This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Fix RT number in perl5260delta.pod
[perl5.git] / pod / perl5260delta.pod
CommitLineData
b5cbfe35
S
1=encoding utf8
2
3=head1 NAME
4
5perl5260delta - what is new for perl v5.26.0
6
7=head1 DESCRIPTION
8
9This document describes the differences between the 5.24.0 release and the
105.26.0 release.
11
12=head1 Notice
13
14This release includes three updates with widespread effects:
15
16=over 4
17
18=item * C<"."> no longer in C<@INC>
19
20For security reasons, the current directory (C<".">) is no longer included
21by default at the end of the module search path (C<@INC>). This may have
22widespread implications for the building, testing and installing of
23modules, and for the execution of scripts. See the section
24L<< Removal of the current directory (C<".">) from C<@INC> >>
25for the full details.
26
27=item * C<do> may now warn
28
29C<do> now gives a deprecation warning when it fails to load a file which
30it would have loaded had C<"."> been in C<@INC>.
31
32=item * In regular expression patterns, a literal left brace C<"{">
33should be escaped
34
35See L</Unescaped literal C<"{"> characters in regular expression patterns are no longer permissible>.
36
37=back
38
39=head1 Core Enhancements
40
41=head2 Lexical subroutines are no longer experimental
42
43Using the C<lexical_subs> feature introduced in v5.18 no longer emits a warning. Existing
44code that disables the C<experimental::lexical_subs> warning category
45that the feature previously used will continue to work. The
46C<lexical_subs> feature has no effect; all Perl code can use lexical
47subroutines, regardless of what feature declarations are in scope.
48
49=head2 Indented Here-documents
50
51This adds a new modifier C<"~"> to here-docs that tells the parser
52that it should look for C</^\s*$DELIM\n/> as the closing delimiter.
53
54These syntaxes are all supported:
55
56 <<~EOF;
57 <<~\EOF;
58 <<~'EOF';
59 <<~"EOF";
60 <<~`EOF`;
61 <<~ 'EOF';
62 <<~ "EOF";
63 <<~ `EOF`;
64
65The C<"~"> modifier will strip, from each line in the here-doc, the
66same whitespace that appears before the delimiter.
67
68Newlines will be copied as-is, and lines that don't include the
69proper beginning whitespace will cause perl to croak.
70
71For example:
72
73 if (1) {
74 print <<~EOF;
75 Hello there
76 EOF
77 }
78
79prints "Hello there\n" with no leading whitespace.
80
81=head2 New regular expression modifier C</xx>
82
83Specifying two C<"x"> characters to modify a regular expression pattern
84does everything that a single one does, but additionally TAB and SPACE
85characters within a bracketed character class are generally ignored and
86can be added to improve readability, like
87S<C</[ ^ A-Z d-f p-x ]/xx>>. Details are at
88L<perlre/E<sol>x and E<sol>xx>.
89
90=head2 C<@{^CAPTURE}>, C<%{^CAPTURE}>, and C<%{^CAPTURE_ALL}>
91
92C<@{^CAPTURE}> exposes the capture buffers of the last match as an
93array. So C<$1> is C<${^CAPTURE}[0]>. This is a more efficient equivalent
94to code like C<substr($matched_string,$-[0],$+[0]-$-[0])>, and you don't
95have to keep track of the C<$matched_string> either. This variable has no
96single character equivalent. Note that, like the other regex magic variables,
97the contents of this variable is dynamic; if you wish to store it beyond
98the lifetime of the match you must copy it to another array.
99
100C<%{^CAPTURE}> is equivalent to C<%+> (I<i.e.>, named captures). Other than
101being more self-documenting there is no difference between the two forms.
102
103C<%{^CAPTURE_ALL}> is equivalent to C<%-> (I<i.e.>, all named captures).
104Other than being more self-documenting there is no difference between the
105two forms.
106
107=head2 Declaring a reference to a variable
108
109As an experimental feature, Perl now allows the referencing operator to come
110after L<C<my()>|perlfunc/my>, L<C<state()>|perlfunc/state>,
111L<C<our()>|perlfunc/our>, or L<C<local()>|perlfunc/local>. This syntax must
112be enabled with C<use feature 'declared_refs'>. It is experimental, and will
113warn by default unless C<no warnings 'experimental::refaliasing'> is in effect.
114It is intended mainly for use in assignments to references. For example:
115
116 use experimental 'refaliasing', 'declared_refs';
117 my \$a = \$b;
118
119See L<perlref/Assigning to References> for more details.
120
121=head2 Unicode 9.0 is now supported
122
123A list of changes is at L<http://www.unicode.org/versions/Unicode9.0.0/>.
124Modules that are shipped with core Perl but not maintained by p5p do not
125necessarily support Unicode 9.0. L<Unicode::Normalize> does work on 9.0.
126
127=head2 Use of C<\p{I<script>}> uses the improved Script_Extensions property
128
129Unicode 6.0 introduced an improved form of the Script (C<sc>) property, and
130called it Script_Extensions (C<scx>). Perl now uses this improved
131version when a property is specified as just C<\p{I<script>}>. This
132should make programs more accurate when determining if a character is
133used in a given script, but there is a slight chance of breakage for
134programs that very specifically needed the old behavior. The meaning of
135compound forms, like C<\p{sc=I<script>}> are unchanged. See
136L<perlunicode/Scripts>.
137
138=head2 Perl can now do default collation in UTF-8 locales on platforms
139that support it
140
141Some platforms natively do a reasonable job of collating and sorting in
142UTF-8 locales. Perl now works with those. For portability and full
143control, L<Unicode::Collate> is still recommended, but now you may
144not need to do anything special to get good-enough results, depending on
145your application. See
146L<perllocale/Category C<LC_COLLATE>: Collation: Text Comparisons and Sorting>.
147
148=head2 Better locale collation of strings containing embedded C<NUL>
149characters
150
151In locales that have multi-level character weights, C<NUL>s are now
152ignored at the higher priority ones. There are still some gotchas in
153some strings, though. See
154L<perllocale/Collation of strings containing embedded C<NUL> characters>.
155
156=head2 C<CORE> subroutines for hash and array functions callable via
157reference
158
159The hash and array functions in the C<CORE> namespace (C<keys>, C<each>,
160C<values>, C<push>, C<pop>, C<shift>, C<unshift> and C<splice>) can now
161be called with ampersand syntax (C<&CORE::keys(\%hash>) and via reference
162(C<< my $k = \&CORE::keys; $k-E<gt>(\%hash) >>). Previously they could only be
163used when inlined.
164
165=head2 New Hash Function For 64-bit Builds
166
167We have switched to a hybrid hash function to better balance
168performance for short and long keys.
169
170For short keys, 16 bytes and under, we use an optimised variant of
171One At A Time Hard, and for longer keys we use Siphash 1-3. For very
172long keys this is a big improvement in performance. For shorter keys
173there is a modest improvement.
174
175=head1 Security
176
177=head2 Removal of the current directory (C<".">) from C<@INC>
178
179The perl binary includes a default set of paths in C<@INC>. Historically
180it has also included the current directory (C<".">) as the final entry,
181unless run with taint mode enabled (C<perl -T>). While convenient, this has
182security implications: for example, where a script attempts to load an
183optional module when its current directory is untrusted (such as F</tmp>),
184it could load and execute code from under that directory.
185
186Starting with v5.26, C<"."> is always removed by default, not just under
187tainting. This has major implications for installing modules and executing
188scripts.
189
190The following new features have been added to help ameliorate these
191issues.
192
193=over
194
195=item * F<Configure -Udefault_inc_excludes_dot>
196
197There is a new F<Configure> option, C<default_inc_excludes_dot> (enabled
198by default) which builds a perl executable without C<".">; unsetting this
199option using C<-U> reverts perl to the old behaviour. This may fix your
200path issues but will reintroduce all the security concerns, so don't
201build a perl executable like this unless you're I<really> confident that
202such issues are not a concern in your environment.
203
204=item * C<PERL_USE_UNSAFE_INC>
205
206There is a new environment variable recognised by the perl interpreter.
207If this variable has the value 1 when the perl interpreter starts up,
208then C<"."> will be automatically appended to C<@INC> (except under tainting).
209
210This allows you restore the old perl interpreter behaviour on a
211case-by-case basis. But note that this is intended to be a temporary crutch,
212and this feature will likely be removed in some future perl version.
213It is currently set by the C<cpan> utility and C<Test::Harness> to
214ease installation of CPAN modules which have not been updated to handle the
215lack of dot. Once again, don't use this unless you are sure that this
216will not reintroduce any security concerns.
217
218=item * A new deprecation warning issued by C<do>.
219
220While it is well-known that C<use> and C<require> use C<@INC> to search
221for the file to load, many people don't realise that C<do "file"> also
222searches C<@INC> if the file is a relative path. With the removal of C<".">,
223a simple C<do "file.pl"> will fail to read in and execute C<file.pl> from
224the current directory. Since this is commonly expected behaviour, a new
225deprecation warning is now issued whenever C<do> fails to load a file which
226it otherwise would have found if a dot had been in C<@INC>.
