similar to C<sv_pos_b2u>,
2459 but supports long strings on 64-bit platforms.
2463 C<PL_exit_flags> can now be used by perl embedders or other XS code to have
2464 perl C<warn> or C<abort> on an attempted exit. [perl #52000]
2468 Compiling with C<-Accflags=-PERL_BOOL_AS_CHAR> now allows C99 and C++
2469 compilers to emulate the aliasing of C<bool> to C<char> that perl does for
2470 C89 compilers. [perl #120314]
2474 The C<sv> argument in L<perlapi/sv_2pv_flags>, L<perlapi/sv_2iv_flags>,
2475 L<perlapi/sv_2uv_flags>, and L<perlapi/sv_2nv_flags> and their older wrappers
2476 sv_2pv, sv_2iv, sv_2uv, sv_2nv, is now non-NULL. Passing NULL now will crash.
2477 When the non-NULL marker was introduced en masse in 5.9.3 the functions
2478 were marked non-NULL, but since the creation of the SV API in 5.0 alpha 2, if
2479 NULL was passed, the functions returned 0 or false-type values. The code that
2480 supports C<sv> argument being non-NULL dates to 5.0 alpha 2 directly, and
2481 indirectly to Perl 1.0 (pre 5.0 api). The lack of documentation that the
2482 functions accepted a NULL C<sv> was corrected in 5.11.0 and between 5.11.0
2483 and 5.19.5 the functions were marked NULLOK. As an optimization the NULLOK code
2484 has now been removed, and the functions became non-NULL marked again, because
2485 core getter-type macros never pass NULL to these functions and would crash
2486 before ever passing NULL.
2488 The only way a NULL C<sv> can be passed to sv_2*v* functions is if XS code
2489 directly calls sv_2*v*. This is unlikely as XS code uses Sv*V* macros to get
2490 the underlying value out of the SV. One possible situation which leads to
2491 a NULL C<sv> being passed to sv_2*v* functions, is if XS code defines its own
2492 getter type Sv*V* macros, which check for NULL B<before> dereferencing and
2493 checking the SV's flags through public API Sv*OK* macros or directly using
2494 private API C<SvFLAGS>, and if C<sv> is NULL, then calling the sv_2*v functions
2495 with a NULL litteral or passing the C<sv> containing a NULL value.
2499 newATTRSUB is now a macro
2501 The public API newATTRSUB was previously a macro to the private
2502 function Perl_newATTRSUB. Function Perl_newATTRSUB has been removed. newATTRSUB
2503 is now macro to a different internal function.
2507 Changes in warnings raised by C<utf8n_to_uvchr()>
2509 This bottom level function decodes the first character of a UTF-8 string
2510 into a code point. It is accessible to C<XS> level code, but it's
2511 discouraged from using it directly. There are higher level functions
2512 that call this that should be used instead, such as
2513 L<perlapi/utf8_to_uvchr_buf>. For completeness though, this documents
2514 some changes to it. Now, tests for malformations are done before any
2515 tests for other potential issues. One of those issues involves code
2516 points so large that they have never appeared in any official standard
2517 (the current standard has scaled back the highest acceptable code point
2518 from earlier versions). It is possible (though not done in CPAN) to
2519 warn and/or forbid these code points, while accepting smaller code
2520 points that are still above the legal Unicode maximum. The warning
2521 message for this now includes the code point if representable on the
2522 machine. Previously it always displayed raw bytes, which is what it
2523 still does for non-representable code points.
2527 Regexp engine changes that affect the pluggable regex engine interface
2529 Many flags that used to be exposed via regexp.h and used to populate the
2530 extflags member of struct regexp have been removed. These fields were
2531 technically private to Perl's own regexp engine and should not have been
2532 exposed there in the first place.
2534 The affected flags are:
2545 As well as the follow flag masks:
2550 All have been renamed to PREGf_ equivalents and moved to regcomp.h.
2552 The behavior previously achieved by setting one or more of the RXf_ANCH_
2553 flags (via the RXf_ANCH mask) have now been replaced by a *single* flag bit
2558 pluggable regex engines which previously used to set these flags should
2559 now set this flag ALONE.
2563 The Perl core now consistently uses C<av_tindex()> ("the top index of an
2564 array") as a more clearly-named synonym for C<av_len()>.
2568 The obscure interpreter variable C<PL_timesbuf> is expected to be removed
2569 early in the 5.21.x development series, so that Perl 5.22.0 will not provide
2570 it to XS authors. While the variable still exists in 5.20.0, we hope that
2571 this advance warning of the deprecation will help anyone who is using that
2576 =head1 Selected Bug Fixes
2578 =head2 Regular Expressions
2584 Fixed a small number of regexp constructions that could either fail to
2585 match or crash perl when the string being matched against was
2586 allocated above the 2GB line on 32-bit systems. [RT #118175]
2590 Various memory leaks involving the parsing of the C<(?[...])> regular
2591 expression construct have been fixed.
2595 C<(?[...])> now allows interpolation of precompiled patterns consisting of
2596 C<(?[...])> with bracketed character classes inside (C<$pat =
2597 S<qr/(?[ [a] ])/;> S</(?[ $pat ])/>>). Formerly, the brackets would
2598 confuse the regular expression parser.
