5 perl5110delta - what is new for perl v5.11.0
9 This document describes differences between the 5.10.0 release and
10 the 5.11.0 development release.
12 =head1 Incompatible Changes
14 =head2 Unicode interpretation of \w, \d, \s, and the POSIX character classes redefined.
16 Previous versions of Perl tried to map POSIX style character class definitions onto
17 Unicode property names so that patterns would "dwim" when matches were made against latin-1 or
18 unicode strings. This proved to be a mistake, breaking character class negation, causing
19 forward compatibility problems (as Unicode keeps updating their property definitions and adding
20 new characters), and other problems.
22 Therefore we have now defined a new set of artificial "unicode" property names which will be
23 used to do unicode matching of patterns using POSIX style character classes and perl short-form
24 escape character classes like \w and \d.
26 The key change here is that \d will no longer match every digit in the unicode standard
27 (there are thousands) nor will \w match every word character in the standard, instead they
28 will match precisely their POSIX or Perl definition.
30 Those needing to match based on Unicode properties can continue to do so by using the \p{} syntax
31 to match whichever property they like, including the new artificial definitions.
33 B<NOTE:> This is a backwards incompatible no-warning change in behaviour. If you are upgrading
34 and you process large volumes of text look for POSIX and Perl style character classes and
35 change them to the relevent property name (by removing the word 'Posix' from the current name).
37 The following table maps the POSIX character class names, the escapes and the old and new
38 Unicode property mappings:
40 POSIX Esc Class New-Property ! Old-Property
41 ----------------------------------------------+-------------
42 alnum [0-9A-Za-z] IsPosixAlnum ! IsAlnum
43 alpha [A-Za-z] IsPosixAlpha ! IsAlpha
44 ascii [\000-\177] IsASCII = IsASCII
45 blank [\011 ] IsPosixBlank !
46 cntrl [\0-\37\177] IsPosixCntrl ! IsCntrl
47 digit \d [0-9] IsPosixDigit ! IsDigit
48 graph [!-~] IsPosixGraph ! IsGraph
49 lower [a-z] IsPosixLower ! IsLower
50 print [ -~] IsPosixPrint ! IsPrint
51 punct [!-/:-@[-`{-~] IsPosixPunct ! IsPunct
52 space [\11-\15 ] IsPosixSpace ! IsSpace
53 \s [\11\12\14\15 ] IsPerlSpace ! IsSpacePerl
54 upper [A-Z] IsPosixUpper ! IsUpper
55 word \w [0-9A-Z_a-z] IsPerlWord ! IsWord
56 xdigit [0-9A-Fa-f] IsXDigit = IsXDigit
58 If you wish to build perl with the old mapping you may do so by setting
60 #define PERL_LEGACY_UNICODE_CHARCLASS_MAPPINGS 1
62 in regcomp.h, and then setting
64 PERL_TEST_LEGACY_POSIX_CC
66 to true your enviornment when testing.
69 =head2 In @INC, move ARCHLIB and PRIVLIB after the current version's site_perl and vendor_perl.
71 =head2 Switch statement changes
73 The handling of complex expressions by the C<given>/C<when> switch
74 statement has been enhanced. These enhancements are also available in
75 5.10.1 and subsequent 5.10 releases. There are two new cases where C<when> now
76 interprets its argument as a boolean, instead of an expression to be used
81 =item flip-flop operators
83 The C<..> and C<...> flip-flop operators are now evaluated in boolean
84 context, following their usual semantics; see L<perlop/"Range Operators">.
86 Note that, as in perl 5.10.0, C<when (1..10)> will not work to test
87 whether a given value is an integer between 1 and 10; you should use
88 C<when ([1..10])> instead (note the array reference).
90 However, contrary to 5.10.0, evaluating the flip-flop operators in boolean
91 context ensures it can now be useful in a C<when()>, notably for
92 implementing bistable conditions, like in:
94 when (/^=begin/ .. /^=end/) {
98 =item defined-or operator
100 A compound expression involving the defined-or operator, as in
101 C<when (expr1 // expr2)>, will be treated as boolean if the first
102 expression is boolean. (This just extends the existing rule that applies
103 to the regular or operator, as in C<when (expr1 || expr2)>.)
107 The next section details more changes brought to the semantics to
108 the smart match operator, that naturally also modify the behaviour
109 of the switch statements where smart matching is implicitly used.
110 These changers were also made for the 5.10.1 release, and will remain in
111 subsequent 5.10 releases.
113 =head2 Smart match changes
115 =head3 Changes to type-based dispatch
117 The smart match operator C<~~> is no longer commutative. The behaviour of
118 a smart match now depends primarily on the type of its right hand
119 argument. Moreover, its semantics have been adjusted for greater
120 consistency or usefulness in several cases. While the general backwards
121 compatibility is maintained, several changes must be noted:
127 Code references with an empty prototype are no longer treated specially.
128 They are passed an argument like the other code references (even if they
129 choose to ignore it).
133 C<%hash ~~ sub {}> and C<@array ~~ sub {}> now test that the subroutine
134 returns a true value for each key of the hash (or element of the
135 array), instead of passing the whole hash or array as a reference to
140 Due to the commutativity breakage, code references are no longer
141 treated specially when appearing on the left of the C<~~> operator,
142 but like any vulgar scalar.
146 C<undef ~~ %hash> is always false (since C<undef> can't be a key in a
147 hash). No implicit conversion to C<""> is done (as was the case in perl
152 C<$scalar ~~ @array> now always distributes the smart match across the
153 elements of the array. It's true if one element in @array verifies
154 C<$scalar ~~ $element>. This is a generalization of the old behaviour
155 that tested whether the array contained the scalar.
159 The full dispatch table for the smart match operator is given in
160 L<perlsyn/"Smart matching in detail">.
162 =head3 Smart match and overloading
164 According to the rule of dispatch based on the rightmost argument type,
165 when an object overloading C<~~> appears on the right side of the
166 operator, the overload routine will always be called (with a 3rd argument
167 set to a true value, see L<overload>.) However, when the object will
168 appear on the left, the overload routine will be called only when the
169 rightmost argument is a simple scalar. This way distributivity of smart match
170 across arrays is not broken, as well as the other behaviours with complex
171 types (coderefs, hashes, regexes). Thus, writers of overloading routines
172 for smart match mostly need to worry only with comparing against a scalar,
173 and possibly with stringification overloading; the other common cases
174 will be automatically handled consistently.
176 C<~~> will now refuse to work on objects that do not overload it (in order
177 to avoid relying on the object's underlying structure). (However, if the
178 object overloads the stringification or the numification operators, and
179 if overload fallback is active, it will be used instead, as usual.)
181 =head2 Labels can't be keywords
183 Labels used as targets for the C<goto>, C<last>, C<next> or C<redo>
184 statements cannot be keywords anymore. This restriction will prevent
185 potential confusion between the C<goto LABEL> and C<goto EXPR> syntaxes:
186 for example, a statement like C<goto print> would jump to a label whose
187 name would be the return value of print(), (usually 1), instead of a
188 label named C<print>. Moreover, the other control flow statements
189 would just ignore any keyword passed to them as a label name. Since
190 such labels cannot be defined anymore, this kind of error will be
193 =head2 Other incompatible changes
199 The semantics of C<use feature :5.10*> have changed slightly.
200 See L<"Modules and Pragmata"> for more information.
204 It is now a run-time error to use the smart match operator C<~~>
205 with an object that has no overload defined for it. (This way
206 C<~~> will not break encapsulation by matching against the
207 object's internal representation as a reference.)
211 The version control system used for the development of the perl
212 interpreter has been switched from Perforce to git. This is mainly an
213 internal issue that only affects people actively working on the perl core;
214 but it may have minor external visibility, for example in some of details
215 of the output of C<perl -V>. See L<perlrepository> for more information.
219 The internal structure of the C<ext/> directory in the perl source has
220 been reorganised. In general, a module C<Foo::Bar> whose source was
221 stored under F<ext/Foo/Bar/> is now located under F<ext/Foo-Bar/>. Also,
222 some modules have been moved from F<lib/> to F<ext/>. This is purely a
223 source tarball change, and should make no difference to the compilation or
224 installation of perl, unless you have a very customised build process that
225 explicitly relies on this structure, or which hard-codes the C<nonxs_ext>
226 F<Configure> parameter. Specifically, this change does not by default
227 alter the location of any files in the final installation.