227
228=back
229
230Here are some things script and module authors may need to do to make
231their software work in the new regime.
232
233=over
234
235=item * Script authors
236
237If the issue is within your own code (rather than within included
238modules), then you have two main options. Firstly, if you are confident
239that your script will only be run within a trusted directory (under which
240you expect to find trusted files and modules), then add C<"."> back into the
241path; I<e.g.>:
242
243 BEGIN {
244 my $dir = "/some/trusted/directory";
245 chdir $dir or die "Can't chdir to $dir: $!\n";
246 # safe now
247 push @INC, '.';
248 }
249
250 use "Foo::Bar"; # may load /some/trusted/directory/Foo/Bar.pm
251 do "config.pl"; # may load /some/trusted/directory/config.pl
252
253On the other hand, if your script is intended to be run from within
254untrusted directories (such as F</tmp>), then your script suddenly failing
255to load files may be indicative of a security issue. You most likely want
256to replace any relative paths with full paths; for example,
257
258 do "foo_config.pl"
259
260might become
261
262 do "$ENV{HOME}/foo_config.pl"
263
264If you are absolutely certain that you want your script to load and
265execute a file from the current directory, then use a C<./> prefix; for
266example:
267
268 do "./foo_config.pl"
269
270=item * Installing and using CPAN modules
271
272If you install a CPAN module using an automatic tool like C<cpan>, then
273this tool will itself set the C<PERL_USE_UNSAFE_INC> environment variable
274while building and testing the module, which may be sufficient to install
275a distribution which hasn't been updated to be dot-aware. If you want to
276install such a module manually, then you'll need to replace the
277traditional invocation:
278
279 perl Makefile.PL && make && make test && make install
280
281with something like
282
283 (export PERL_USE_UNSAFE_INC=1; \
284 perl Makefile.PL && make && make test && make install)
285
286Note that this only helps build and install an unfixed module. It's
287possible for the tests to pass (since they were run under
288C<PERL_USE_UNSAFE_INC=1>), but for the module itself to fail to perform
289correctly in production. In this case, you may have to temporarily modify
290your script until a fixed version of the module is released.
291For example:
292
293 use Foo::Bar;
294 {
295 local @INC = (@INC, '.');
296 # assuming read_config() needs '.' in @INC
297 $config = Foo::Bar->read_config();
298 }
299
300This is only rarely expected to be necessary. Again, if doing this,
301assess the resultant risks first.
302
303=item * Module Authors
304
305If you maintain a CPAN distribution, it may need updating to run in
306a dotless environment. Although C<cpan> and other such tools will
307currently set the C<PERL_USE_UNSAFE_INC> during module build, this is a
308temporary workaround for the set of modules which rely on C<"."> being in
309C<@INC> for installation and testing, and this may mask deeper issues. It
310could result in a module which passes tests and installs, but which
311fails at run time.
312
313During build, test, and install, it will normally be the case that any perl
314processes will be executing directly within the root directory of the
315untarred distribution, or a known subdirectory of that, such as F<t/>. It
316may well be that F<Makefile.PL> or F<t/foo.t> will attempt to include
317local modules and configuration files using their direct relative
318filenames, which will now fail.
319
320However, as described above, automatic tools like F<cpan> will (for now)
321set the C<PERL_USE_UNSAFE_INC> environment variable, which introduces
322dot during a build.
323
324This makes it likely that your existing build and test code will work, but
325this may mask issues with your code which only manifest when used after
326install. It is prudent to try and run your build process with that
327variable explicitly disabled:
328
329 (export PERL_USE_UNSAFE_INC=0; \
330 perl Makefile.PL && make && make test && make install)
331
332This is more likely to show up any potential problems with your module's
333build process, or even with the module itself. Fixing such issues will
334ensure both that your module can again be installed manually, and that
335it will still build once the C<PERL_USE_UNSAFE_INC> crutch goes away.
336
337When fixing issues in tests due to the removal of dot from C<@INC>,
338reinsertion of dot into C<@INC> should be performed with caution, for this
339too may suppress real errors in your runtime code. You are encouraged
340wherever possible to apply the aforementioned approaches with explicit
341absolute/relative paths, or to relocate your needed files into a
342subdirectory and insert that subdirectory into C<@INC> instead.
343
344If your runtime code has problems under the dotless C<@INC>, then the comments
345above on how to fix for script authors will mostly apply here too. Bear in
346mind though that it is considered bad form for a module to globally add a dot to
347C<@INC>, since it introduces both a security risk and hides issues of
348accidentally requiring dot in C<@INC>, as explained above.
349
350=back
351
352=head2 Escaped colons and relative paths in PATH
353
354On Unix systems, Perl treats any relative paths in the C<PATH> environment
355variable as tainted when starting a new process. Previously, it was
356allowing a backslash to escape a colon (unlike the OS), consequently
357allowing relative paths to be considered safe if the PATH was set to
358something like C</\:.>. The check has been fixed to treat C<"."> as tainted
359in that example.
360
361=head2 New C<-Di> switch is now required for PerlIO debugging output
362
363This is used for debugging of code within PerlIO to avoid recursive
364calls. Previously this output would be sent to the file specified
365by the C<PERLIO_DEBUG> environment variable if perl wasn't running
366setuid and the C<-T> or C<-t> switches hadn't been parsed yet.
367
368If perl performed output at a point where it hadn't yet parsed its
369switches this could result in perl creating or overwriting the file
370named by C<PERLIO_DEBUG> even when the C<-T> switch had been supplied.
371
372Perl now requires the C<-Di> switch to be present before it will produce
373PerlIO debugging
374output. By default this is written to C<stderr>, but can optionally
375be redirected to a file by setting the C<PERLIO_DEBUG> environment
376variable.
377
378If perl is running setuid or the C<-T> switch was supplied,
379C<PERLIO_DEBUG> is ignored and the debugging output is sent to
380C<stderr> as for any other C<-D> switch.
381
382=head1 Incompatible Changes
383
384=head2 Unescaped literal C<"{"> characters in regular expression
385patterns are no longer permissible
386
387You have to now say something like C<"\{"> or C<"[{]"> to specify to
388match a LEFT CURLY BRACKET; otherwise, it is a fatal pattern compilation
389error. This change will allow future extensions to the language.
390
391These have been deprecated since v5.16, with a deprecation message
392raised for some uses starting in v5.22. Unfortunately, the code added
393to raise the message was buggy and failed to warn in some cases where
394it should have. Therefore, enforcement of this ban for these cases is
395deferred until Perl 5.30, but the code has been fixed to raise a
396default-on deprecation message for them in the meantime.
397
398Some uses of literal C<"{"> occur in contexts where we do not foresee
399the meaning ever being anything but the literal, such as the very first
400character in the pattern, or after a C<"|"> meaning alternation. Thus
401
402 qr/{fee|{fie/
403
404matches either of the strings C<{fee> or C<{fie>. To avoid forcing
405unnecessary code changes, these uses do not need to be escaped, and no
406warning is raised about them, and there are no current plans to change this.
407
408But it is always correct to escape C<"{">, and the simple rule to
409remember is to always do so.
410
411See L<Unescaped left brace in regex is illegal here|perldiag/Unescaped left brace in regex is illegal here in regex; marked by S<E<lt>-- HERE> in mE<sol>%sE<sol>>.
412
413=head2 C<scalar(%hash)> return signature changed
414
415The value returned for C<scalar(%hash)> will no longer show information about
416the buckets allocated in the hash. It will simply return the count of used
417keys. It is thus equivalent to C<0+keys(%hash)>.
418
419A form of backward compatibility is provided via
420L<C<Hash::Util::bucket_ratio()>|Hash::Util/bucket_ratio> which provides
421the same behavior as
422C<scalar(%hash)> provided in Perl 5.24 and earlier.
423
424=head2 C<keys> returned from an lvalue subroutine
425
426C<keys> returned from an lvalue subroutine can no longer be assigned
427to in list context.
428
429 sub foo : lvalue { keys(%INC) }
430 (foo) = 3; # death
431 sub bar : lvalue { keys(@_) }
432 (bar) = 3; # also an error
433
434This makes the lvalue sub case consistent with C<(keys %hash) = ...> and
435C<(keys @_) = ...>, which are also errors.
436L<[perl #128187]|https://rt.perl.org/Public/Bug/Display.html?id=128187>
437
438=head2 The C<${^ENCODING}> facility has been removed
439
440The special behaviour associated with assigning a value to this variable
441has been removed. As a consequence, the L<encoding> pragma's default mode
442is no longer supported. If
443you still need to write your source code in encodings other than UTF-8, use a
444source filter such as L<Filter::Encoding> on CPAN or L<encoding>'s C<Filter>
445option.