2602 The "Quantifier unexpected on zero-length expression" warning message could
2603 appear twice starting in Perl v5.10 for a regular expression also
2604 containing alternations (e.g., "a|b") triggering the trie optimisation.
2608 Perl v5.18 inadvertently introduced a bug whereby interpolating mixed up-
2609 and down-graded UTF-8 strings in a regex could result in malformed UTF-8
2610 in the pattern: specifically if a downgraded character in the range
2611 C<\x80..\xff> followed a UTF-8 string, e.g.
2613 utf8::upgrade( my $u = "\x{e5}");
2614 utf8::downgrade(my $d = "\x{e5}");
2621 In regular expressions containing multiple code blocks, the values of
2622 C<$1>, C<$2>, etc., set by nested regular expression calls would leak from
2623 one block to the next. Now these variables always refer to the outer
2624 regular expression at the start of an embedded block [perl #117917].
2628 C</$qr/p> was broken in Perl 5.18.0; the C</p> flag was ignored. This has been
2629 fixed. [perl #118213]
2633 Starting in Perl 5.18.0, a construct like C</[#](?{})/x> would have its C<#>
2634 incorrectly interpreted as a comment. The code block would be skipped,
2635 unparsed. This has been corrected.
2639 Starting in Perl 5.001, a regular expression like C</[#$a]/x> or C</[#]$a/x>
2640 would have its C<#> incorrectly interpreted as a comment, so the variable would
2641 not interpolate. This has been corrected. [perl #45667]
2645 Perl 5.18.0 inadvertently made dereferenced regular expressions
2646 S<(C<${ qr// }>)> false as booleans. This has been fixed.
2650 The use of C<\G> in regular expressions, where it's not at the start of the
2651 pattern, is now slightly less buggy (although it is still somewhat
2656 Where a regular expression included code blocks (C</(?{...})/>), and where the
2657 use of constant overloading triggered a re-compilation of the code block, the
2658 second compilation didn't see its outer lexical scope. This was a regression
2663 The string position set by C<pos> could shift if the string changed
2664 representation internally to or from utf8. This could happen, e.g., with
2665 references to objects with string overloading.
2669 Taking references to the return values of two C<pos> calls with the same
2670 argument, and then assigning a reference to one and C<undef> to the other,
2671 could result in assertion failures or memory leaks.
2675 Elements of @- and @+ now update correctly when they refer to non-existent
2676 captures. Previously, a referenced element (C<$ref = \$-[1]>) could refer to
2677 the wrong match after subsequent matches.
2681 The code that parses regex backrefs (or ambiguous backref/octals) such as \123
2682 did a simple atoi(), which could wrap round to negative values on long digit
2683 strings and cause segmentation faults. This has now been fixed. [perl
2688 Assigning another typeglob to C<*^R> no longer makes the regular expression
2693 The C<\N> regular expression escape, when used without the curly braces (to
2694 mean C<[^\n]>), was ignoring a following C<*> if followed by whitespace
2695 under /x. It had been this way since C<\N> to mean C<[^\n]> was introduced
2700 C<s///>, C<tr///> and C<y///> now work when a wide character is used as the
2701 delimiter. [perl #120463]
2705 Some cases of unterminated (?...) sequences in regular expressions (e.g.,
2706 C</(?</>) have been fixed to produce the proper error message instead of
2707 "panic: memory wrap". Other cases (e.g., C</(?(/>) have yet to be fixed.
2711 When a reference to a reference to an overloaded object was returned from
2712 a regular expression C<(??{...})> code block, an incorrect implicit
2713 dereference could take place if the inner reference had been returned by
2714 a code block previously.
2718 A tied variable returned from C<(??{...})> sees the inner values of match
2719 variables (i.e., the $1 etc. from any matches inside the block) in its
2720 FETCH method. This was not the case if a reference to an overloaded object
2721 was the last thing assigned to the tied variable. Instead, the match
2722 variables referred to the outer pattern during the FETCH call.
2726 Fix unexpected tainting via regexp using locale. Previously, under certain
2727 conditions, the use of character classes could cause tainting when it
2728 shouldn't. Some character classes are locale-dependent, but before this
2729 patch, sometimes tainting was happening even for character classes that
2730 don't depend on the locale. [perl #120675]
2734 Under certain conditions, Perl would throw an error if in an lookbehind
2735 assertion in a regexp, the assertion referred to a named subpattern,
2736 complaining the lookbehind was variable when it wasn't. This has been
2737 fixed. [perl #120600], [perl #120618]. The current fix may be improved
2742 C<$^R> wasn't available outside of the regular expression that
2743 initialized it. [perl #121070]
2747 A large set of fixes and refactoring for re_intuit_start() was merged,
2754 Fixed a panic when compiling the regular expression
2755 C</\x{100}[xy]\x{100}{2}/>.
2759 Fixed a performance regression when performing a global pattern match
2760 against a UTF-8 string. [perl #120692]
2764 Fixed another performance issue where matching a regular expression
2765 like C</ab.{1,2}x/> against a long UTF-8 string would unnecessarily
2766 calculate byte offsets for a large portion of the string. [perl
2773 Fixed an alignment error when compiling regular expressions when built
2774 with GCC on HP-UX 64-bit.