231 As part of the C<Test::Harness> 2.x to 3.x upgrade, the experimental
232 C<Test::Harness::Straps> module has been removed.
233 See L</"Updated Modules"> for more details.
237 As part of the C<ExtUtils::MakeMaker> upgrade, the
238 C<ExtUtils::MakeMaker::bytes> and C<ExtUtils::MakeMaker::vmsish> modules
239 have been removed from this distribution.
243 C<Module::CoreList> no longer contains the C<%:patchlevel> hash.
247 This one is actually a change introduced in 5.10.0, but it was missed
248 from that release's perldelta, so it is mentioned here instead.
250 A bugfix related to the handling of the C</m> modifier and C<qr> resulted
251 in a change of behaviour between 5.8.x and 5.10.0:
253 # matches in 5.8.x, doesn't match in 5.10.0
254 $re = qr/^bar/; "foo\nbar" =~ /$re/m;
258 C<length undef> now returns undef.
262 =head1 Core Enhancements
264 =head2 Unicode Character Database 5.1.0
266 The copy of the Unicode Character Database included in Perl 5.11.0 has
267 been updated to 5.1.0 from 5.0.0. See
268 L<http://www.unicode.org/versions/Unicode5.1.0/#Notable_Changes> for the
271 =head2 A proper interface for pluggable Method Resolution Orders
273 As of Perl 5.11.0 there is a new interface for plugging and using method
274 resolution orders other than the default (linear depth first search).
275 The C3 method resolution order added in 5.10.0 has been re-implemented as
276 a plugin, without changing its Perl-space interface. See L<perlmroapi> for
279 =head2 The C<overloading> pragma
281 This pragma allows you to lexically disable or enable overloading
282 for some or all operations. (Yuval Kogman)
284 =head2 C<\N> regex escape
286 A new regex escape has been added, C<\N>. It will match any character that
287 is not a newline, independently from the presence or absence of the single
288 line match modifier C</s>. (If C<\N> is followed by an opening brace and
289 by a letter, perl will still assume that a Unicode character name is
290 coming, so compatibility is preserved.) (Rafael Garcia-Suarez)
292 =head2 Implicit strictures
294 Using the C<use VERSION> syntax with a version number greater or equal
295 to 5.11.0 will also lexically enable strictures just like C<use strict>
296 would do (in addition to enabling features.) So, the following:
305 =head2 Parallel tests
307 The core distribution can now run its regression tests in parallel on
308 Unix-like platforms. Instead of running C<make test>, set C<TEST_JOBS> in
309 your environment to the number of tests to run in parallel, and run
310 C<make test_harness>. On a Bourne-like shell, this can be done as
312 TEST_JOBS=3 make test_harness # Run 3 tests in parallel
314 An environment variable is used, rather than parallel make itself, because
315 L<TAP::Harness> needs to be able to schedule individual non-conflicting test
316 scripts itself, and there is no standard interface to C<make> utilities to
317 interact with their job schedulers.
319 Note that currently some test scripts may fail when run in parallel (most
320 notably C<ext/IO/t/io_dir.t>). If necessary run just the failing scripts
321 again sequentially and see if the failures go away.
323 =head2 The C<...> operator
325 A new operator, C<...>, nicknamed the Yada Yada operator, has been added.
326 It is intended to mark placeholder code, that is not yet implemented.
327 See L<perlop/"Yada Yada Operator">. (chromatic)
329 =head2 DTrace support
331 Some support for DTrace has been added. See "DTrace support" in F<INSTALL>.
333 =head2 Support for C<configure_requires> in CPAN module metadata
335 Both C<CPAN> and C<CPANPLUS> now support the C<configure_requires> keyword
336 in the F<META.yml> metadata file included in most recent CPAN distributions.
337 This allows distribution authors to specify configuration prerequisites that
338 must be installed before running F<Makefile.PL> or F<Build.PL>.
340 See the documentation for C<ExtUtils::MakeMaker> or C<Module::Build> for more
341 on how to specify C<configure_requires> when creating a distribution for CPAN.
343 =head2 The C<each> function can now operate on arrays
345 =head2 Perl's core time-related functions are now Y2038 compliand
347 =head2 The variable C<$,> may now be tied
349 =head2 // now behaves like || in when clauses
351 =head2 You can now set C<-W> from the C<PERL5OPT> environment varialbe
353 =head1 Modules and Pragmata
355 =head2 New Modules and Pragmata
361 This is a new lexically-scoped alternative for the C<Fatal> module.
362 The bundled version is 2.06_01. Note that in this release, using a string
363 eval when C<autodie> is in effect can cause the autodie behaviour to leak
364 into the surrounding scope. See L<autodie/"BUGS"> for more details.
366 =item C<Compress::Raw::Bzip2>
368 This has been added to the core (version 2.020).
372 This pragma establishes an ISA relationship with base classes at compile
373 time. It provides the key feature of C<base> without the feature creep.
375 =item C<Parse::CPAN::Meta>
377 This has been added to the core (version 1.39).
381 =head2 Pragmata Changes
387 See L</"The C<overloading> pragma"> above.
391 Upgraded from version 0.08 to 0.09.
395 The C<attrs> pragma has been removed. It had been marked as deprecated since
400 Upgraded from version 2.13 to 2.14. See L<parent> for a replacement.
404 Upgraded from version 0.22 to 0.23.
408 Upgraded from version 0.22 to 0.23.
412 Upgraded from version 0.22 to 0.23.
416 Upgraded from version 1.06 to 1.07.
418 The Unicode F<NameAliases.txt> database file has been added. This has the
419 effect of adding some extra C<\N> character names that formerly wouldn't
420 have been recognised; for example, C<"\N{LATIN CAPITAL LETTER GHA}">.
424 Upgraded from version 1.13 to 1.19. Some code has been shifted from run time to
425 compile time, and the amount of MRO cache flushing has been minimised.
429 The meaning of the C<:5.10> and C<:5.10.X> feature bundles has
430 changed slightly. The last component, if any (i.e. C<X>) is simply ignored.
431 This is predicated on the assumption that new features will not, in
432 general, be added to maintenance releases. So C<:5.10> and C<:5.10.X>
433 have identical effect. This is a change to the behaviour documented for
438 Upgraded from version 2.13 to 2.14 (this was just a version bump; there
439 were no functional changes).
443 Upgraded from version 0.5565 to 0.62.
447 Upgraded from version 1.06 to 1.07.
451 Upgraded from version 1.06 to 1.07.
455 See L</"The C<overloading> pragma"> above.
459 Upgraded from version 0.74 to 0.77.
463 =head2 Updated Modules
467 =item C<Archive::Extract>
469 Upgraded from version 0.24 to 0.34.
471 =item C<Archive::Tar>
473 Upgraded from version 1.38 to 1.52.
475 =item C<Attribute::Handlers>
477 Upgraded from version 0.79 to 0.85.
481 Upgraded from version 5.63 to 5.68.
485 Upgraded from version 1.05 to 1.06.
489 Upgraded from version 1.17 to 1.22.
493 Upgraded from version 1.05 to 1.11.
497 Upgraded from version 0.83 to 0.89.
501 Upgraded from version 1.09 to 1.11.
505 Upgraded from version 1.01 to 1.02.
509 Upgraded from version 1.10 to 1.11.
513 Upgraded from version 1.08 to 1.11.
515 L<Carp> now includes all the necessary code to function. Previously, it
516 used to be a lightweight placeholder that loaded the actual code from
517 C<Carp::Heavy> on demand. C<Carp::Heavy> is now a simple, empty module
518 kept for backwards compatibility for programs that used to pre-load it.
522 Upgraded from version 3.29 to 3.43.
523 (also includes the "default_value for popup_menu()" fix from 3.45).
525 =item C<Compress::Zlib>
527 Upgraded from version 2.008 to 2.020.
531 Upgraded from version 1.9205 to 1.9402. C<CPAN::FTP> has a local fix to
532 stop it being too verbose on download failure.