446
447=head2 C<POSIX::tmpnam()> has been removed
448
449The fundamentally unsafe C<tmpnam()> interface was deprecated in
450Perl 5.22 and has now been removed. In its place, you can use,
451for example, the L<File::Temp> interfaces.
452
453=head2 require ::Foo::Bar is now illegal.
454
455Formerly, C<require ::Foo::Bar> would try to read F</Foo/Bar.pm>. Now any
456bareword require which starts with a double colon dies instead.
457
458=head2 Literal control character variable names are no longer permissible
459
460A variable name may no longer contain a literal control character under
461any circumstances. These previously were allowed in single-character
462names on ASCII platforms, but have been deprecated there since Perl
4635.20. This affects things like C<$I<\cT>>, where I<\cT> is a literal
464control (such as a C<NAK> or C<NEGATIVE ACKNOWLEDGE> character) in the
465source code.
466
467=head2 C<NBSP> is no longer permissible in C<\N{...}>
468
469The name of a character may no longer contain non-breaking spaces. It
470has been deprecated to do so since Perl 5.22.
471
472=head1 Deprecations
473
474=head2 String delimiters that aren't stand-alone graphemes are now deprecated
475
476For Perl to eventually allow string delimiters to be Unicode
477grapheme clusters (which look like a single character, but may be
478a sequence of several ones), we have to stop allowing a single character
479delimiter that isn't a grapheme by itself. These are unlikely to exist
480in actual code, as they would typically display as attached to the
481character in front of them.
482
483=head2 C<\cI<X>> that maps to a printable is no longer deprecated
484
485This means we have no plans to remove this feature. It still raises a
486warning, but only if syntax warnings are enabled. The feature was
487originally intended to be a way to express non-printable characters that
488don't have a mnemonic (C<\t> and C<\n> are mnemonics for two
489non-printable characters, but most non-printables don't have a
490mnemonic.) But the feature can be used to specify a few printable
491characters, though those are more clearly expressed as the printable
492itself. See
493L<http://www.nntp.perl.org/group/perl.perl5.porters/2017/02/msg242944.html>.
494
495=head1 Performance Enhancements
496
497=over 4
498
499=item *
500
501A hash in boolean context is now sometimes faster, I<e.g.>
502
503 if (!%h) { ... }
504
505This was already special-cased, but some cases were missed (such as
506C<grep %$_, @AoH>), and even the ones which weren't have been improved.
507
508=item * New Faster Hash Function on 64 bit builds
509
510We use a different hash function for short and long keys. This should
511improve performance and security, especially for long keys.
512
513=item * readline is faster
514
515Reading from a file line-by-line with C<readline()> or C<< E<lt>E<gt> >> should
516now typically be faster due to a better implementation of the code that
517searches for the next newline character.
518
519=item *
520
521Assigning one reference to another, I<e.g.> C<$ref1 = $ref2> has been
522optimized in some cases.
523
524=item *
525
526Remove some exceptions to creating Copy-on-Write strings. The string
527buffer growth algorithm has been slightly altered so that you're less
528likely to encounter a string which can't be COWed.
529
530=item *
531
532Better optimise array and hash assignment: where an array or hash appears
533in the LHS of a list assignment, such as C<(..., @a) = (...);>, it's
534likely to be considerably faster, especially if it involves emptying the
535array/hash. For example, this code runs about a third faster compared to
536Perl 5.24.0:
537
538 my @a;
539 for my $i (1..10_000_000) {
540 @a = (1,2,3);
541 @a = ();
542 }
543
544=item *
545
546Converting a single-digit string to a number is now substantially faster.
547
548=item *
549
550The C<split> builtin is now slightly faster in many cases: in particular
551for the two specially-handled forms
552
553 my @a = split ...;
554 local @a = split ...;
555
556=item *
557
558The rather slow implementation for the experimental subroutine signatures
559feature has been made much faster; it is now comparable in speed with the
560traditional C<my ($a, $b, @c) = @_>.
561
562=item *
563
564Bareword constant strings are now permitted to take part in constant
565folding. They were originally exempted from constant folding in August 1999,
566during the development of Perl 5.6, to ensure that C<use strict "subs">
567would still apply to bareword constants. That has now been accomplished a
568different way, so barewords, like other constants, now gain the performance
569benefits of constant folding.
570
571This also means that void-context warnings on constant expressions of
572barewords now report the folded constant operand, rather than the operation;
573this matches the behaviour for non-bareword constants.
574
575=back
576
577=head1 Modules and Pragmata
578
579=head2 Updated Modules and Pragmata
580
581=over 4
582
583=item *
584
585IO::Compress has been upgraded from version 2.069 to 2.074.
586
587=item *
588
589L<Archive::Tar> has been upgraded from version 2.04 to 2.24.
590
591=item *
592
593L<arybase> has been upgraded from version 0.11 to 0.12.
594
595=item *
596
597L<attributes> has been upgraded from version 0.27 to 0.29.
598
599The deprecation message for the C<:unique> and C<:locked> attributes
600now mention that they will disappear in Perl 5.28.
601
602=item *
603
604L<B> has been upgraded from version 1.62 to 1.68.
605
606=item *
607
608L<B::Concise> has been upgraded from version 0.996 to 0.999.
609
610Its output is now more descriptive for C<op_private> flags.
611
612=item *
613
614L<B::Debug> has been upgraded from version 1.23 to 1.24.
615
616=item *
617
618L<B::Deparse> has been upgraded from version 1.37 to 1.40.
619
620=item *
621
622L<B::Xref> has been upgraded from version 1.05 to 1.06.
623
624It now uses 3-arg C<open()> instead of 2-arg C<open()>.
625L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
626
627=item *
628
629L<base> has been upgraded from version 2.23 to 2.25.
630
631=item *
632
633L<bignum> has been upgraded from version 0.42 to 0.47.
634
635=item *
636
637L<Carp> has been upgraded from version 1.40 to 1.42.
638
639=item *
640
641L<charnames> has been upgraded from version 1.43 to 1.44.
642
643=item *
644
645L<Compress::Raw::Bzip2> has been upgraded from version 2.069 to 2.074.
646
647=item *
648
649L<Compress::Raw::Zlib> has been upgraded from version 2.069 to 2.074.
650
651=item *
652
653L<Config::Perl::V> has been upgraded from version 0.25 to 0.28.
654
655=item *
656
657L<CPAN> has been upgraded from version 2.11 to 2.18.
658
659=item *
660
661L<CPAN::Meta> has been upgraded from version 2.150005 to 2.150010.
662
663=item *
664
665L<Data::Dumper> has been upgraded from version 2.160 to 2.167.
666
667The XS implementation now supports Deparse.
668
669=item *
670
671L<DB_File> has been upgraded from version 1.835 to 1.840.
672
673=item *
674
675L<Devel::Peek> has been upgraded from version 1.23 to 1.26.
676
677=item *
678
679L<Devel::PPPort> has been upgraded from version 3.32 to 3.35.
680
681=item *
682
683L<Devel::SelfStubber> has been upgraded from version 1.05 to 1.06.
684
685It now uses 3-arg C<open()> instead of 2-arg C<open()>.
686L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
687
688=item *
689
690L<diagnostics> has been upgraded from version 1.34 to 1.36.
691
692It now uses 3-arg C<open()> instead of 2-arg C<open()>.
693L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
694
695=item *
696
697L<Digest> has been upgraded from version 1.17 to 1.17_01.
698
699=item *
700
701L<Digest::MD5> has been upgraded from version 2.54 to 2.55.
702
703=item *
704
705L<Digest::SHA> has been upgraded from version 5.95 to 5.96.
706
707=item *
708
709L<DynaLoader> has been upgraded from version 1.38 to 1.42.
710
711=item *
712
713L<Encode> has been upgraded from version 2.80 to 2.88.
714
715=item *
716
717L<encoding> has been upgraded from version 2.17 to 2.19.
718
719This module's default mode is no longer supported. It now
720dies when imported, unless the C<Filter> option is being used.
721
722=item *
723
724L<encoding::warnings> has been upgraded from version 0.12 to 0.13.
725
726This module is no longer supported. It emits a warning to
727that effect and then does nothing.
728
729=item *
730
731L<Errno> has been upgraded from version 1.25 to 1.28.
732
733It now documents that using C<%!> automatically loads Errno for you.
734
735It now uses 3-arg C<open()> instead of 2-arg C<open()>.
736L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
737
738=item *
739
740L<ExtUtils::Embed> has been upgraded from version 1.33 to 1.34.
741
742It now uses 3-arg C<open()> instead of 2-arg C<open()>.
743L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
744
745=item *
746
747L<ExtUtils::MakeMaker> has been upgraded from version 7.10_01 to 7.24.
748
749=item *
750
751L<ExtUtils::Miniperl> has been upgraded from version 1.05 to 1.06.
752
753=item *
754
755L<ExtUtils::ParseXS> has been upgraded from version 3.31 to 3.34.