2778 On 64-bit platforms C<pos> can now be set to a value higher than 2**31-1.
2783 =head2 Perl 5 Debugger and -d
2789 The debugger's C<man> command been fixed. It was broken in the v5.18.0
2790 release. The C<man> command is aliased to the names C<doc> and C<perldoc> -
2795 C<@_> is now correctly visible in the debugger, fixing a regression
2796 introduced in v5.18.0's debugger. [RT #118169]
2800 Under copy-on-write builds (the default as of 5.20.0) C<< ${'_<-e'}[0] >>
2801 no longer gets mangled. This is the first line of input saved for the
2802 debugger's use for one-liners [perl #118627].
2806 On non-threaded builds, setting C<${"_E<lt>filename"}> to a reference or
2807 typeglob no longer causes C<__FILE__> and some error messages to produce a
2808 corrupt string, and no longer prevents C<#line> directives in string evals from
2809 providing the source lines to the debugger. Threaded builds were unaffected.
2813 Starting with Perl 5.12, line numbers were off by one if the B<-d> switch was
2814 used on the #! line. Now they are correct.
2818 C<*DB::DB = sub {} if 0> no longer stops Perl's debugging mode from finding
2819 C<DB::DB> subs declared thereafter.
2823 C<%{'_<...'}> hashes now set breakpoints on the corresponding C<@{'_<...'}>
2824 rather than whichever array C<@DB::dbline> is aliased to. [perl #119799]
2828 Call set-magic when setting $DB::sub. [perl #121255]
2832 The debugger's "n" command now respects lvalue subroutines and steps over
2833 them [perl #118839].
2837 =head2 Lexical Subroutines
2843 Lexical constants (C<my sub a() { 42 }>) no longer crash when inlined.
2847 Parameter prototypes attached to lexical subroutines are now respected when
2848 compiling sub calls without parentheses. Previously, the prototypes were
2849 honoured only for calls I<with> parentheses. [RT #116735]
2853 Syntax errors in lexical subroutines in combination with calls to the same
2854 subroutines no longer cause crashes at compile time.
2858 Deep recursion warnings no longer crash lexical subroutines. [RT #118521]
2862 The dtrace sub-entry probe now works with lexical subs, instead of
2863 crashing [perl #118305].
2867 Undefining an inlinable lexical subroutine (C<my sub foo() { 42 } undef
2868 &foo>) would result in a crash if warnings were turned on.
2872 An undefined lexical sub used as an inherited method no longer crashes.
2876 The presence of a lexical sub named "CORE" no longer stops the CORE::
2877 prefix from working.
2881 =head2 Everything Else
2887 The OP allocation code now returns correctly aligned memory in all cases
2888 for C<struct pmop>. Previously it could return memory only aligned to a
2889 4-byte boundary, which is not correct for an ithreads build with 64 bit IVs
2890 on some 32 bit platforms. Notably, this caused the build to fail completely
2891 on sparc GNU/Linux. [RT #118055]
2895 Evaluating large hashes in scalar context is now much faster, as the number
2896 of used chains in the hash is now cached for larger hashes. Smaller hashes
2897 continue not to store it and calculate it when needed, as this saves one IV.
2898 That would be 1 IV overhead for every object built from a hash. [RT #114576]
2902 Perl v5.16 inadvertently introduced a bug whereby calls to XSUBs that were
2903 not visible at compile time were treated as lvalues and could be assigned
2904 to, even when the subroutine was not an lvalue sub. This has been fixed.
2909 In Perl v5.18.0 dualvars that had an empty string for the string part but a
2910 non-zero number for the number part starting being treated as true. In
2911 previous versions they were treated as false, the string representation
2912 taking precedeence. The old behaviour has been restored. [RT #118159]
2916 Since Perl v5.12, inlining of constants that override built-in keywords of
2917 the same name had countermanded C<use subs>, causing subsequent mentions of
2918 the constant to use the built-in keyword instead. This has been fixed.
2922 The warning produced by C<-l $handle> now applies to IO refs and globs, not
2923 just to glob refs. That warning is also now UTF8-clean. [RT #117595]
2927 C<delete local $ENV{nonexistent_env_var}> no longer leaks memory.
2931 C<sort> and C<require> followed by a keyword prefixed with C<CORE::> now
2932 treat it as a keyword, and not as a subroutine or module name. [RT #24482]
2936 Through certain conundrums, it is possible to cause the current package to
2937 be freed. Certain operators (C<bless>, C<reset>, C<open>, C<eval>) could
2938 not cope and would crash. They have been made more resilient. [RT #117941]
2942 Aliasing filehandles through glob-to-glob assignment would not update
2943 internal method caches properly if a package of the same name as the
2944 filehandle existed, resulting in filehandle method calls going to the
2945 package instead. This has been fixed.
2949 C<./Configure -de -Dusevendorprefix> didn't default. [RT #64126]
2953 The C<Statement unlikely to be reached> warning was listed in
2954 L<perldiag> as an C<exec>-category warning, but was enabled and disabled
2955 by the C<syntax> category. On the other hand, the C<exec> category
2956 controlled its fatal-ness. It is now entirely handled by the C<exec>
2961 The "Replacement list is longer that search list" warning for C<tr///> and
2962 C<y///> no longer occurs in the presence of the C</c> flag. [RT #118047]
2966 Stringification of NVs are not cached so that the lexical locale controls
2967 stringification of the decimal point. [perl #108378] [perl #115800]
2971 There have been several fixes related to Perl's handling of locales. perl
2972 #38193 was described above in L</Internal Changes>.