536 Upgraded from version 0.84 to 0.88.
538 =item C<CPANPLUS::Dist::Build>
540 Upgraded from version 0.06_02 to 0.36.
544 Upgraded from version 3.25_01 to 3.30.
546 =item C<Data::Dumper>
548 Upgraded from version 2.121_14 to 2.124.
552 Upgraded from version 1.01 to 1.02.
556 Upgraded from version 1.816_1 to 1.820.
558 =item C<Devel::PPPort>
560 Upgraded from version 3.13 to 3.19.
564 Upgraded from version 2.36_01 to 2.39.
568 Upgraded from version 5.45 to 5.47.
572 Upgraded from version 1.01 to 1.03.
576 Upgraded from version 1.12 to 1.13.
580 Upgraded from version 1.08 to 1.10.
584 Upgraded from version 2.23 to 2.35.
588 Upgraded from version 1.10 to 1.11.
592 Upgraded from version 5.62 to 5.63.
594 =item C<ExtUtils::CBuilder>
596 Upgraded from version 0.21 to 0.2602.
598 =item C<ExtUtils::Command>
600 Upgraded from version 1.13 to 1.16.
602 =item C<ExtUtils::Constant>
604 Upgraded from 0.20 to 0.22. (Note that neither of these versions are
607 =item C<ExtUtils::Embed>
609 Upgraded from version 1.27 to 1.28.
611 =item C<ExtUtils::Install>
613 Upgraded from version 1.44 to 1.54.
615 =item C<ExtUtils::MakeMaker>
617 Upgraded from version 6.42 to 6.55_02.
619 Note that C<ExtUtils::MakeMaker::bytes> and C<ExtUtils::MakeMaker::vmsish>
620 have been removed from this distribution.
622 =item C<ExtUtils::Manifest>
624 Upgraded from version 1.51_01 to 1.56.
626 =item C<ExtUtils::ParseXS>
628 Upgraded from version 2.18_02 to 2.2002.
632 Upgraded from version 1.05 to 2.06_01. See also the new pragma C<autodie>.
634 =item C<File::Basename>
636 Upgraded from version 2.76 to 2.77.
638 =item C<File::Compare>
640 Upgraded from version 1.1005 to 1.1006.
644 Upgraded from version 2.11 to 2.16.
646 File::Copy now always return 0 (not "") on failure.
648 FIXME - describe C<cp>
652 Upgraded from version 0.14 to 0.20.
656 Upgraded from version 1.12 to 1.14.
660 Upgraded from version 2.04 to 2.07_03.
664 Upgraded from version 3.2501 to 3.30.
668 Upgraded from version 1.00 to 1.01.
670 Added -X overloading, -M, -C and -A.
674 Upgraded from version 0.18 to 0.22.
678 Upgraded from version 1.07 to 1.08.
682 Upgraded from version 2.01 to 2.02.
684 =item C<Filter::Simple>
686 Upgraded from version 0.82 to 0.84.
688 =item C<Filter::Util::Call>
690 Upgraded from version 1.07 to 1.08.
694 Upgraded from version 1.49 to 1.50.
698 Upgraded from version 1.08 to 1.09.
700 =item C<Getopt::Long>
702 Upgraded from version 2.37 to 2.38.
704 =item C<Hash::Util::FieldHash>
706 Upgraded from version 1.03 to 1.04. This fixes a memory leak.
708 =item C<I18N::Collate>
710 Upgraded from version 1.00 to 1.01.
714 Upgraded from version 1.23_01 to 1.25.
716 This makes non-blocking mode work on Windows in C<IO::Socket::INET>
719 =item C<IO::Compress::*>
721 Upgraded from version 2.008 to 2.020.
725 Upgraded from version 1.06 to 1.07.
729 Upgraded from version 1.27 to 1.28.
733 Upgraded from version 1.30_01 to 1.31.
737 Upgraded from version 1.07 to 1.09.
741 Upgraded from version 0.40_1 to 0.46.
745 Upgraded from version 1.02 to 1.04.
749 Upgraded from version 1.05 to 2.01.
753 Upgraded from version 0.5565 to 0.62.
757 Upgraded from version 1.19 to 1.21.
759 =item C<Locale::MakeText>
761 Upgraded from version 1.12 to 1.13.
763 =item C<Log::Message>
765 Upgraded from version 0.01 to 0.02.
767 =item C<Math::BigFloat>
769 Upgraded from version 1.59 to 1.60.
771 =item C<Math::BigInt>
773 Upgraded from version 1.88 to 1.89.
775 =item C<Math::BigInt::FastCalc>
777 Upgraded from version 0.16 to 0.19.
779 =item C<Math::BigRat>
781 Upgraded from version 0.21 to 0.22.
783 =item C<Math::Complex>
785 Upgraded from version 1.37 to 1.56.
789 Upgraded from version 1.04 to 1.20.
793 Upgraded from version 1.01_02 to 1.01_03 (just a minor documentation
796 =item C<Module::Build>
798 Upgraded from version 0.2808_01 to 0.34_02.
800 =item C<Module::CoreList>
802 Upgraded from version 2.13 to 2.18. This release no longer contains the
803 C<%Module::CoreList::patchlevel> hash.
805 =item C<Module::Load>
807 Upgraded from version 0.12 to 0.16.
809 =item C<Module::Load::Conditional>
811 Upgraded from version 0.22 to 0.30.
813 =item C<Module::Loaded>
815 Upgraded from version 0.01 to 0.02.
817 =item C<Module::Pluggable>
819 Upgraded from version 3.6 to 3.9.
823 Upgraded from version 1.07 to 1.08.
827 Upgraded from version 2.33 to 2.36.
831 Upgraded from version 0.60_01 to 0.64.
833 =item C<Object::Accessor>
835 Upgraded from version 0.32 to 0.34.
839 Upgraded from version 1.03 to 1.04.
841 =item C<Package::Constants>
843 Upgraded from version 0.01 to 0.02.
847 Upgraded from version 1.04 to 1.06.
851 Upgraded from version 0.04 to 0.07.
855 Upgraded from version 2.16 to 2.22.
859 Upgraded from version 1.35 to 1.37.
863 Upgrade from version 3.14_02 to 3.15.
867 Upgraded from version 3.05 to 3.07.
871 Upgraded from version 3.08 to 3.13.
875 Upgraded from version 1.13 to 1.17.
879 Upgraded from 2.12 to 2.18.
881 =item C<Scalar::Util>
883 Upgraded from version 1.19 to 1.21.
887 Upgraded from 1.01 to 1.02.
891 Upgraded from 1.11 to 1.17.
895 Upgraded from 1.80 to 1.84.
897 As of 1.84, C<Socket> can now handle abstract namespace sockets on Linux.
902 Upgraded from 2.18 to 2.20.
906 Upgraded from version 2.13 to 2.14. Please see L</Deprecations>.
910 Upgraded from version 1.06 to 1.07.
914 Upgraded from version 0.22 to 0.27.
916 =item C<Term::ANSIColor>
918 Upgraded from version 1.12 to 2.01.
920 =item C<Term::ReadLine>
922 Upgraded from version 1.03 to 1.04.
926 Upgraded from version 0.18 to 0.20.
928 =item C<Test::Harness>
930 Upgraded from version 2.64 to 3.17.
932 Note that one side-effect of the 2.x to 3.x upgrade is that the
933 experimental C<Test::Harness::Straps> module (and its supporting
934 C<Assert>, C<Iterator>, C<Point> and C<Results> modules) have been
935 removed. If you still need this, then they are available in the
936 (unmaintained) C<Test-Harness-Straps> distribution on CPAN.
938 =item C<Test::Simple>
940 Upgraded from version 0.72 to 0.92.
942 =item C<Text::ParseWords>
944 Upgraded from version 3.26 to 3.27.
948 Upgraded from version 2007.1117 to 2009.0305.
952 Upgraded from version 2006.1117 to 2009.0305.
954 =item C<Thread::Queue>
956 Upgraded from version 2.00 to 2.11.
958 =item C<Thread::Semaphore>
960 Upgraded from version 2.01 to 2.09.
964 Upgraded from version 1.67 to 1.73.
966 =item C<threads::shared>
968 Upgraded from version 1.14 to 1.29.
970 =item C<Tie::RefHash>
972 Upgraded from version 1.37 to 1.38.
974 =item C<Tie::StdHandle>
976 This has documentation changes, and has been assigned a version for the
977 first time: version 4.2.