756
757=item *
758
759L<ExtUtils::Typemaps> has been upgraded from version 3.31 to 3.34.
760
761=item *
762
763L<feature> has been upgraded from version 1.42 to 1.47.
764
765=item *
766
767L<File::Copy> has been upgraded from version 2.31 to 2.32.
768
769=item *
770
771L<File::Fetch> has been upgraded from version 0.48 to 0.52.
772
773=item *
774
775L<File::Glob> has been upgraded from version 1.26 to 1.28.
776
777It now Issues a deprecation message for C<File::Glob::glob()>.
778
779=item *
780
781L<File::Spec> has been upgraded from version 3.63 to 3.67.
782
783=item *
784
785L<FileHandle> has been upgraded from version 2.02 to 2.03.
786
787=item *
788
789L<Filter::Simple> has been upgraded from version 0.92 to 0.93.
790
791It no longer treats C<no MyFilter> immediately following C<use MyFilter> as
792end-of-file.
793L<[perl #107726]|https://rt.perl.org/Public/Bug/Display.html?id=107726>
794
795=item *
796
797L<Getopt::Long> has been upgraded from version 2.48 to 2.49.
798
799=item *
800
801L<Getopt::Std> has been upgraded from version 1.11 to 1.12.
802
803=item *
804
805L<Hash::Util> has been upgraded from version 0.19 to 0.22.
806
807=item *
808
809L<HTTP::Tiny> has been upgraded from version 0.056 to 0.070.
810
811Internal 599-series errors now include the redirect history.
812
813=item *
814
815L<I18N::LangTags> has been upgraded from version 0.40 to 0.42.
816
817It now uses 3-arg C<open()> instead of 2-arg C<open()>.
818L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
819
820=item *
821
822L<IO> has been upgraded from version 1.36 to 1.38.
823
824=item *
825
826L<IO::Socket::IP> has been upgraded from version 0.37 to 0.38.
827
828=item *
829
830L<IPC::Cmd> has been upgraded from version 0.92 to 0.96.
831
832=item *
833
834L<IPC::SysV> has been upgraded from version 2.06_01 to 2.07.
835
836=item *
837
838L<JSON::PP> has been upgraded from version 2.27300 to 2.27400_02.
839
840=item *
841
842L<lib> has been upgraded from version 0.63 to 0.64.
843
844It now uses 3-arg C<open()> instead of 2-arg C<open()>.
845L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
846
847=item *
848
849L<List::Util> has been upgraded from version 1.42_02 to 1.46_02.
850
851=item *
852
853L<Locale::Codes> has been upgraded from version 3.37 to 3.42.
854
855=item *
856
857L<Locale::Maketext> has been upgraded from version 1.26 to 1.28.
858
859=item *
860
861L<Locale::Maketext::Simple> has been upgraded from version 0.21 to 0.21_01.
862
863=item *
864
865L<Math::BigInt> has been upgraded from version 1.999715 to 1.999806.
866
867=item *
868
869L<Math::BigInt::FastCalc> has been upgraded from version 0.40 to 0.5005.
870
871=item *
872
873L<Math::BigRat> has been upgraded from version 0.260802 to 0.2611.
874
875=item *
876
877L<Math::Complex> has been upgraded from version 1.59 to 1.5901.
878
879=item *
880
881L<Memoize> has been upgraded from version 1.03 to 1.03_01.
882
883=item *
884
885L<Module::CoreList> has been upgraded from version 5.20170420 to 5.20170530.
886
887=item *
888
889L<Module::Load::Conditional> has been upgraded from version 0.64 to 0.68.
890
891=item *
892
893L<Module::Metadata> has been upgraded from version 1.000031 to 1.000033.
894
895=item *
896
897L<mro> has been upgraded from version 1.18 to 1.20.
898
899=item *
900
901L<Net::Ping> has been upgraded from version 2.43 to 2.55.
902
903IPv6 addresses and C<AF_INET6> sockets are now supported, along with several
904other enhancements.
905
906=item *
907
908L<NEXT> has been upgraded from version 0.65 to 0.67.
909
910=item *
911
912L<Opcode> has been upgraded from version 1.34 to 1.39.
913
914=item *
915
916L<open> has been upgraded from version 1.10 to 1.11.
917
918=item *
919
920L<OS2::Process> has been upgraded from version 1.11 to 1.12.
921
922It now uses 3-arg C<open()> instead of 2-arg C<open()>.
923L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
924
925=item *
926
927L<overload> has been upgraded from version 1.26 to 1.28.
928
929Its compilation speed has been improved slightly.
930
931=item *
932
933L<parent> has been upgraded from version 0.234 to 0.236.
934
935=item *
936
937L<perl5db.pl> has been upgraded from version 1.50 to 1.51.
938
939It now ignores F</dev/tty> on non-Unix systems.
940L<[perl #113960]|https://rt.perl.org/Public/Bug/Display.html?id=113960>
941
942=item *
943
944L<Perl::OSType> has been upgraded from version 1.009 to 1.010.
945
946=item *
947
948L<perlfaq> has been upgraded from version 5.021010 to 5.021011.
949
950=item *
951
952L<PerlIO> has been upgraded from version 1.09 to 1.10.
953
954=item *
955
956L<PerlIO::encoding> has been upgraded from version 0.24 to 0.25.
957
958=item *
959
960L<PerlIO::scalar> has been upgraded from version 0.24 to 0.26.
961
962=item *
963
964L<Pod::Checker> has been upgraded from version 1.60 to 1.73.
965
966=item *
967
968L<Pod::Functions> has been upgraded from version 1.10 to 1.11.
969
970=item *
971
972L<Pod::Html> has been upgraded from version 1.22 to 1.2202.
973
974=item *
975
976L<Pod::Perldoc> has been upgraded from version 3.25_02 to 3.28.
977
978=item *
979
980L<Pod::Simple> has been upgraded from version 3.32 to 3.35.
981
982=item *
983
984L<Pod::Usage> has been upgraded from version 1.68 to 1.69.
985
986=item *
987
988L<POSIX> has been upgraded from version 1.65 to 1.76.
989
990This remedies several defects in making its symbols exportable.
991L<[perl #127821]|https://rt.perl.org/Public/Bug/Display.html?id=127821>
992
993The C<POSIX::tmpnam()> interface has been removed,
994see L</"POSIX::tmpnam() has been removed">.
995
996The following deprecated functions have been removed:
997
998 POSIX::isalnum
999 POSIX::isalpha
1000 POSIX::iscntrl
1001 POSIX::isdigit
1002 POSIX::isgraph
1003 POSIX::islower
1004 POSIX::isprint
1005 POSIX::ispunct
1006 POSIX::isspace
1007 POSIX::isupper
1008 POSIX::isxdigit
1009 POSIX::tolower
1010 POSIX::toupper
1011
1012Trying to import POSIX subs that have no real implementations
1013(like C<POSIX::atend()>) now fails at import time, instead of
1014waiting until runtime.
1015
1016=item *
1017
1018L<re> has been upgraded from version 0.32 to 0.34
1019
1020This adds support for the new L<C<E<47>xx>|perlre/E<sol>x and E<sol>xx>
1021regular expression pattern modifier, and a change to the L<S<C<use re
1022'strict'>>|re/'strict' mode> experimental feature. When S<C<re
1023'strict'>> is enabled, a warning now will be generated for all
1024unescaped uses of the two characters C<"}"> and C<"]"> in regular
1025expression patterns (outside bracketed character classes) that are taken
1026literally. This brings them more in line with the C<")"> character which
1027is always a metacharacter unless escaped. Being a metacharacter only
1028sometimes, depending on an action at a distance, can lead to silently
1029having the pattern mean something quite different than was intended,
1030which the S<C<re 'strict'>> mode is intended to minimize.
1031
1032=item *
1033
1034L<Safe> has been upgraded from version 2.39 to 2.40.
1035
1036=item *
1037
1038L<Scalar::Util> has been upgraded from version 1.42_02 to 1.46_02.
1039
1040=item *
1041
1042L<Storable> has been upgraded from version 2.56 to 2.62.
1043
1044Fixes
1045L<[perl #130098]|https://rt.perl.org/Public/Bug/Display.html?id=130098>.
1046
1047=item *
1048
1049L<Symbol> has been upgraded from version 1.07 to 1.08.
1050
1051=item *
1052
1053L<Sys::Syslog> has been upgraded from version 0.33 to 0.35.
1054
1055=item *
1056
1057L<Term::ANSIColor> has been upgraded from version 4.04 to 4.06.
1058
1059=item *
1060
1061L<Term::ReadLine> has been upgraded from version 1.15 to 1.16.
1062
1063It now uses 3-arg C<open()> instead of 2-arg C<open()>.
1064L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
1065
1066=item *
1067
1068L<Test> has been upgraded from version 1.28 to 1.30.
1069
1070It now uses 3-arg C<open()> instead of 2-arg C<open()>.