2974 #118197, where the radix (decimal point) character had to be an ASCII
2975 character (which doesn't work for some non-Western languages);
2976 and #115808, in which C<POSIX::setlocale()> on failure returned an
2977 C<undef> which didn't warn about not being defined even if those
2978 warnings were enabled.
2982 Compiling a C<split> operator whose third argument is a named constant
2983 evaluating to 0 no longer causes the constant's value to change.
2987 A named constant used as the second argument to C<index> no longer gets
2988 coerced to a string if it is a reference, regular expression, dualvar, etc.
2992 A named constant evaluating to the undefined value used as the second
2993 argument to C<index> no longer produces "uninitialized" warnings at compile
2994 time. It will still produce them at run time.
2998 When a scalar was returned from a subroutine in @INC, the referenced scalar
2999 was magically converted into an IO thingy, possibly resulting in "Bizarre
3000 copy" errors if that scalar continued to be used elsewhere. Now Perl uses
3001 an internal copy of the scalar instead.
3005 Certain uses of the C<sort> operator are optimised to modify an array in
3006 place, such as C<@a = sort @a>. During the sorting, the array is made
3007 read-only. If a sort block should happen to die, then the array remained
3008 read-only even outside the C<sort>. This has been fixed.
3012 C<$a> and C<$b> inside a sort block are aliased to the actual arguments to
3013 C<sort>, so they can be modified through those two variables. This did not
3014 always work, e.g., for lvalue subs and C<$#ary>, and probably many other
3015 operators. It works now.
3019 The arguments to C<sort> are now all in list context. If the C<sort>
3020 itself were called in void or scalar context, then I<some>, but not all, of
3021 the arguments used to be in void or scalar context.
3025 Subroutine prototypes with Unicode characters above U+00FF were getting
3026 mangled during closure cloning. This would happen with subroutines closing
3027 over lexical variables declared outside, and with lexical subs.
3031 C<UNIVERSAL::can> now treats its first argument the same way that method
3032 calls do: Typeglobs and glob references with non-empty IO slots are treated
3033 as handles, and strings are treated as filehandles, rather than packages,
3034 if a handle with that name exists [perl #113932].
3038 Method calls on typeglobs (e.g., C<< *ARGV->getline >>) used to stringify
3039 the typeglob and then look it up again. Combined with changes in Perl
3040 5.18.0, this allowed C<< *foo->bar >> to call methods on the "foo" package
3041 (like C<< foo->bar >>). In some cases it could cause the method to be
3042 called on the wrong handle. Now a typeglob argument is treated as a
3043 handle (just like C<< (\*foo)->bar >>), or, if its IO slot is empty, an
3048 Assigning a vstring to a tied variable or to a subroutine argument aliased
3049 to a nonexistent hash or array element now works, without flattening the
3050 vstring into a regular string.
3054 C<pos>, C<tie>, C<tied> and C<untie> did not work
3055 properly on subroutine arguments aliased to nonexistent
3056 hash and array elements [perl #77814, #27010].
3060 The C<< => >> fat arrow operator can now quote built-in keywords even if it
3061 occurs on the next line, making it consistent with how it treats other
3066 Autovivifying a subroutine stub via C<\&$glob> started causing crashes in Perl
3067 5.18.0 if the $glob was merely a copy of a real glob, i.e., a scalar that had
3068 had a glob assigned to it. This has been fixed. [perl #119051]
3072 Perl used to leak an implementation detail when it came to referencing the
3073 return values of certain operators. C<for ($a+$b) { warn \$_; warn \$_ }> used
3074 to display two different memory addresses, because the C<\> operator was
3075 copying the variable. Under threaded builds, it would also happen for
3076 constants (C<for(1) { ... }>). This has been fixed. [perl #21979, #78194,
3077 #89188, #109746, #114838, #115388]
3081 The range operator C<..> was returning the same modifiable scalars with each
3082 call, unless it was the only thing in a C<foreach> loop header. This meant
3083 that changes to values within the list returned would be visible the next time
3084 the operator was executed. [perl #3105]
3088 Constant folding and subroutine inlining no longer cause operations that would
3089 normally return new modifiable scalars to return read-only values instead.
3093 Closures of the form C<sub () { $some_variable }> are no longer inlined,
3094 causing changes to the variable to be ignored by callers of the subroutine.
3099 Return values of certain operators such as C<ref> would sometimes be shared
3100 between recursive calls to the same subroutine, causing the inner call to
3101 modify the value returned by C<ref> in the outer call. This has been fixed.
3105 C<__PACKAGE__> and constants returning a package name or hash key are now
3106 consistently read-only. In various previous Perl releases, they have become
3107 mutable under certain circumstances.
3111 Enabling "used once" warnings no longer causes crashes on stash circularities
3112 created at compile time (C<*Foo::Bar::Foo:: = *Foo::>).
3116 Undef constants used in hash keys (C<use constant u =E<gt> undef; $h{+u}>) no
3117 longer produce "uninitialized" warnings at compile time.
3121 Modifying a substitution target inside the substitution replacement no longer
3126 The first statement inside a string eval used to use the wrong pragma setting
3127 sometimes during constant folding. C<eval 'uc chr 0xe0'> would randomly choose
3128 between Unicode, byte, and locale semantics. This has been fixed.