981 Upgraded from version 1.9711 to 1.9719.
985 Upgraded from version 1.18 to 1.1901.
989 Upgraded from version 1.12 to 1.15.
991 =item C<Unicode::Normalize>
993 Upgraded from version 1.02 to 1.03.
995 =item C<Unicode::UCD>
997 Upgraded from version 0.25 to 0.27.
999 C<charinfo()> now works on Unified CJK code points added to later versions
1002 C<casefold()> has new fields returned to provide both a simpler interface
1003 and previously missing information. The old fields are retained for
1004 backwards compatibility. Information about Turkic-specific code points is
1007 The documentation has been corrected and expanded.
1011 Upgraded from version 1.04 to 1.05.
1013 UNIVERSAL->import() is now deprecated.
1017 Upgraded from version 0.34 to 0.39.
1019 =item C<Win32API::File>
1021 Upgraded from version 0.1001_01 to 0.1101.
1025 Upgraded from version 0.08 to 0.10.
1029 =head1 Utility Changes
1035 Now looks in C<include-fixed> too, which is a recent addition to gcc's
1040 No longer incorrectly treats enum values like macros (Daniel Burr).
1042 Now handles C++ style constants (C<//>) properly in enums. (A patch from
1043 Rainer Weikusat was used; Daniel Burr also proposed a similar fix).
1047 C<LVALUE> subroutines now work under the debugger.
1049 The debugger now correctly handles proxy constant subroutines, and
1054 F<perlbug> now uses C<%Module::CoreList::bug_tracker> to print out upstream bug
1057 Where the user names a module that their bug report is about, and we know the
1058 URL for its upstream bug tracker, provide a message to the user explaining
1059 that the core copies the CPAN version directly, and provide the URL for
1060 reporting the bug directly to upstream.
1064 Perl 5.11.0 added a new utility F<perlthanks>, which is a variant of
1065 F<perlbug>, but for sending non-bug-reports to the authors and maintainers
1066 of Perl. Getting nothing but bug reports can become a bit demoralising:
1067 we'll see if this changes things.
1071 =head1 New Documentation
1077 This contains instructions on how to build perl for the Haiku platform.
1081 This describes the new interface for pluggable Method Resolution Orders.
1085 This document, by Richard Foley, provides an introduction to the use of
1086 performance and optimization techniques which can be used with particular
1087 reference to perl programs.
1089 =item L<perlrepository>
1091 This describes how to access the perl source using the I<git> version
1096 This describes the new F<perlthanks> utility.
1100 =head1 Changes to Existing Documentation
1102 The various large F<Changes*> files (which listed every change made to perl
1103 over the last 18 years) have been removed, and replaced by a small file,
1104 also called F<Changes>, which just explains how that same information may
1105 be extracted from the git version control system.
1107 The file F<Porting/patching.pod> has been deleted, as it mainly described
1108 interacting with the old Perforce-based repository, which is now obsolete.
1109 Information still relevant has been moved to L<perlrepository>.
1111 L<perlapi>, L<perlintern>, L<perlmodlib> and L<perltoc> are now all
1112 generated at build time, rather than being shipped as part of the release.
1114 =head2 Documented -X overloading.
1116 =head2 Documented that when() treats specially most of the filetest operators
1118 =head2 Documented when as a syntax modifier
1120 =head2 Eliminated "Old Perl threads tutorial", which describes 5005 threads.
1122 pod/perlthrtut.pod is the same material reworked for ithreads.
1124 =head2 Correct previous documentation: v-strings are not deprecated
1126 With version objects, we need them to use MODULE VERSION syntax. This
1127 patch removes the deprecation note.
1129 =head1 Performance Enhancements
1135 A new internal cache means that C<isa()> will often be faster.
1139 The implementation of C<C3> Method Resolution Order has been optimised -
1140 linearisation for classes with single inheritance is 40% faster. Performance
1141 for multiple inheritance is unchanged.
1145 Under C<use locale>, the locale-relevant information is now cached on
1146 read-only values, such as the list returned by C<keys %hash>. This makes
1147 operations such as C<sort keys %hash> in the scope of C<use locale> much
1152 Empty C<DESTROY> methods are no longer called.
1156 Faster sv_utf8_upgrade()
1160 Speed up keys() on empty hash
1164 =head1 Installation and Configuration Improvements
1166 =head2 F<ext/> reorganisation
1168 The layout of directories in F<ext> has been revised. Specifically, all
1169 extensions are now flat, and at the top level, with C</> in pathnames
1170 replaced by C<->, so that F<ext/Data/Dumper/> is now F<ext/Data-Dumper/>,
1171 etc. The names of the extensions as specified to F<Configure>, and as
1172 reported by C<%Config::Config> under the keys C<dynamic_ext>,
1173 C<known_extensions>, C<nonxs_ext> and C<static_ext> have not changed, and
1174 still use C</>. Hence this change will not have any affect once perl is
1175 installed. However, C<Attribute::Handlers>, C<Safe> and C<mro> have now
1176 become extensions in their own right, so if you run F<Configure> with
1177 options to specify an exact list of extensions to build, you will need to
1178 change it to account for this.
1180 For 5.11.1, it is planned that many dual-life modules will have been moved
1181 from F<lib> to F<ext>; again this will have no effect on an installed
1182 perl, but will matter if you invoke F<Configure> with a pre-canned list of
1183 extensions to build.
1185 =head2 Configuration improvements
1187 If C<vendorlib> and C<vendorarch> are the same, then they are only added to
1190 C<$Config{usedevel}> and the C-level C<PERL_USE_DEVEL> are now defined if
1191 perl is built with C<-Dusedevel>.
1193 F<Configure> will enable use of C<-fstack-protector>, to provide protection
1194 against stack-smashing attacks, if the compiler supports it.
1196 F<Configure> will now determine the correct prototypes for re-entrant
1197 functions, and for C<gconvert>, if you are using a C++ compiler rather
1200 On Unix, if you build from a tree containing a git repository, the
1201 configuration process will note the commit hash you have checked out, for
1202 display in the output of C<perl -v> and C<perl -V>. Unpushed local commits
1203 are automatically added to the list of local patches displayed by
1206 =head2 Compilation improvements
1208 As part of the flattening of F<ext>, all extensions on all platforms are
1209 built by F<make_ext.pl>. This replaces the Unix-specific
1210 F<ext/util/make_ext>, VMS-specific F<make_ext.com> and Win32-specific
1211 F<win32/buildext.pl>.
1213 =head2 Platform Specific Changes
1219 Removed F<libbsd> for AIX 5L and 6.1. Only flock() was used from F<libbsd>.
1221 Removed F<libgdbm> for AIX 5L and 6.1. The F<libgdbm> is delivered as an
1222 optional package with the AIX Toolbox. Unfortunately the 64 bit version
1225 Hints changes mean that AIX 4.2 should work again.
1229 On Cygwin we now strip the last number from the DLL. This has been the
1230 behaviour in the cygwin.com build for years. The hints files have been
1235 Support for Apollo DomainOS was removed in Perl 5.11.0
1239 The hints files now identify the correct threading libraries on FreeBSD 7
1244 We now work around a bizarre preprocessor bug in the Irix 6.5 compiler:
1245 C<cc -E -> unfortunately goes into K&R mode, but C<cc -E file.c> doesn't.
1249 Patches from the Haiku maintainers have been merged in. Perl should now
1254 Support for Atari MiNT was removed in Perl 5.11.0.
1258 Perl should now build on MirOS BSD.
1262 Hints now supports versions 5.*.
1266 Various changes from Stratus have been merged in.
1270 There is now support for Symbian S60 3.2 SDK and S60 5.0 SDK.
1274 Improved message window handling means that C<alarm> and C<kill> messages
1275 will no longer be dropped under race conditions.
1279 Reads from the in-memory temporary files of C<PerlIO::scalar> used to fail
1280 if C<$/> was set to a numeric reference (to indicate record-style reads).