1071L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
1072
1073=item *
1074
1075L<Test::Harness> has been upgraded from version 3.36 to 3.38.
1076
1077=item *
1078
1079L<Test::Simple> has been upgraded from version 1.001014 to 1.302073.
1080
1081=item *
1082
1083L<Thread::Queue> has been upgraded from version 3.09 to 3.12.
1084
1085=item *
1086
1087L<Thread::Semaphore> has been upgraded from 2.12 to 2.13.
1088
1089Added the C<down_timed> method.
1090
1091=item *
1092
1093L<threads> has been upgraded from version 2.07 to 2.15.
1094
1095=item *
1096
1097L<threads::shared> has been upgraded from version 1.51 to 1.56.
1098
1099=item *
1100
1101L<Tie::Hash::NamedCapture> has been upgraded from version 0.09 to 0.10.
1102
1103=item *
1104
1105L<Time::HiRes> has been upgraded from version 1.9733 to 1.9741.
1106
1107It now builds on systems with C++11 compilers (such as G++ 6 and Clang++
11083.9).
1109
1110Now uses C<clockid_t>.
1111
1112=item *
1113
1114L<Time::Local> has been upgraded from version 1.2300 to 1.25.
1115
1116=item *
1117
1118L<Unicode::Collate> has been upgraded from version 1.14 to 1.19.
1119
1120=item *
1121
1122L<Unicode::UCD> has been upgraded from version 0.64 to 0.68.
1123
1124It now uses 3-arg C<open()> instead of 2-arg C<open()>.
1125L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
1126
1127=item *
1128
1129L<version> has been upgraded from version 0.9916 to 0.9917.
1130
1131=item *
1132
1133L<VMS::DCLsym> has been upgraded from version 1.06 to 1.08.
1134
1135It now uses 3-arg C<open()> instead of 2-arg C<open()>.
1136L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
1137
1138=item *
1139
1140L<warnings> has been upgraded from version 1.36 to 1.37.
1141
1142=item *
1143
1144L<XS::Typemap> has been upgraded from version 0.14 to 0.15.
1145
1146=item *
1147
1148L<XSLoader> has been upgraded from version 0.21 to 0.27.
1149
1150Fixed a security hole in which binary files could be loaded from a path
1151outside of L<C<@INC>|perlvar/@INC>.
1152
1153It now uses 3-arg C<open()> instead of 2-arg C<open()>.
1154L<[perl #130122]|https://rt.perl.org/Public/Bug/Display.html?id=130122>
1155
1156=back
1157
1158=head1 Documentation
1159
1160=head2 New Documentation
1161
1162=head3 L<perldeprecation>
1163
1164This file documents all upcoming deprecations, and some of the deprecations
1165which already have been removed. The purpose of this documentation is
1166two-fold: document what will disappear, and by which version, and serve
1167as a guide for people dealing with code which has features that no longer
1168work after an upgrade of their perl.
1169
1170=head2 Changes to Existing Documentation
1171
1172We have attempted to update the documentation to reflect the changes
1173listed in this document. If you find any we have missed, send email to
1174L<perlbug@perl.org|mailto:perlbug@perl.org>.
1175
1176Additionally, all references to Usenet have been removed, and the
1177following selected changes have been made:
1178
1179=head3 L<perlfunc>
1180
1181=over 4
1182
1183=item *
1184
1185Removed obsolete text about L<C<defined()>|perlfunc/defined>
1186on aggregates that should have been deleted earlier, when the feature
1187was removed.
1188
1189=item *
1190
1191Corrected documentation of L<C<eval()>|perlfunc/eval>,
1192and L<C<evalbytes()>|perlfunc/evalbytes>.
1193
1194=item *
1195
1196Clarified documentation of L<C<seek()>|perlfunc/seek>,
1197L<C<tell()>|perlfunc/tell> and L<C<sysseek()>|perlfunc/sysseek>
1198emphasizing that positions are in bytes and not characters.
1199L<[perl #128607]|https://rt.perl.org/Public/Bug/Display.html?id=128607>
1200
1201=item *
1202
1203Clarified documentation of L<C<sort()>|perlfunc/sort LIST> concerning
1204the variables C<$a> and C<$b>.
1205
1206=item *
1207
1208In L<C<split()>|perlfunc/split> noted that certain pattern modifiers are
1209legal, and added a caution about its use in Perls before v5.11.
1210
1211=item *
1212
1213Removed obsolete documentation of L<C<study()>|perlfunc/study>, noting
1214that it is now a no-op.
1215
1216=item *
1217
1218Noted that L<C<vec()>|perlfunc/vec> doesn't work well when the string
1219contains characters whose code points are above 255.
1220
1221=back
1222
1223=head3 L<perlguts>
1224
1225=over 4
1226
1227=item *
1228
1229Added advice on
1230L<formatted printing of operands of C<Size_t> and C<SSize_t>|perlguts/Formatted Printing of Size_t and SSize_t>
1231
1232=back
1233
1234=head3 L<perlhack>
1235
1236=over 4
1237
1238=item *
1239
1240Clarify what editor tab stop rules to use, and note that we are
1241migrating away from using tabs, replacing them with sequences of SPACE
1242characters.
1243
1244=back
1245
1246=head3 L<perlhacktips>
1247
1248=over 4
1249
1250=item *
1251
1252Give another reason to use C<cBOOL> to cast an expression to boolean.
1253
1254=item *
1255
1256Note that the macros C<TRUE> and C<FALSE> are available to express
1257boolean values.
1258
1259=back
1260
1261=head3 L<perlinterp>
1262
1263=over 4
1264
1265=item *
1266
1267L<perlinterp> has been expanded to give a more detailed example of how to
1268hunt around in the parser for how a given operator is handled.
1269
1270=back
1271
1272=head3 L<perllocale>
1273
1274=over 4
1275
1276=item *
1277
1278Some locales aren't compatible with Perl. Note that these can cause
1279core dumps.
1280
1281=back
1282
1283=head3 L<perlmod>
1284
1285=over 4
1286
1287=item *
1288
1289Various clarifications have been added.
1290
1291=back
1292
1293=head3 L<perlmodlib>
1294
1295=over 4
1296
1297=item *
1298
1299Updated the site mirror list.
1300
1301=back
1302
1303=head3 L<perlobj>
1304
1305=over 4
1306
1307=item *
1308
1309Added a section on calling methods using their fully qualified names.
1310
1311=item *
1312
1313Do not discourage manual C<@ISA>.
1314
1315=back
1316
1317=head3 L<perlootut>
1318
1319=over 4
1320
1321=item *
1322
1323Mention C<Moo> more.
1324
1325=back
1326
1327=head3 L<perlop>
1328
1329=over 4
1330
1331=item *
1332
1333Note that white space must be used for quoting operators if the
1334delimiter is a word character (I<i.e.>, matches C<\w>).
1335
1336=item *
1337
1338Clarify that in regular expression patterns delimited by single quotes,
1339no variable interpolation is done.
1340
1341=back
1342
1343=head3 L<perlre>
1344
1345=over 4
1346
1347=item *
1348
1349The first part was extensively rewritten to incorporate various basic
1350points, that in earlier versions were mentioned in sort of an appendix
1351on Version 8 regular expressions.
1352
1353=item *
1354
1355Note that it is common to have the C</x> modifier and forget that this
1356means that C<"#"> has to be escaped.
1357
1358=back
1359
1360=head3 L<perlretut>
1361
1362=over 4
1363
1364=item *
1365
1366Add introductory material.
1367
1368=item *
1369
1370Note that a metacharacter occurring in a context where it can't mean
1371that, silently loses its meta-ness and matches literally.
1372L<C<use re 'strict'>|re/'strict' mode> can catch some of these.
1373
1374=back
1375
1376=head3 L<perlunicode>
1377
1378=over 4
1379
1380=item *
1381
1382Corrected the text about Unicode BYTE ORDER MARK handling.
1383
1384=item *
1385
1386Updated the text to correspond with changes in Unicode UTS#18, concerning
1387regular expressions, and Perl compatibility with what it says.
1388
1389=back
1390
1391=head3 L<perlvar>
1392
1393=over 4
1394
1395=item *
1396
1397Document C<@ISA>. It was documented in other places, but not in L<perlvar>.