3132 The handling of return values of @INC filters (subroutines returned by
3133 subroutines in @INC) has been fixed in various ways. Previously tied variables
3134 were mishandled, and setting $_ to a reference or typeglob could result in
3139 The C<SvPVbyte> XS function has been fixed to work with tied scalars returning
3140 something other than a string. It used to return utf8 in those cases where
3145 Perl 5.18.0 inadvertently made C<--> and C<++> crash on dereferenced regular
3146 expressions, and stopped C<++> from flattening vstrings.
3150 C<bless> no longer dies with "Can't bless non-reference value" if its first
3151 argument is a tied reference.
3155 C<reset> with an argument no longer skips copy-on-write scalars, regular
3156 expressions, typeglob copies, and vstrings. Also, when encountering those or
3157 read-only values, it no longer skips any array or hash with the same name.
3161 C<reset> with an argument now skips scalars aliased to typeglobs
3162 (C<for $z (*foo) { reset "z" }>). Previously it would corrupt memory or crash.
3166 C<ucfirst> and C<lcfirst> were not respecting the bytes pragma. This was a
3167 regression from Perl 5.12. [perl #117355]
3171 Changes to C<UNIVERSAL::DESTROY> now update DESTROY caches in all classes,
3172 instead of causing classes that have already had objects destroyed to continue
3173 using the old sub. This was a regression in Perl 5.18. [perl #114864]
3177 All known false-positive occurrences of the deprecation warning "Useless use of
3178 '\'; doesn't escape metacharacter '%c'", added in Perl 5.18.0, have been
3179 removed. [perl #119101]
3183 The value of $^E is now saved across signal handlers on Windows. [perl #85104]
3187 A lexical filehandle (as in C<open my $fh...>) is usually given a name based on
3188 the current package and the name of the variable, e.g. "main::$fh". Under
3189 recursion, the filehandle was losing the "$fh" part of the name. This has been
3194 Uninitialized values returned by XSUBs are no longer exempt from uninitialized
3195 warnings. [perl #118693]
3199 C<elsif ("")> no longer erroneously produces a warning about void context.
3204 Passing C<undef> to a subroutine now causes @_ to contain the same read-only
3205 undefined scalar that C<undef> returns. Furthermore, C<exists $_[0]> will now
3206 return true if C<undef> was the first argument. [perl #7508, #109726]
3210 Passing a non-existent array element to a subroutine does not usually
3211 autovivify it unless the subroutine modifies its argument. This did not work
3212 correctly with negative indices and with non-existent elements within the
3213 array. The element would be vivified immediately. The delayed vivification
3214 has been extended to work with those. [perl #118691]
3218 Assigning references or globs to the scalar returned by $#foo after the @foo
3219 array has been freed no longer causes assertion failures on debugging builds
3220 and memory leaks on regular builds.
3224 On 64-bit platforms, large ranges like 1..1000000000000 no longer crash, but
3225 eat up all your memory instead. [perl #119161]
3229 C<__DATA__> now puts the C<DATA> handle in the right package, even if the
3230 current package has been renamed through glob assignment.
3234 When C<die>, C<last>, C<next>, C<redo>, C<goto> and C<exit> unwind the scope,
3235 it is possible for C<DESTROY> recursively to call a subroutine or format that
3236 is currently being exited. It that case, sometimes the lexical variables
3237 inside the sub would start out having values from the outer call, instead of
3238 being undefined as they should. This has been fixed. [perl #119311]
3242 ${^MPEN} is no longer treated as a synonym for ${^MATCH}.
3246 Perl now tries a little harder to return the correct line number in
3247 C<(caller)[2]>. [perl #115768]
3251 Line numbers inside multiline quote-like operators are now reported correctly.
3256 C<#line> directives inside code embedded in quote-like operators are now
3261 Line numbers are now correct inside the second here-doc when two here-doc
3262 markers occur on the same line.
3266 An optimization in Perl 5.18 made incorrect assumptions causing a bad
3267 interaction with the L<Devel::CallParser> CPAN module. If the module was
3268 loaded then lexical variables declared in separate statements following a
3269 C<my(...)> list might fail to be cleared on scope exit.
3273 C<&xsub> and C<goto &xsub> calls now allow the called subroutine to autovivify
3278 C<&xsub> and C<goto &xsub> no longer crash if *_ has been undefined and has no
3279 ARRAY entry (i.e. @_ does not exist).
3283 C<&xsub> and C<goto &xsub> now work with tied @_.
3287 Overlong identifiers no longer cause a buffer overflow (and a crash). They
3288 started doing so in Perl 5.18.
3292 The warning "Scalar value @hash{foo} better written as $hash{foo}" now produces
3293 far fewer false positives. In particular, C<@hash{+function_returning_a_list}>
3294 and C<@hash{ qw "foo bar baz" }> no longer warn. The same applies to array
3295 slices. [perl #28380, #114024]
3299 C<$! = EINVAL; waitpid(0, WNOHANG);> no longer goes into an internal infinite
3304 A possible segmentation fault in filehandle duplication has been fixed.
3308 A subroutine in @INC can return a reference to a scalar containing the initial
3309 contents of the file. However, that scalar was freed prematurely if not
3310 referenced elsewhere, giving random results.