1283 VMS now supports C<getgrgid>.
1285 Many improvements and cleanups have been made to the VMS file name handling
1286 and conversion code.
1288 Enabling the C<PERL_VMS_POSIX_EXIT> logical name now encodes a POSIX exit
1289 status in a VMS condition value for better interaction with GNV's bash
1290 shell and other utilities that depend on POSIX exit values. See
1291 L<perlvms/"$?"> for details.
1293 C<File::Copy> now detects Unix compatibility mode on VMS.
1297 =head1 Selected Bug Fixes
1303 C<-I> on shebang line now adds directories in front of @INC.
1304 as documented, and as does C<-I> when specified on the command-line.
1309 C<kill> is now fatal when called on non-numeric process identifiers.
1310 Previously, an 'undef' process identifier would be interpreted as a request to
1311 kill process "0", which would terminate the current process group on POSIX
1312 systems. Since process identifiers are always integers, killing a non-numeric
1313 process is now fatal.
1317 5.10.0 inadvertently disabled an optimisation, which caused a measurable
1318 performance drop in list assignment, such as is often used to assign
1319 function parameters from C<@_>. The optimisation has been re-instated, and
1320 the performance regression fixed.
1324 Fixed memory leak on C<while (1) { map 1, 1 }> [RT #53038].
1328 Some potential coredumps in PerlIO fixed [RT #57322,54828].
1332 The debugger now works with lvalue subroutines.
1336 The debugger's C<m> command was broken on modules that defined constants
1341 C<crypt()> and string complement could return tainted values for untainted
1342 arguments [RT #59998].
1346 The C<-i>I<.suffix> command-line switch now recreates the file using
1347 restricted permissions, before changing its mode to match the original
1348 file. This eliminates a potential race condition [RT #60904].
1352 On some UNIX systems, the value in C<$?> would not have the top bit set
1353 (C<$? & 128>) even if the child core dumped.
1357 Under some circumstances, C<$^R> could incorrectly become undefined
1362 In the XS API, various hash functions, when passed a pre-computed hash where
1363 the key is UTF-8, might result in an incorrect lookup.
1367 XS code including F<XSUB.h> before F<perl.h> gave a compile-time error
1372 C<< $object->isa('Foo') >> would report false if the package C<Foo> didn't
1373 exist, even if the object's C<@ISA> contained C<Foo>.
1377 Various bugs in the new-to 5.10.0 mro code, triggered by manipulating
1378 C<@ISA>, have been found and fixed.
1382 Bitwise operations on references could crash the interpreter, e.g.
1383 C<$x=\$y; $x |= "foo"> [RT #54956].
1387 Patterns including alternation might be sensitive to the internal UTF-8
1388 representation, e.g.
1390 my $byte = chr(192);
1391 my $utf8 = chr(192); utf8::upgrade($utf8);
1392 $utf8 =~ /$byte|X}/i; # failed in 5.10.0
1396 Within UTF8-encoded Perl source files (i.e. where C<use utf8> is in
1397 effect), double-quoted literal strings could be corrupted where a C<\xNN>,
1398 C<\0NNN> or C<\N{}> is followed by a literal character with ordinal value
1399 greater than 255 [RT #59908].
1403 C<B::Deparse> failed to correctly deparse various constructs:
1404 C<readpipe STRING> [RT #62428], C<CORE::require(STRING)> [RT #62488],
1405 C<sub foo(_)> [RT #62484].
1409 Using C<setpgrp()> with no arguments could corrupt the perl stack.
1413 The block form of C<eval> is now specifically trappable by C<Safe> and
1414 C<ops>. Previously it was erroneously treated like string C<eval>.
1418 In 5.10.0, the two characters C<[~> were sometimes parsed as the smart
1419 match operator (C<~~>) [RT #63854].
1423 In 5.10.0, the C<*> quantifier in patterns was sometimes treated as
1424 C<{0,32767}> [RT #60034, #60464]. For example, this match would fail:
1426 ("ab" x 32768) =~ /^(ab)*$/
1430 C<shmget> was limited to a 32 bit segment size on a 64 bit OS [RT #63924].
1434 Using C<next> or C<last> to exit a C<given> block no longer produces a
1435 spurious warning like the following:
1437 Exiting given via last at foo.pl line 123
1441 On Windows, C<'.\foo'> and C<'..\foo'> were treated differently than
1442 C<'./foo'> and C<'../foo'> by C<do> and C<require> [RT #63492].
1446 Assigning a format to a glob could corrupt the format; e.g.:
1448 *bar=*foo{FORMAT}; # foo format now bad
1452 Attempting to coerce a typeglob to a string or number could cause an
1453 assertion failure. The correct error message is now generated,
1454 C<Can't coerce GLOB to I<$type>>.
1458 Under C<use filetest 'access'>, C<-x> was using the wrong access mode. This
1459 has been fixed [RT #49003].
1463 C<length> on a tied scalar that returned a Unicode value would not be
1464 correct the first time. This has been fixed.
1468 Using an array C<tie> inside in array C<tie> could SEGV. This has been
1473 A race condition inside C<PerlIOStdio_close()> has been identified and
1474 fixed. This used to cause various threading issues, including SEGVs.
1478 In C<unpack>, the use of C<()> groups in scalar context was internally
1479 placing a list on the interpreter's stack, which manifested in various
1480 ways, including SEGVs. This is now fixed [RT #50256].
1484 Magic was called twice in C<substr>, C<\&$x>, C<tie $x, $m> and C<chop>.
1485 These have all been fixed.
1489 A 5.10.0 optimisation to clear the temporary stack within the implicit
1490 loop of C<s///ge> has been reverted, as it turned out to be the cause of
1491 obscure bugs in seemingly unrelated parts of the interpreter [commit
1496 The line numbers for warnings inside C<elsif> are now correct.
1500 The C<..> operator now works correctly with ranges whose ends are at or
1501 close to the values of the smallest and largest integers.
1505 C<binmode STDIN, ':raw'> could lead to segmentation faults on some platforms.
1506 This has been fixed [RT #54828].
1510 An off-by-one error meant that C<index $str, ...> was effectively being
1511 executed as C<index "$str\0", ...>. This has been fixed [RT #53746].
1515 Various leaks associated with named captures in regexes have been fixed
1520 A weak reference to a hash would leak. This was affecting C<DBI>
1525 Using (?|) in a regex could cause a segfault [RT #59734].
1529 Use of a UTF-8 C<tr//> within a closure could cause a segfault [RT #61520].
1533 Calling C<sv_chop()> or otherwise upgrading an SV could result in an
1534 unaligned 64-bit access on the SPARC architecture [RT #60574].
1538 In the 5.10.0 release, C<inc_version_list> would incorrectly list
1539 C<5.10.*> after C<5.8.*>; this affected the C<@INC> search order
1544 In 5.10.0, C<pack "a*", $tainted_value> returned a non-tainted value
1549 In 5.10.0, C<printf> and C<sprintf> could produce the fatal error
1550 C<panic: utf8_mg_pos_cache_update> when printing UTF-8 strings
1555 In the 5.10.0 release, a dynamically created C<AUTOLOAD> method might be
1556 missed (method cache issue) [RT #60220,60232].
1560 In the 5.10.0 release, a combination of C<use feature> and C<//ee> could
1561 cause a memory leak [RT #63110].
1565 C<-C> on the shebang (C<#!>) line is once more permitted if it is also
1566 specified on the command line. C<-C> on the shebang line used to be a
1567 silent no-op I<if> it was not also on the command line, so perl 5.10.0
1568 disallowed it, which broke some scripts. Now perl checks whether it is
1569 also on the command line and only dies if it is not [RT #67880].
1573 In 5.10.0, certain types of re-entrant regular expression could crash,
1574 or cause the following assertion failure [RT #60508]:
1576 Assertion rx->sublen >= (s - rx->subbeg) + i failed
1580 =head1 New or Changed Diagnostics
1584 =item C<panic: sv_chop %s>
1586 This new fatal error occurs when the C routine C<Perl_sv_chop()> was
1587 passed a position that is not within the scalar's string buffer. This
1588 could be caused by buggy XS code, and at this point recovery is not
1591 =item C<Can't locate package %s for the parents of %s>
1593 This warning has been removed. In general, it only got produced in
1594 conjunction with other warnings, and removing it allowed an ISA lookup
1595 optimisation to be added.