1398
1399=back
1400
1401=head1 Diagnostics
1402
1403=head2 New Diagnostics
1404
1405=head3 New Errors
1406
1407=over 4
1408
1409=item *
1410
1411L<A signature parameter must start with C<'$'>, C<'@'> or C<'%'>
1412|perldiag/A signature parameter must start with C<'$'>, C<'@'> or C<'%'>>
1413
1414=item *
1415
1416L<Bareword in require contains "%s"|perldiag/"Bareword in require contains "%s"">
1417
1418=item *
1419
1420L<Bareword in require maps to empty filename|perldiag/"Bareword in require maps to empty filename">
1421
1422=item *
1423
1424L<Bareword in require maps to disallowed filename "%s"|perldiag/"Bareword in require maps to disallowed filename "%s"">
1425
1426=item *
1427
1428L<Bareword in require must not start with a double-colon: "%s"|perldiag/"Bareword in require must not start with a double-colon: "%s"">
1429
1430=item *
1431
1432L<%s: command not found|perldiag/"%s: command not found">
1433
1434(A) You've accidentally run your script through B<bash> or another shell
1435instead of Perl. Check the C<#!> line, or manually feed your script into
1436Perl yourself. The C<#!> line at the top of your file could look like:
1437
1438 #!/usr/bin/perl
1439
1440=item *
1441
1442L<%s: command not found: %s|perldiag/"%s: command not found: %s">
1443
1444(A) You've accidentally run your script through B<zsh> or another shell
1445instead of Perl. Check the C<#!> line, or manually feed your script into
1446Perl yourself. The C<#!> line at the top of your file could look like:
1447
1448 #!/usr/bin/perl
1449
1450=item *
1451
1452L<The experimental declared_refs feature is not enabled|perldiag/"The experimental declared_refs feature is not enabled">
1453
1454(F) To declare references to variables, as in C<my \%x>, you must first enable
1455the feature:
1456
1457 no warnings "experimental::declared_refs";
1458 use feature "declared_refs";
1459
1460See L</Declaring a reference to a variable>.
1461
1462=item *
1463
1464L<Illegal character following sigil in a subroutine signature
1465|perldiag/Illegal character following sigil in a subroutine signature>
1466
1467=item *
1468
1469L<Indentation on line %d of here-doc doesn't match delimiter
1470|perldiag/Indentation on line %d of here-doc doesn't match delimiter>
1471
1472=item *
1473
1474L<Infinite recursion via empty pattern|perldiag/"Infinite recursion via empty pattern">.
1475
1476Using the empty pattern (which re-executes the last successfully-matched
1477pattern) inside a code block in another regex, as in C</(?{ s!!new! })/>, has
1478always previously yielded a segfault. It now produces this error.
1479
1480=item *
1481
1482L<Malformed UTF-8 string in "%s"
1483|perldiag/Malformed UTF-8 string in "%s">
1484
1485=item *
1486
1487L<Multiple slurpy parameters not allowed
1488|perldiag/Multiple slurpy parameters not allowed>
1489
1490=item *
1491
1492L<C<'#'> not allowed immediately following a sigil in a subroutine signature
1493|perldiag/C<'#'> not allowed immediately following a sigil in a subroutine signature>
1494
1495=item *
1496
1497L<panic: unknown OA_*: %x
1498|perldiag/panic: unknown OA_*: %x>
1499
1500=item *
1501
1502L<Unescaped left brace in regex is illegal here|perldiag/Unescaped left brace in regex is illegal here in regex; marked by S<E<lt>-- HERE> in mE<sol>%sE<sol>>
1503
1504Unescaped left braces are now illegal in some contexts in regular expression
1505patterns. In other contexts, they are still just deprecated; they will
1506be illegal in Perl 5.30.
1507
1508=item *
1509
1510L<Version control conflict marker|perldiag/"Version control conflict marker">
1511
1512(F) The parser found a line starting with C<E<lt>E<lt>E<lt>E<lt>E<lt>E<lt>E<lt>>,
1513C<E<gt>E<gt>E<gt>E<gt>E<gt>E<gt>E<gt>>, or C<=======>. These may be left by a
1514version control system to mark conflicts after a failed merge operation.
1515
1516=back
1517
1518=head3 New Warnings
1519
1520=over 4
1521
1522=item *
1523
1524L<Can't determine class of operator %s, assuming C<BASEOP>
1525|perldiag/Can't determine class of operator %s, assuming C<BASEOP>>
1526
1527=item *
1528
1529L<Declaring references is experimental|perldiag/"Declaring references is experimental">
1530
1531(S experimental::declared_refs) This warning is emitted if you use a reference
1532constructor on the right-hand side of C<my()>, C<state()>, C<our()>, or
1533C<local()>. Simply suppress the warning if you want to use the feature, but
1534know that in doing so you are taking the risk of using an experimental feature
1535which may change or be removed in a future Perl version:
1536
1537 no warnings "experimental::declared_refs";
1538 use feature "declared_refs";
1539 $fooref = my \$foo;
1540
1541See L</Declaring a reference to a variable>.
1542
1543=item *
1544
1545L<do "%s" failed, '.' is no longer in @INC|perldiag/do "%s" failed, '.' is no longer in @INC; did you mean do ".E<sol>%s"?>
1546
1547Since C<"."> is now removed from C<@INC> by default, C<do> will now trigger a warning recommending to fix the C<do> statement.
1548
1549=item *
1550
1551L<C<File::Glob::glob()> will disappear in perl 5.30. Use C<File::Glob::bsd_glob()> instead.
1552|perldiag/C<File::Glob::glob()> will disappear in perl 5.30. Use C<File::Glob::bsd_glob()> instead.>
1553
1554=item *
1555
1556L<Unescaped literal '%c' in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>
1557|perldiag/Unescaped literal '%c' in regex; marked by <-- HERE in mE<sol>%sE<sol>>
1558
1559=item *
1560
1561L<Use of unassigned code point or non-standalone grapheme for a delimiter will be a fatal error starting in Perl 5.30|perldiag/"Use of unassigned code point or non-standalone grapheme for a delimiter will be a fatal error starting in Perl 5.30">
1562
1563See L</Deprecations>
1564
1565=back
1566
1567=head2 Changes to Existing Diagnostics
1568
1569=over 4
1570
1571=item *
1572
1573When a C<require> fails, we now do not provide C<@INC> when the C<require>
1574is for a file instead of a module.
1575
1576=item *
1577
1578When C<@INC> is not scanned for a C<require> call, we no longer display
1579C<@INC> to avoid confusion.
1580
1581=item *
1582
1583L<Attribute "locked" is deprecated, and will disappear in Perl 5.28
1584|perldiag/Attribute "locked" is deprecated, and will disappear in Perl 5.28>
1585
1586This existing warning has had the I<and will disappear> text added in this
1587release.
1588
1589=item *
1590
1591L<Attribute "unique" is deprecated, and will disappear in Perl 5.28
1592|perldiag/Attribute "unique" is deprecated, and will disappear in Perl 5.28>
1593
1594This existing warning has had the I<and will disappear> text added in this
1595release.
1596
1597=item *
1598
1599Calling POSIX::%s() is deprecated
1600
1601This warning has been removed, as the deprecated functions have been
1602removed from POSIX.
1603
1604=item *
1605
1606L<Constants from lexical variables potentially modified elsewhere are deprecated. This will not be allowed in Perl 5.32
1607|perldiag/Constants from lexical variables potentially modified elsewhere are deprecated. This will not be allowed in Perl 5.32>
1608
1609This existing warning has had the I<this will not be allowed> text added
1610in this release.
1611
1612=item *
1613
1614L<Deprecated use of C<my()> in false conditional. This will be a fatal error in Perl 5.30
1615|perldiag/Deprecated use of C<my()> in false conditional. This will be a fatal error in Perl 5.30>
1616
1617This existing warning has had the I<this will be a fatal error> text added
1618in this release.
1619
1620=item *
1621
1622L<C<dump()> better written as C<CORE::dump()>. C<dump()> will no longer be available in Perl 5.30
1623|perldiag/C<dump()> better written as C<CORE::dump()>. C<dump()> will no longer be available in Perl 5.30>
1624
1625This existing warning has had the I<no longer be available> text added in
1626this release.
1627
1628=item *
1629
1630L<Experimental %s on scalar is now forbidden
1631|perldiag/Experimental %s on scalar is now forbidden>
1632
1633This message is now followed by more helpful text.
1634L<[perl #127976]|https://rt.perl.org/Public/Bug/Display.html?id=127976>
1635
1636=item *
1637
1638Experimental "%s" subs not enabled
1639
1640This warning was been removed, as lexical subs are no longer experimental.
1641
1642=item *
1643
1644Having more than one /%c regexp modifier is deprecated
1645
1646This deprecation warning has been removed, since C</xx> now has a new
1647meaning.
1648
1649=item *
1650
1651L<%s() is deprecated on C<:utf8> handles. This will be a fatal error in Perl 5.30
1652|perldiag/%s() is deprecated on C<:utf8> handles. This will be a fatal error in Perl 5.30>.
1653
1654where "%s" is one of C<sysread>, C<recv>, C<syswrite>, or C<send>.
1655
1656This existing warning has had the I<this will be a fatal error> text added
1657in this release.
1658
1659This warning is now enabled by default, as all C<deprecated> category
1660warnings should be.
1661
1662=item *
1663
1664L<C<$*> is no longer supported. Its use will be fatal in Perl 5.30
1665|perldiag/C<$*> is no longer supported. Its use will be fatal in Perl 5.30>
1666
1667This existing warning has had the I<its use will be fatal> text added in
1668this release.