3314 C<last> no longer returns values that the same statement has accumulated so
3315 far, fixing amongst other things the long-standing bug that C<push @a, last>
3316 would try to return the @a, copying it like a scalar in the process and
3317 resulting in the error, "Bizarre copy of ARRAY in last." [perl #3112]
3321 In some cases, closing file handles opened to pipe to or from a process, which
3322 had been duplicated into a standard handle, would call perl's internal waitpid
3323 wrapper with a pid of zero. With the fix for [perl #85228] this zero pid was
3324 passed to C<waitpid>, possibly blocking the process. This wait for process
3325 zero no longer occurs. [perl #119893]
3329 C<select> used to ignore magic on the fourth (timeout) argument, leading to
3330 effects such as C<select> blocking indefinitely rather than the expected sleep
3331 time. This has now been fixed. [perl #120102]
3335 The class name in C<for my class $foo> is now parsed correctly. In the case of
3336 the second character of the class name being followed by a digit (e.g. 'a1b')
3337 this used to give the error "Missing $ on loop variable". [perl #120112]
3341 Perl 5.18.0 accidentally disallowed C<-bareword> under C<use strict> and
3342 C<use integer>. This has been fixed. [perl #120288]
3346 C<-a> at the start of a line (or a hyphen with any single letter that is
3347 not a filetest operator) no longer produces an erroneous 'Use of "-a"
3348 without parentheses is ambiguous' warning. [perl #120288]
3352 Lvalue context is now properly propagated into bare blocks and C<if> and
3353 C<else> blocks in lvalue subroutines. Previously, arrays and hashes would
3354 sometimes incorrectly be flattened when returned in lvalue list context, or
3355 "Bizarre copy" errors could occur. [perl #119797]
3359 Lvalue context is now propagated to the branches of C<||> and C<&&> (and
3360 their alphabetic equivalents, C<or> and C<and>). This means
3361 C<foreach (pos $x || pos $y) {...}> now allows C<pos> to be modified
3366 C<stat> and C<readline> remember the last handle used; the former
3367 for the special C<_> filehandle, the latter for C<${^LAST_FH}>.
3368 C<eval "*foo if 0"> where *foo was the last handle passed to C<stat>
3369 or C<readline> could cause that handle to be forgotten if the
3370 handle were not opened yet. This has been fixed.
3374 Various cases of C<delete $::{a}>, C<delete $::{ENV}> etc. causing a crash
3375 have been fixed. [perl #54044]
3379 Setting C<$!> to EACCESS before calling C<require> could affect
3380 C<require>'s behaviour. This has been fixed.
3384 The "Can't use \1 to mean $1 in expression" warning message now only occurs
3385 on the right-hand (replacement) part of a substitution. Formerly it could
3386 happen in code embedded in the left-hand side, or in any other quote-like
3391 Blessing into a reference (C<bless $thisref, $thatref>) has long been
3392 disallowed, but magical scalars for the second like C<$/> and those tied
3393 were exempt. They no longer are. [perl #119809]
3397 Blessing into a reference was accidentally allowed in 5.18 if the class
3398 argument were a blessed reference with stale method caches (i.e., whose
3399 class had had subs defined since the last method call). They are
3400 disallowed once more, as in 5.16.
3404 C<< $x->{key} >> where $x was declared as C<my Class $x> no longer crashes
3405 if a Class::FIELDS subroutine stub has been declared.
3409 C<@$obj{'key'}> and C<${$obj}{key}> used to be exempt from compile-time
3410 field checking ("No such class field"; see L<fields>) but no longer are.
3414 A nonexistent array element with a large index passed to a subroutine that
3415 ties the array and then tries to access the element no longer results in a
3420 Declaring a subroutine stub named NEGATIVE_INDICES no longer makes negative
3421 array indices crash when the current package is a tied array class.
3425 Declaring a C<require>, C<glob>, or C<do> subroutine stub in the
3426 CORE::GLOBAL:: package no longer makes compilation of calls to the
3427 corresponding functions crash.
3431 Aliasing CORE::GLOBAL:: functions to constants stopped working in Perl 5.10
3432 but has now been fixed.
3436 When C<`...`> or C<qx/.../> calls a C<readpipe> override, double-quotish
3437 interpolation now happens, as is the case when there is no override.
3438 Previously, the presence of an override would make these quote-like
3439 operators act like C<q{}>, suppressing interpolation. [perl #115330]
3443 C<<<<`...`> here-docs (with backticks as the delimiters) now call
3444 C<readpipe> overrides. [perl #119827]
3448 C<&CORE::exit()> and C<&CORE::die()> now respect L<vmsish> hints.
3452 Undefining a glob that triggers a DESTROY method that undefines the same
3453 glob is now safe. It used to produce "Attempt to free unreferenced glob
3454 pointer" warnings and leak memory.
3458 If subroutine redefinition (C<eval 'sub foo{}'> or C<newXS> for XS code)
3459 triggers a DESTROY method on the sub that is being redefined, and that
3460 method assigns a subroutine to the same slot (C<*foo = sub {}>), C<$_[0]>
3461 is no longer left pointing to a freed scalar. Now DESTROY is delayed until
3462 the new subroutine has been installed.