1597 =item C<v-string in use/require is non-portable>
1599 This warning has been removed.
1601 =item C<Deep recursion on subroutine "%s">
1603 It is now possible to change the depth threshold for this warning from the
1604 default of 100, by recompiling the F<perl> binary, setting the C
1605 pre-processor macro C<PERL_SUB_DEPTH_WARN> to the desired value.
1609 =head1 Changed Internals
1615 The J.R.R. Tolkien quotes at the head of C source file have been checked and
1616 proper citations added, thanks to a patch from Tom Christiansen.
1620 C<vcroak()> now accepts a null first argument. In addition, a full audit
1621 was made of the "not NULL" compiler annotations, and those for several
1622 other internal functions were corrected.
1626 New macros C<dSAVEDERRNO>, C<dSAVE_ERRNO>, C<SAVE_ERRNO>, C<RESTORE_ERRNO>
1627 have been added to formalise the temporary saving of the C<errno>
1632 The function C<Perl_sv_insert_flags> has been added to augment
1637 The function C<Perl_newSV_type(type)> has been added, equivalent to
1638 C<Perl_newSV()> followed by C<Perl_sv_upgrade(type)>.
1642 The function C<Perl_newSVpvn_flags()> has been added, equivalent to
1643 C<Perl_newSVpvn()> and then performing the action relevant to the flag.
1645 Two flag bits are currently supported.
1651 This will call C<SvUTF8_on()> for you. (Note that this does not convert an
1652 sequence of ISO 8859-1 characters to UTF-8). A wrapper, C<newSVpvn_utf8()>
1653 is available for this.
1657 Call C<sv_2mortal()> on the new SV.
1661 There is also a wrapper that takes constant strings, C<newSVpvs_flags()>.
1665 The function C<Perl_croak_xs_usage> has been added as a wrapper to
1670 The functions C<PerlIO_find_layer> and C<PerlIO_list_alloc> are now
1675 C<PL_na> has been exterminated from the core code, replaced by local STRLEN
1676 temporaries, or C<*_nolen()> calls. Either approach is faster than C<PL_na>,
1677 which is a pointer deference into the interpreter structure under ithreads,
1678 and a global variable otherwise.
1682 C<Perl_mg_free()> used to leave freed memory accessible via SvMAGIC() on
1683 the scalar. It now updates the linked list to remove each piece of magic
1688 Under ithreads, the regex in C<PL_reg_curpm> is now reference counted. This
1689 eliminates a lot of hackish workarounds to cope with it not being reference
1694 C<Perl_mg_magical()> would sometimes incorrectly turn on C<SvRMAGICAL()>.
1695 This has been fixed.
1699 The I<public> IV and NV flags are now not set if the string value has
1700 trailing "garbage". This behaviour is consistent with not setting the
1701 public IV or NV flags if the value is out of range for the type.
1705 SV allocation tracing has been added to the diagnostics enabled by C<-Dm>.
1706 The tracing can alternatively output via the C<PERL_MEM_LOG> mechanism, if
1707 that was enabled when the F<perl> binary was compiled.
1711 Uses of C<Nullav>, C<Nullcv>, C<Nullhv>, C<Nullop>, C<Nullsv> etc have been
1712 replaced by C<NULL> in the core code, and non-dual-life modules, as C<NULL>
1713 is clearer to those unfamiliar with the core code.
1717 A macro C<MUTABLE_PTR(p)> has been added, which on (non-pedantic) gcc will
1718 not cast away C<const>, returning a C<void *>. Macros C<MUTABLE_SV(av)>,
1719 C<MUTABLE_SV(cv)> etc build on this, casting to C<AV *> etc without
1720 casting away C<const>. This allows proper compile-time auditing of
1721 C<const> correctness in the core, and helped picked up some errors (now
1726 Macros C<mPUSHs()> and C<mXPUSHs()> have been added, for pushing SVs on the
1727 stack and mortalizing them.
1731 Use of the private structure C<mro_meta> has changed slightly. Nothing
1732 outside the core should be accessing this directly anyway.
1736 A new tool, F<Porting/expand-macro.pl> has been added, that allows you
1737 to view how a C preprocessor macro would be expanded when compiled.
1738 This is handy when trying to decode the macro hell that is the perl
1745 Many modules updated from CPAN incorporate new tests.
1747 Several tests that have the potential to hang forever if they fail now
1748 incorporate a "watchdog" functionality that will kill them after a timeout,
1749 which helps ensure that C<make test> and C<make test_harness> run to
1750 completion automatically. (Jerry Hedden).
1752 Some core-specific tests have been added:
1756 =item t/comp/retainedlines.t
1758 Check that the debugger can retain source lines from C<eval>.
1760 =item t/io/perlio_fail.t
1762 Check that bad layers fail.
1764 =item t/io/perlio_leaks.t
1766 Check that PerlIO layers are not leaking.
1768 =item t/io/perlio_open.t
1770 Check that certain special forms of open work.
1774 General PerlIO tests.
1778 Check that there is no unexpected interaction between the internal types
1779 C<PVBM> and C<PVGV>.
1781 =item t/mro/package_aliases.t
1783 Check that mro works properly in the presence of aliased packages.
1787 Tests for C<dbmopen> and C<dbmclose>.
1789 =item t/op/index_thr.t
1791 Tests for the interaction of C<index> and threads.
1793 =item t/op/pat_thr.t
1795 Tests for the interaction of esoteric patterns and threads.
1799 Test that C<qr> doesn't leak.
1801 =item t/op/reg_email_thr.t
1803 Tests for the interaction of regex recursion and threads.
1805 =item t/op/regexp_qr_embed_thr.t
1807 Tests for the interaction of patterns with embedded C<qr//> and threads.
1809 =item t/op/regexp_unicode_prop.t
1811 Tests for Unicode properties in regular expressions.
1813 =item t/op/regexp_unicode_prop_thr.t
1815 Tests for the interaction of Unicode properties and threads.
1817 =item t/op/reg_nc_tie.t
1819 Test the tied methods of C<Tie::Hash::NamedCapture>.
1821 =item t/op/reg_posixcc.t
1823 Check that POSIX character classes behave consistently.
1827 Check that exportable C<re> functions in F<universal.c> work.
1829 =item t/op/setpgrpstack.t
1831 Check that C<setpgrp> works.
1833 =item t/op/substr_thr.t
1835 Tests for the interaction of C<substr> and threads.
1837 =item t/op/upgrade.t
1839 Check that upgrading and assigning scalars works.
1841 =item t/uni/lex_utf8.t
1843 Check that Unicode in the lexer works.
1847 Check that Unicode and C<tie> work.
1851 =head1 Known Problems
1853 This is a list of some significant unfixed bugs, which are regressions
1854 from either 5.10.0 or 5.8.x.
1860 C<List::Util::first> misbehaves in the presence of a lexical C<$_>
1861 (typically introduced by C<my $_> or implicitly by C<given>). The variable
1862 which gets set for each iteration is the package variable C<$_>, not the
1863 lexical C<$_> [RT #67694].
1865 A similar issue may occur in other modules that provide functions which
1866 take a block as their first argument, like
1868 foo { ... $_ ...} list
1872 The C<charnames> pragma may generate a run-time error when a regex is
1873 interpolated [RT #56444]:
1875 use charnames ':full';
1876 my $r1 = qr/\N{THAI CHARACTER SARA I}/;
1877 "foo" =~ $r1; # okay
1878 "foo" =~ /$r1+/; # runtime error
1880 A workaround is to generate the character outside of the regex:
1882 my $a = "\N{THAI CHARACTER SARA I}";
1887 Some regexes may run much more slowly when run in a child thread compared
1888 with the thread the pattern was compiled into [RT #55600].
1894 The following items are now deprecated.
1900 C<Switch> is buggy and should be avoided. From perl 5.11.0 onwards, it is
1901 intended that any use of the core version of this module will emit a
1902 warning, and that the module will eventually be removed from the core
1903 (probably in perl 5.14.0). See L<perlsyn/"Switch statements"> for its
1908 C<suidperl> has been removed. It used to provide a mechanism to
1909 emulate setuid permission bits on systems that don't support it properly.