1669
1670=item *
1671
1672L<C<$#> is no longer supported. Its use will be fatal in Perl 5.30
1673|perldiag/C<$#> is no longer supported. Its use will be fatal in Perl 5.30>
1674
1675This existing warning has had the I<its use will be fatal> text added in
1676this release.
1677
1678=item *
1679
1680L<Malformed UTF-8 character%s
1681|perldiag/Malformed UTF-8 character%s>
1682
1683Details as to the exact problem have been added at the end of this
1684message
1685
1686=item *
1687
1688L<Missing or undefined argument to %s
1689|perldiag/Missing or undefined argument to %s>
1690
1691This warning used to warn about C<require>, even if it was actually C<do>
1692which being executed. It now gets the operation name right.
1693
1694=item *
1695
1696NO-BREAK SPACE in a charnames alias definition is deprecated
1697
1698This warning has been removed as the behavior is now an error.
1699
1700=item *
1701
1702L<Odd nameE<sol>value argument for subroutine '%s'
1703|perldiag/"Odd nameE<sol>value argument for subroutine '%s'">
1704
1705This warning now includes the name of the offending subroutine.
1706
1707=item *
1708
1709L<Opening dirhandle %s also as a file. This will be a fatal error in Perl 5.28
1710|perldiag/Opening dirhandle %s also as a file. This will be a fatal error in Perl 5.28>
1711
1712This existing warning has had the I<this will be a fatal error> text added
1713in this release.
1714
1715=item *
1716
1717L<Opening filehandle %s also as a directory. This will be a fatal error in Perl 5.28
1718|perldiag/Opening filehandle %s also as a directory. This will be a fatal error in Perl 5.28>
1719
1720This existing warning has had the I<this will be a fatal error> text added
1721in this release.
1722
1723=item *
1724
1725panic: ck_split, type=%u
1726
1727panic: pp_split, pm=%p, s=%p
1728
1729These panic errors have been removed.
1730
1731=item *
1732
1733Passing malformed UTF-8 to "%s" is deprecated
1734
1735This warning has been changed to the fatal
1736L<Malformed UTF-8 string in "%s"
1737|perldiag/Malformed UTF-8 string in "%s">
1738
1739=item *
1740
1741L<Setting C<< $E<sol> >> to a reference to %s as a form of slurp is deprecated, treating as undef. This will be fatal in Perl 5.28
1742|perldiag/Setting C<< $E<sol> >> to a reference to %s as a form of slurp is deprecated, treating as undef. This will be fatal in Perl 5.28>
1743
1744This existing warning has had the I<this will be fatal> text added in
1745this release.
1746
1747=item *
1748
1749L<C<${^ENCODING}> is no longer supported. Its use will be fatal in Perl 5.28|perldiag/"${^ENCODING} is no longer supported. Its use will be fatal in Perl 5.28">
1750
1751This warning used to be: "Setting C<${^ENCODING}> is deprecated".
1752
1753The special action of the variable C<${^ENCODING}> was formerly used to
1754implement the C<encoding> pragma. As of Perl 5.26, rather than being
1755deprecated, assigning to this variable now has no effect except to issue
1756the warning.
1757
1758=item *
1759
1760L<Too few arguments for subroutine '%s'
1761|perldiag/Too few arguments for subroutine '%s'>
1762
1763This warning now includes the name of the offending subroutine.
1764
1765=item *
1766
1767L<Too many arguments for subroutine '%s'
1768|perldiag/Too many arguments for subroutine '%s'>
1769
1770This warning now includes the name of the offending subroutine.
1771
1772=item *
1773
1774L<Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through in regex; marked by S<< E<lt>-- HERE >> in mE<sol>%sE<sol>
1775|perldiag/Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through in regex; marked by S<< E<lt>-- HERE >> in mE<sol>%sE<sol>>
1776
1777This existing warning has had the I<here (and will be fatal...)> text
1778added in this release.
1779
1780=item *
1781
1782L<Unknown charname '' is deprecated. Its use will be fatal in Perl 5.28
1783|perldiag/Unknown charname '' is deprecated. Its use will be fatal in Perl 5.28>
1784
1785This existing warning has had the I<its use will be fatal> text added in
1786this release.
1787
1788=item *
1789
1790L<Use of bare E<lt>E<lt> to mean E<lt>E<lt>"" is deprecated. Its use will be fatal in Perl 5.28
1791|perldiag/Use of bare E<lt>E<lt> to mean E<lt>E<lt>"" is deprecated. Its use will be fatal in Perl 5.28>
1792
1793This existing warning has had the I<its use will be fatal> text added in
1794this release.
1795
1796=item *
1797
1798L<Use of code point 0x%s is deprecated; the permissible max is 0x%s. This will be fatal in Perl 5.28
1799|perldiag/Use of code point 0x%s is deprecated; the permissible max is 0x%s. This will be fatal in Perl 5.28>
1800
1801This existing warning has had the I<this will be fatal> text added in
1802this release.
1803
1804=item *
1805
1806L<Use of comma-less variable list is deprecated. Its use will be fatal in Perl 5.28
1807|perldiag/Use of comma-less variable list is deprecated. Its use will be fatal in Perl 5.28>
1808
1809This existing warning has had the I<its use will be fatal> text added in
1810this release.
1811
1812=item *
1813
1814L<Use of inherited C<AUTOLOAD> for non-method %s() is deprecated. This will be fatal in Perl 5.28
1815|perldiag/Use of inherited C<AUTOLOAD> for non-method %s() is deprecated. This will be fatal in Perl 5.28>
1816
1817This existing warning has had the I<this will be fatal> text added in
1818this release.
1819
1820=item *
1821
1822L<Use of strings with code points over 0xFF as arguments to %s operator is deprecated. This will be a fatal error in Perl 5.28
1823|perldiag/Use of strings with code points over 0xFF as arguments to %s operator is deprecated. This will be a fatal error in Perl 5.28>
1824
1825This existing warning has had the I<this will be a fatal error> text added in
1826this release.
1827
1828=back
1829
1830=head1 Utility Changes
1831
1832=head2 F<c2ph> and F<pstruct>
1833
1834=over 4
1835
1836=item *
1837
1838These old utilities have long since superceded by L<h2xs>, and are
1839now gone from the distribution.
1840
1841=back
1842
1843=head2 F<Porting/pod_lib.pl>
1844
1845=over 4
1846
1847=item *
1848
1849Removed spurious executable bit.
1850
1851=item *
1852
1853Account for the possibility of DOS file endings.
1854
1855=back
1856
1857=head2 F<Porting/sync-with-cpan>
1858
1859=over 4
1860
1861=item *
1862
1863Many improvements.
1864
1865=back
1866
1867=head2 F<perf/benchmarks>
1868
1869=over 4
1870
1871=item *
1872
1873Tidy file, rename some symbols.
1874
1875=back
1876
1877=head2 F<Porting/checkAUTHORS.pl>
1878
1879=over 4
1880
1881=item *
1882
1883Replace obscure character range with C<\w>.
1884
1885=back
1886
1887=head2 F<t/porting/regen.t>
1888
1889=over 4
1890
1891=item *
1892
1893Try to be more helpful when tests fail.
1894
1895=back
1896
1897=head2 F<utils/h2xs.PL>
1898
1899=over 4
1900
1901=item *
1902
1903Avoid infinite loop for enums.
1904
1905=back
1906
1907=head2 L<perlbug>
1908
1909=over 4
1910
1911=item *
1912
1913Long lines in the message body are now wrapped at 900 characters, to stay
1914well within the 1000-character limit imposed by SMTP mail transfer agents.
1915This is particularly likely to be important for the list of arguments to
1916F<Configure>, which can readily exceed the limit if, for example, it names
1917several non-default installation paths. This change also adds the first unit
1918tests for perlbug.
1919L<[perl #128020]|https://rt.perl.org/Public/Bug/Display.html?id=128020>
1920
1921=back
1922
1923=head1 Configuration and Compilation
1924
1925=over 4
1926
1927=item *
1928
1929C<-Ddefault_inc_excludes_dot> has added, and enabled by default.
1930
1931=item *
1932
1933The C<dtrace> build process has further changes
1934L<[perl #130108]|https://rt.perl.org/Public/Bug/Display.html?id=130108>:
1935
1936=over
1937
1938=item *
1939
1940If the C<-xnolibs> is available, use that so a F<dtrace> perl can be
1941built within a FreeBSD jail.
1942
1943=item *
1944
1945On systems that build a F<dtrace> object file (FreeBSD, Solaris, and
1946SystemTap's dtrace emulation), copy the input objects to a separate
1947directory and process them there, and use those objects in the link,
1948since C<dtrace -G> also modifies these objects.
1949
1950=item *
1951
1952Add F<libelf> to the build on FreeBSD 10.x, since F<dtrace> adds
1953references to F<libelf> symbols.