3466 On Windows, perl no longer calls CloseHandle() on a socket handle. This makes
3467 debugging easier on Windows by removing certain irrelevant bad handle
3468 exceptions. It also fixes a race condition that made socket functions randomly
3469 fail in a Perl process with multiple OS threads, and possible test failures in
3470 F<dist/IO/t/cachepropagate-tcp.t>. [perl #120091/118059]
3474 Formats involving UTF-8 encoded strings, or strange vars like ties,
3475 overloads, or stringified refs (and in recent
3476 perls, pure NOK vars) would generally do the wrong thing in formats
3477 when the var is treated as a string and repeatedly chopped, as in
3478 C<< ^<<<~~ >> and similar. This has now been resolved.
3479 [perl #33832/45325/113868/119847/119849/119851]
3483 C<< semctl(..., SETVAL, ...) >> would set the semaphore to the top
3484 32-bits of the supplied integer instead of the bottom 32-bits on
3485 64-bit big-endian systems. [perl #120635]
3489 C<< readdir() >> now only sets C<$!> on error. C<$!> is no longer set
3490 to C<EBADF> when then terminating C<undef> is read from the directory
3491 unless the system call sets C<$!>. [perl #118651]
3495 C<&CORE::glob> no longer causes an intermittent crash due to perl's stack
3496 getting corrupted. [perl #119993]
3500 C<open> with layers that load modules (e.g., "<:encoding(utf8)") no longer
3501 runs the risk of crashing due to stack corruption.
3505 Perl 5.18 broke autoloading via C<< ->SUPER::foo >> method calls by looking
3506 up AUTOLOAD from the current package rather than the current package's
3507 superclass. This has been fixed. [perl #120694]
3511 A longstanding bug causing C<do {} until CONSTANT>, where the constant
3512 holds a true value, to read unallocated memory has been resolved. This
3513 would usually happen after a syntax error. In past versions of Perl it has
3514 crashed intermittently. [perl #72406]
3518 Fix HP-UX C<$!> failure. HP-UX strerror() returns an empty string for an
3519 unknown error code. This caused an assertion to fail under DEBUGGING
3520 builds. Now instead, the returned string for C<"$!"> contains text
3521 indicating the code is for an unknown error.
3525 Individually-tied elements of @INC (as in C<tie $INC[0]...>) are now
3526 handled correctly. Formerly, whether a sub returned by such a tied element
3527 would be treated as a sub depended on whether a FETCH had occurred
3532 C<getc> on a byte-sized handle after the same C<getc> operator had been
3533 used on a utf8 handle used to treat the bytes as utf8, resulting in erratic
3534 behavior (e.g., malformed UTF-8 warnings).
3538 An initial C<{> at the beginning of a format argument line was always
3539 interpreted as the beginning of a block prior to v5.18. In Perl v5.18, it
3540 started being treated as an ambiguous token. The parser would guess
3541 whether it was supposed to be an anonymous hash constructor or a block
3542 based on the contents. Now the previous behavious has been restored.
3547 In Perl v5.18 C<undef *_; goto &sub> and C<local *_; goto &sub> started
3548 crashing. This has been fixed. [perl #119949]
3552 Backticks (C< `` > or C< qx// >) combined with multiple threads on
3553 Win32 could result in output sent to stdout on one thread being
3554 captured by backticks of an external command in another thread.
3556 This could occur for pseudo-forked processes too, as Win32's
3557 pseudo-fork is implemented in terms of threads. [perl #77672]
3561 C<< open $fh, ">+", undef >> no longer leaks memory when TMPDIR is set
3562 but points to a directory a temporary file cannot be created in. [perl
3567 C< for ( $h{k} || '' ) > no longer auto-vivifies C<$h{k}>. [perl
3572 On Windows machines, Perl now emulates the POSIX use of the environment
3573 for locale initialization. Previously, the environment was ignored.
3574 See L<perllocale/ENVIRONMENT>.
3578 Fixed a crash when destroying a self-referencing GLOB. [perl #121242]
3582 =head1 Known Problems
3588 L<IO::Socket> is known to fail tests on AIX 5.3. There is
3589 L<a patch|https://rt.perl.org/Ticket/Display.html?id=120835> in the request
3590 tracker, #120835, which may be applied to future releases.
3594 The following modules are known to have test failures with this version of
3595 Perl. Patches have been submitted, so there will hopefully be new releases
3602 L<Data::Structure::Util> version 0.15
3606 L<HTML::StripScripts> version 1.05
3610 L<List::Gather> version 0.08.
3618 Diana Rosa, 27, of Rio de Janeiro, went to her long rest on May 10,
3619 2014, along with the plush camel she kept hanging on her computer screen
3620 all the time. She was a passionate Perl hacker who loved the language and its
3621 community, and who never missed a Rio.pm event. She was a true artist, an
3622 enthusiast about writing code, singing arias and graffiting walls. We'll never
3625 Greg McCarroll died on August 28, 2013.
3627 Greg was well known for many good reasons. He was one of the organisers of
3628 the first YAPC::Europe, which concluded with an unscheduled auction where he
3629 frantically tried to raise extra money to avoid the conference making a
3630 loss. It was Greg who mistakenly arrived for a london.pm meeting a week
3631 late; some years later he was the one who sold the choice of official
3632 meeting date at a YAPC::Europe auction, and eventually as glorious leader of
3633 london.pm he got to inherit the irreverent confusion that he had created.