1913 Deprecate assignment to $[
1917 Remove attrs, which has been deprecated since 1999/10/02.
1921 Deprecate use of the attribute :locked on subroutines.
1925 Deprecate using "locked" with the attributes pragma.
1929 Deprecate using "unique" with the attributes pragma.
1933 warn if ++ or -- are unable to change the value because it's beyond the limit of representation
1935 This uses a new warnings category: "imprecision".
1939 Make lc/uc/lcfirst/ucfirst warn when passed undef.
1943 Show constant in "Useless use of a constant in void context"
1947 Make the new warning report undef constants as undef
1951 Add a new warning, "Prototype after '%s'"
1955 Tweak the "Illegal character in prototype" warning so it's more precise when reporting illegal characters after _
1959 Unintented interpolation of $\ in regex
1963 Make overflow warnings in gmtime/localtime only occur when warnings are on
1967 Improve mro merging error messages.
1969 They are now very similar to those produced by Algorithm::C3.
1973 Amelioration of the error message "Unrecognized character %s in column %d"
1975 Changes the error message to "Unrecognized character %s; marked by <--
1976 HERE after %s<-- HERE near column %d". This should make it a little
1977 simpler to spot and correct the suspicious character.
1981 Explicitely point to $. when it causes an uninitialized warning for ranges in scalar context
1985 =head1 Acknowledgements
1987 Some of the work in this release was funded by a TPF grant.
1989 Nicholas Clark officially retired from maintenance pumpking duty at the
1990 end of 2008; however in reality he has put much effort in since then to
1991 help get 5.10.1 into a fit state to be released, including writing a
1992 considerable chunk of this perldelta.
1994 Steffen Mueller and David Golden in particular helped getting CPAN modules
1995 polished and synchronised with their in-core equivalents.
1997 Craig Berry was tireless in getting maint to run under VMS, no matter how
1998 many times we broke it for him.
2000 The other core committers contributed most of the changes, and applied most
2001 of the patches sent in by the hundreds of contributors listed in F<AUTHORS>.
2003 Much of the work of categorizing changes in this perldelta file was contributed
2004 by the following porters using changelogger.bestpractical.com:
2006 Nicholas Clark, leon, shawn, alexm, rjbs, rafl, Pedro Melo, brunorc,
2007 anonymous, ☄, Tom Hukins, anonymous, Jesse, dagolden, Moritz Onken,
2008 Mark Fowler, chorny, anonymous, tmtm
2010 Finally, thanks to Larry Wall, without whom none of this would be
2013 =head1 Reporting Bugs
2015 If you find what you think is a bug, you might check the articles
2016 recently posted to the comp.lang.perl.misc newsgroup and the perl
2017 bug database at http://rt.perl.org/perlbug/ . There may also be
2018 information at http://www.perl.org/ , the Perl Home Page.
2020 If you believe you have an unreported bug, please run the B<perlbug>
2021 program included with your release. Be sure to trim your bug down
2022 to a tiny but sufficient test case. Your bug report, along with the
2023 output of C<perl -V>, will be sent off to perlbug@perl.org to be
2024 analysed by the Perl porting team.
2026 If the bug you are reporting has security implications, which make it
2027 inappropriate to send to a publicly archived mailing list, then please send
2028 it to perl5-security-report@perl.org. This points to a closed subscription
2029 unarchived mailing list, which includes all the core committers, who be able
2030 to help assess the impact of issues, figure out a resolution, and help
2031 co-ordinate the release of patches to mitigate or fix the problem across all
2032 platforms on which Perl is supported. Please only use this address for
2033 security issues in the Perl core, not for modules independently
2034 distributed on CPAN.
2038 The F<Changes> file for an explanation of how to view exhaustive details
2041 The F<INSTALL> file for how to build Perl.
2043 The F<README> file for general stuff.
2045 The F<Artistic> and F<Copying> files for copyright information.
2052 The following changes are a filtered list of changes which weren't backported
2053 to 5.10.1. They were run through changelogger.bestpractical.com and triaged
2054 by a set of Perl 5 Porters. Changes to blead after 5945e41e have not yet
2055 been triaged or integrated.
2057 The following changes need to be
2059 1) deleted if they don't merit inclusion
2063 2) lightly copyedited and integrated into the perldelta above. Unfortunately, many of our
2064 commit messages are somewhat terse and require a bit more help to turn into something readable
2071 =head2 Move the reg_stringify logic to Perl_sv_2pv_flags
2073 =head2 mg_copy ought to take an I32
2075 =head2 Perl_store_cop_label() isn't meant to be part of the public API.
2077 =head2 Perl_gv_fetchmethod{,_autoload,_flags} are actually never* called with a non-NULL stash.
2078 So change the parameter to NN.
2081 =head2 Promote Perl_setdefout() to the public API.
2083 =head2 Add get_cvs() as a shortcut for STR_WITH_LEN() and Perl_get_cvn_flags()
2085 =head2 In Perl_newCONSTSUB(), sv should not be NULL.
2087 =head2 GvUNIQUE* have been defined as 0 since 2005/06/30 - high time to remove them.
2089 =head2 invert and rename PERL_MEM_LOG_STDERR to PERL_MEM_LOG_NOIMPL
2091 Most users who want PERL_MEM_LOG want the default implementation,
2092 give it to them. Users providing their own implementation can
2093 obtain current behavior by adding -DPERL_MEM_LOG_NOIMPL.
2094 Frankly, the average user probably wants _ENV by default too.
2096 =head2 simplify PERL_MEM_LOG
2098 This combines multiple environment variable reads into 1,
2099 where it looks for values like "2mst"
2100 -2 leading digits are atoi()d to get FD
2101 -m memory logging please
2103 -t timestamp those please.
2105 Combining these reduces overhead such that it seemed
2106 worthwhile to drop all the ifdefs. TBD whether this works
2107 in the environment that drove the original tradeoffs.
2109 If it isnt enough, Id be tempted by a global static ptr,
2110 and on 1st use, is read, seen 0, a lock is taken, and getenvar
2111 run to populate it, unlocked, proceed. This would remove
2112 iterative overheads.
2114 =head2 Add a parameter "destructing" to Gv_AMupdate()
2116 This boolean parameter indicates if the function has been called
2117 to update the overload magic table while looking up the DESTROY
2118 method. In this case, it's probably best to avoid croaking if
2119 those tables could not be updated (for example due to a method
2120 that could not be loaded.)
2122 =head2 Modify the return value of Gv_AMupdate to indicate a compilation error
2124 This way we'll restore most of the performance on object desctruction
2125 lost by the previous commit
2128 =head2 local $SIG{FOO} = sub {...}; sets signal handler to SIG_DFL
2130 Re: [perl #60360] [PATCH] UPDATED: local $SIG{FOO} = sub {...}; sets signal handler to SIG_DFL
2131 Message-ID: <20081112234504.GI2062@tytlal.topaz.cx>
2133 Updated patch to retain source compatibility.
2135 Plus using the correct PERL_ARGS_ASSERT_SAVE_HELEM_FLAGS
2136 macro and running make regen.
2138 =head2 Respecting inc_version_list while processing PERL_VENDORLIB_STEM
2140 Respecting inc_version_list while processing PERL_VENDORLIB_STEM
2141 From: "Mandalemula, Rajesh" <Rajesh.Mandalemula@deshaw.com>
2143 =head2 Change PL_debug behaviour
2146 String eval lines are now saved whenever
2147 a subroutine is defined, even if the eval'd string has subsequent
2148 syntax errors. This allows the debugger to single step into these
2152 =head2 Tied filehandles now have an additional method EOF which provides the EOF type
2154 =head2 Perl is now smarter about adding a -I dir to the beginning or end of @INC
2156 =head2 On scope end, delete localized array elements that should not exist anymore, so that the array recovers its previous length. Honour EXISTS and DELETE for tied arrays.
2158 =head2 When a glob is deleted, mark its sub as ANON.
2160 =head2 Require a space or a newline after a "#line XXX" directive
2162 =head2 Forbid using "foreach" as an attribute
2164 (like all other control flow statements)
2167 =head2 Unregister signal handlers before destroying my_perl
2169 If the signal handler runs after perl_destruct() has been called, it
2170 will get an invalid (or NULL) my_perl when it asks for the
2171 thread-specific interpreter struct. This patch resets the signal
2172 handler for any signal previously handled by PL_csighandlerp to SIG_DFL
2173 before calling perl_destruct().
2175 =head2 Perl_magic_clearsig() needs to remove magic, else delete $SIG{INT} returns undef
2177 Perl_magic_clearsig() needs to remove magic, else delete $SIG{INT} returns undef
2178 instead of the now-removed INT handler.
2180 =head2 [perl #66452] TMPDIR not honored when opening an anonymous temporary file
2182 [perl #66452] TMPDIR not honored when opening an anonymous temporary file
2184 =head2 The attached patch to perlio.c fixes the problem of errno getting set.
2186 While I am firmly in the school of "do not look at $! except immediately
2187 after a failure", I also agree that spuriously setting it is messy. But
2188 there is just no way of knowing where your errno might have been.
2190 The problem was that PerlIO_fast_gets() (and other nearby similar
2191 capability-checking PerlIO routines) set the errno (and it was being
2192 called a lot, from sv_gets()). I think setting the errno here was
2193 a mistake: checking for "can has FOO" should not set external state,
2194 such as the errno. The patch removes that errno trashing from all those
2197 =head2 Trim all trailing / from "." in @INC when filling %INC
2199 This fixes bug #66942 : as a / was left in the directory name,
2200 $INC{"Foo.pm"} for a file loaded from the current directory
2201 was given the incorrect value "/Foo.pm".
2203 =head2 Don't enqueue pending signals during global destruction
2205 Global destruction is not signal-safe. PL_psig_pend may already
2206 be gone when the signal handler is called (with destruct_level > 0).
2207 NULL it before freeing it to prevent a race condition.
2209 =head2 Eliminate struct regexp_allocated and xpvio_allocated.
2211 Calculate memory allocation using regexp and XPVIO, and the offset of the first
2212 real structure member. This avoids tripping over alignment differences between
2213 X* and x*_allocated, because x*_allocated doesn't have a double in it.
2218 =head2 [perl #47047] Use of inherited AUTOLOAD for non-method is deprecated
2220 =head2 Remove the definitions of Null(), Nullch, Nullfp, Nullsv and PL_na when code is within the perl source tree
2222 =head2 Replace our assert-which-can-be-caught-by-eval with the real deal from the standard C library.
2224 =head2 Tweak Perl_sv_upgrade() so that references can upgrade to SVt_PV
2226 =head2 Eliminate prelen from struct regexp.
2228 =head2 Change Perl_av_iter_p() to return IV* rather than I32* (which means
2230 =head2 Reorder the external regexp flags to get RXf_PMf_STD_PMMOD into the
2232 lowest 4 bits (which saves a shift), and the "flags indicating special
2233 patterns" into contiguous bits. This makes everything a little tidier,
2234 and saves 88 bytes (woohoo!) of object file with -Os on x86 FreeBSD.
2237 =head2 Re-implement the SvOOK() hack to store the offset as a BER encoded number in the part of the PVX that is being released.
2239 fit, as chopping off 1 byte gives just enough space for recording a
2240 delta of up to 127). This allows SvOOK() to co-exist with SvIOK_on(),
2241 which means all the calls to SvOOK_off() [with the possibility of a
2242 call to sv_backoff()] in SvIOK_on() can be removed. This ought to make
2243 a lot of straight line code a little bit simpler.
2244 OOK()d scalars can now be SVt_PV, as the IVX isn't needed.
2246 =head2 Abolish wraplen from struct regexp. We're already storing it in SvCUR.
2248 =head2 Make Perl_pregcomp() use SvUTF8() of the pattern, rather than the flag bit in pmflags, to decide whether the pattern is UTF-8.
2250 =head2 Abolish RXf_UTF8. Store the UTF-8-ness of the pattern with SvUTF8().
2252 =head2 In struct regexp move the member paren_names to the IV union.
2254 =head2 Make REGEXP a type distinct from SV. (Much like AV, CV, GV, HV).
2256 =head2 Allow sv_setsv_flags() to copy SVt_REGEXP much like it copies SVt_FORMAT - the just string buffer.
2259 =head2 Correct a long-standing ithreads reference counting anonamly
2261 The reference count only needs "doubling" when the scalar is pushed onto
2262 PL_regex_padav for the second time.
2265 =head2 In PL_regexp_padav, store regexps via real references, rather than hiding them within IVs.
2267 We can do this now that they are real SV pointers.
2269 =head2 With regexps stored as real RVs, we can eliminate SvREPADTMP().
2271 =head2 REGEXPs are now stored directly in PL_regex_padav, rather than indirectly via RVs.
2273 =head2 Remove code that protected pp_qr against REGEXPs going away during global destruction whilst they were stored via true references in PL_regex_padav.
2275 =head2 Remove PM_GETRE_SAFE and PM_SETRE_SAFE as nothing uses them.
2277 =head2 Note the U8 sized space created by removing -P, and check that it is now an illegal command line flag.
2279 =head2 Pack the recycled pad offsets into an SV at PL_regex_pad[0].
2281 =head2 Re-order so that the !SvOK() case is last (which should be rare)
2283 =head2 Extend PUSHFORMAT() to take a second parameter to set retop, to save NULLing it and then reassigning.
2285 =head2 Split struct block_sub into struct block_sub and struct block_format.
2287 Split struct block_sub into struct block_sub and struct block_format.
2288 (CXt_SUB and CXt_FORMAT were using some comon members, but some members
2289 were only for one or the other.)
2291 =head2 Change the wantarray result from caller from IV to bool for the SCALAR/ARRAY case.
2293 This doesn't contradict the documentation, as there isn't any. Oops.
2296 =head2 Give G_VOID, G_SCALAR and G_ARRAY the same numeric values as OPf_WANT_VOID, OPf_WANT_SCALAR and OPf_WANT_LIST.
2299 =head2 Squeeze the context type down to 4 bits, and move the private flags to fit within the next 4 bits.
2302 =head2 In struct block change blku_type from U8 to U16, and the "spare" U8 to U16, with the lockstep changes in struct subst.
2304 struct block_sub, and instead store it in the U16 in struct block.
2307 =head2 In struct block_eval, eliminate old_in_eval and old_op_type by storing the data in blk_u16.
2310 =head2 The layout for struct block_loop under ithreads can be simplified.
2312 Instead of wedging the pad offset into a void* iterdata, and always
2313 storing PL_comppad even when it isn't used, instead do this:
2315 PAD *oldcomppad; /* Also used for the GV, if targoffset is 0 */
2316 /* This is also accesible via cx->blk_loop.my_op->op_targ */
2317 PADOFFSET targoffset;
2319 and store the GV pointer in oldcompad. Pointers to pointers seems
2320 cleaner. This also allows us to eliminate the flag bit CXp_PADVAR.
2323 =head2 In XS_PerlIO_get_layers() take advantage of the implementation of
2325 In XS_PerlIO_get_layers() take advantage of the implementation of
2326 PerlIO_get_layers(), by co-opting the new SVs it creates, rather than
2330 =head2 Micro-optimise the order of the context types. [Because I can :-)]
2332 =head2 [patch] optimize OP_IS_(FILETEST|SOCKET) macros
2334 =head2 Eliminate ck_lengthconst.
2336 =head2 Chainsaw DEBUG_S out, as suggested by Vincent Pit.
2338 =head2 Unsupported private API functions are now declared "static" to prevent leakage to the public API
2340 =head2 Perl_cv_ckproto() is not part of the public API, and not used anywhere. It has been removed
2342 =head2 Remove all the 5005threads specific mutex macros, which are now vestigial.
2344 =head2 Do not honor TMPDIR for anonymous temporary files when tainting
2346 Use a default of /tmp on Unixes when TMPDIR is unset or empty, or
2347 when creation of a temporary file in it fails
2349 =head2 Add a pluggable hook in op_free()