3635 Always helpful, friendly and cheerfully optimistic, you will be missed, but
3638 =head1 Acknowledgements
3640 Perl 5.20.0 represents approximately 12 months of development since Perl 5.18.0
3641 and contains approximately 470,000 lines of changes across 2,900 files from 124
3644 Excluding auto-generated files, documentation and release tools, there were
3645 approximately 280,000 lines of changes to 1,800 .pm, .t, .c and .h files.
3647 Perl continues to flourish into its third decade thanks to a vibrant community
3648 of users and developers. The following people are known to have contributed the
3649 improvements that became Perl 5.20.0:
3651 Aaron Crane, Abhijit Menon-Sen, Abigail, Abir Viqar, Alan Haggai Alavi, Alan
3652 Hourihane, Alexander Voronov, Alexandr Ciornii, Andy Dougherty, Anno Siegel,
3653 Aristotle Pagaltzis, Arthur Axel 'fREW' Schmidt, Brad Gilbert, Brendan Byrd,
3654 Brian Childs, Brian Fraser, Brian Gottreu, Chris 'BinGOs' Williams, Christian
3655 Millour, Colin Kuskie, Craig A. Berry, Dabrien 'Dabe' Murphy, Dagfinn Ilmari
3656 Mannsåker, Daniel Dragan, Darin McBride, David Golden, David Leadbeater, David
3657 Mitchell, David Nicol, David Steinbrunner, Dennis Kaarsemaker, Dominic
3658 Hargreaves, Ed Avis, Eric Brine, Evan Zacks, Father Chrysostomos, Florian
3659 Ragwitz, François Perrad, Gavin Shelley, Gideon Israel Dsouza, Gisle Aas,
3660 Graham Knop, H.Merijn Brand, Hauke D, Heiko Eissfeldt, Hiroo Hayashi, Hojung
3661 Youn, James E Keenan, Jarkko Hietaniemi, Jerry D. Hedden, Jess Robinson, Jesse
3662 Luehrs, Johan Vromans, John Gardiner Myers, John Goodyear, John P. Linderman,
3663 John Peacock, kafka, Kang-min Liu, Karen Etheridge, Karl Williamson, Keedi Kim,
3664 Kent Fredric, kevin dawson, Kevin Falcone, Kevin Ryde, Leon Timmermans, Lukas
3665 Mai, Marc Simpson, Marcel Grünauer, Marco Peereboom, Marcus Holland-Moritz,
3666 Mark Jason Dominus, Martin McGrath, Matthew Horsfall, Max Maischein, Mike
3667 Doherty, Moritz Lenz, Nathan Glenn, Nathan Trapuzzano, Neil Bowers, Neil
3668 Williams, Nicholas Clark, Niels Thykier, Niko Tyni, Olivier Mengué, Owain G.
3669 Ainsworth, Paul Green, Paul Johnson, Peter John Acklam, Peter Martini, Peter
3670 Rabbitson, Petr Písař, Philip Boulain, Philip Guenther, Piotr Roszatycki,
3671 Rafael Garcia-Suarez, Reini Urban, Reuben Thomas, Ricardo Signes, Ruslan
3672 Zakirov, Sergey Alekseev, Shirakata Kentaro, Shlomi Fish, Slaven Rezic,
3673 Smylers, Steffen Müller, Steve Hay, Sullivan Beck, Thomas Sibley, Tobias
3674 Leich, Toby Inkster, Tokuhiro Matsuno, Tom Christiansen, Tom Hukins, Tony Cook,
3675 Victor Efimov, Viktor Turskyi, Vladimir Timofeev, YAMASHINA Hio, Yves Orton,
3676 Zefram, Zsbán Ambrus, Ævar Arnfjörð Bjarmason.
3678 The list above is almost certainly incomplete as it is automatically generated
3679 from version control history. In particular, it does not include the names of
3680 the (very much appreciated) contributors who reported issues to the Perl bug
3683 Many of the changes included in this version originated in the CPAN modules
3684 included in Perl's core. We're grateful to the entire CPAN community for
3685 helping Perl to flourish.
3687 For a more complete list of all of Perl's historical contributors, please see
3688 the F<AUTHORS> file in the Perl source distribution.
3690 =head1 Reporting Bugs
3692 If you find what you think is a bug, you might check the articles recently
3693 posted to the comp.lang.perl.misc newsgroup and the perl bug database at
3694 http://rt.perl.org/perlbug/ . There may also be information at
3695 http://www.perl.org/ , the Perl Home Page.
3697 If you believe you have an unreported bug, please run the L<perlbug> program
3698 included with your release. Be sure to trim your bug down to a tiny but
3699 sufficient test case. Your bug report, along with the output of C<perl -V>,
3700 will be sent off to perlbug@perl.org to be analysed by the Perl porting team.
3702 If the bug you are reporting has security implications, which make it
3703 inappropriate to send to a publicly archived mailing list, then please send it
3704 to perl5-security-report@perl.org. This points to a closed subscription
3705 unarchived mailing list, which includes all the core committers, who will be
3706 able to help assess the impact of issues, figure out a resolution, and help
3707 co-ordinate the release of patches to mitigate or fix the problem across all
3708 platforms on which Perl is supported. Please only use this address for
3709 security issues in the Perl core, not for modules independently distributed on
3714 The F<Changes> file for an explanation of how to view exhaustive details on
3717 The F<INSTALL> file for how to build Perl.
3719 The F<README> file for general stuff.
3721 The F<Artistic> and F<Copying> files for copyright information.