1954
1955=item *
1956
1957Generate a dummy F<dtrace_main.o> if C<dtrace -G> fails to build it. A
1958default build on Solaris generates probes from the unused inline
1959functions, while they don't on FreeBSD, which causes C<dtrace -G> to
1960fail.
1961
1962=back
1963
1964=item *
1965
1966You can now disable perl's use of the C<PERL_HASH_SEED> and
1967C<PERL_PERTURB_KEYS> environment variables by configuring perl with
1968C<-Accflags=NO_PERL_HASH_ENV>.
1969
1970=item *
1971
1972You can now disable perl's use of the C<PERL_HASH_SEED_DEBUG> environment
1973variable by configuring perl with
1974C<-Accflags=-DNO_PERL_HASH_SEED_DEBUG>.
1975
1976=item *
1977
1978F<Configure> now zeroes out the alignment bytes when calculating the bytes
1979for 80-bit C<NaN> and C<Inf> to make builds more reproducible.
1980L<[perl #130133]|https://rt.perl.org/Public/Bug/Display.html?id=130133>
1981
1982=item *
1983
1984Since v5.18, for testing purposes we have included support for
1985building perl with a variety of non-standard, and non-recommended
1986hash functions. Since we do not recommend the use of these functions,
1987we have removed them and their corresponding build options. Specifically
1988this includes the following build options:
1989
1990 PERL_HASH_FUNC_SDBM
1991 PERL_HASH_FUNC_DJB2
1992 PERL_HASH_FUNC_SUPERFAST
1993 PERL_HASH_FUNC_MURMUR3
1994 PERL_HASH_FUNC_ONE_AT_A_TIME
1995 PERL_HASH_FUNC_ONE_AT_A_TIME_OLD
1996 PERL_HASH_FUNC_MURMUR_HASH_64A
1997 PERL_HASH_FUNC_MURMUR_HASH_64B
1998
1999=item *
2000
2001Remove "Warning: perl appears in your path"
2002
2003This install warning is more or less obsolete, since most platforms already
2004B<will> have a F</usr/bin/perl> or similar provided by the OS.
2005
2006=item *
2007
2008Reduce verbosity of C<make install.man>
2009
2010Previously, two progress messages were emitted for each manpage: one by
2011installman itself, and one by the function in F<install_lib.pl> that it calls to
2012actually install the file. Disabling the second of those in each case saves
2013over 750 lines of unhelpful output.
2014
2015=item *
2016
2017Cleanup for C<clang -Weverything> support.
2018L<[perl #129961]|https://rt.perl.org/Public/Bug/Display.html?id=129961>
2019
2020=item *
2021
2022F<Configure>: signbit scan was assuming too much, stop assuming negative 0.
2023
2024=item *
2025
2026Various compiler warnings have been silenced.
2027
2028=item *
2029
2030Several smaller changes have been made to remove impediments to compiling
2031under C++11.
2032
2033=item *
2034
2035Builds using C<USE_PAD_RESET> now work again; this configuration had
2036bit-rotted.
2037
2038=item *
2039
2040A probe for C<gai_strerror> was added to F<Configure> that checks if
2041the C<gai_strerror()> routine is available and can be used to
2042translate error codes returned by C<getaddrinfo()> into human
2043readable strings.
2044
2045=item *
2046
2047F<Configure> now aborts if both C<-Duselongdouble> and C<-Dusequadmath> are
2048requested.
2049L<[perl #126203]|https://rt.perl.org/Public/Bug/Display.html?id=126203>
2050
2051=item *
2052
2053Fixed a bug in which F<Configure> could append C<-quadmath> to the
2054archname even if it was already present.
2055L<[perl #128538]|https://rt.perl.org/Public/Bug/Display.html?id=128538>
2056
2057=item *
2058
2059Clang builds with C<-DPERL_GLOBAL_STRUCT> or
2060C<-DPERL_GLOBAL_STRUCT_PRIVATE> have
2061been fixed (by disabling Thread Safety Analysis for these configurations).
2062
2063=item *
2064
2065F<make_ext.pl> no longer updates a module's F<pm_to_blib> file when no
2066files require updates. This could cause dependencies, F<perlmain.c>
2067in particular, to be rebuilt unnecessarily.
2068L<[perl #126710]|https://rt.perl.org/Public/Bug/Display.html?id=126710>
2069
2070=item *
2071
2072The output of C<perl -V> has been reformatted so that each configuration
2073and compile-time option is now listed one per line, to improve
2074readability.
2075
2076=item *
2077
2078F<Configure> now builds C<miniperl> and C<generate_uudmap> if you
2079invoke it with C<-Dusecrosscompiler> but not C<-Dtargethost=somehost>.
2080This means you can supply your target platform C<config.sh>, generate
2081the headers and proceed to build your cross-target perl.
2082L<[perl #127234]|https://rt.perl.org/Public/Bug/Display.html?id=127234>
2083
2084=item *
2085
2086Perl built with C<-Accflags=-DPERL_TRACE_OPS> now only dumps the operator
2087counts when the environment variable C<PERL_TRACE_OPS> is set to a
2088non-zero integer. This allows C<make test> to pass on such a build.
2089
2090=item *
2091
2092When building with GCC 6 and link-time optimization (the C<-flto> option to
2093C<gcc>), F<Configure> was treating all probed symbols as present on the
2094system, regardless of whether they actually exist. This has been fixed.
2095L<[perl #128131]|https://rt.perl.org/Public/Bug/Display.html?id=128131>
2096
2097=item *
2098
2099The F<t/test.pl> library is used for internal testing of Perl itself, and
2100also copied by several CPAN modules. Some of those modules must work on
2101older versions of Perl, so F<t/test.pl> must in turn avoid newer Perl
2102features. Compatibility with Perl 5.8 was inadvertently removed some time
2103ago; it has now been restored.
2104L<[perl #128052]|https://rt.perl.org/Public/Bug/Display.html?id=128052>
2105
2106=item *
2107
2108The build process no longer emits an extra blank line before building each
2109"simple" extension (those with only F<*.pm> and F<*.pod> files).
2110
2111=back
2112
2113=head1 Testing
2114
2115Tests were added and changed to reflect the other additions and changes
2116in this release. Furthermore, these substantive changes were made:
2117
2118=over 4
2119
2120=item *
2121
2122A new test script, F<comp/parser_run.t>, has been added that is like
2123F<comp/parser.t> but with F<test.pl> included so that C<runperl()> and the
2124like are available for use.
2125
2126=item *
2127
2128Tests for locales were erroneously using locales incompatible with Perl.
2129
2130=item *
2131
2132Some parts of the test suite that try to exhaustively test edge cases in the
2133regex implementation have been restricted to running for a maximum of five
2134minutes. On slow systems they could otherwise take several hours, without
2135significantly improving our understanding of the correctness of the code
2136under test.
2137
2138=item *
2139
2140A new internal facility allows analysing the time taken by the individual
2141tests in Perl's own test suite; see F<Porting/harness-timer-report.pl>.
2142
2143=item *
2144
2145F<t/re/regexp_nonull.t> has been added to test that the regular expression
2146engine can handle scalars that do not have a null byte just past the end of
2147the string.
2148
2149=item *
2150
2151A new test script, F<t/op/decl-refs.t>, has been added to test the new feature
2152L</Declaring a reference to a variable>.
2153
2154=item *
2155
2156A new test script, F<t/re/keep_tabs.t> has been added to contain tests
2157where C<\t> characters should not be expanded into spaces.
2158
2159=item *
2160
2161A new test script, F<t/re/anyof.t>, has been added to test that the ANYOF nodes
2162generated by bracketed character classes are as expected.
2163
2164=item *
2165
2166There is now more extensive testing of the Unicode-related API macros
2167and functions.
2168
2169=item *
2170
2171Several of the longer running API test files have been split into
2172multiple test files so that they can be run in parallel.
2173
2174=item *
2175
2176F<t/harness> now tries really hard not to run tests which are located
2177outside of the Perl source tree.
2178L<[perl #124050]|https://rt.perl.org/Public/Bug/Display.html?id=124050>
2179
2180=item *
2181
2182Prevent debugger tests (F<lib/perl5db.t>) from failing due to the contents
2183of C<$ENV{PERLDB_OPTS}>.
2184L<[perl #130445]|https://rt.perl.org/Public/Bug/Display.html?id=130445>
2185
2186=back
2187
2188=head1 Platform Support
2189
2190=head2 New Platforms
2191
2192=over 4
2193
2194=item NetBSD/VAX
2195
2196Perl now compiles under NetBSD on VAX machines. However, it's not
2197possible for that platform to implement floating-point infinities and
2198NaNs compatible with most modern systems, which implement the IEEE-754
2199floating point standard. The hexadecimal floating point (C<0x...p[+-]n>
2200literals, C<printf %a>) is not implemented, either.
2201The C<make test> passes 98% of tests.
2202
2203=over 4
2204
2205=item *
2206
2207Test fixes and minor updates.
2208
2209=item *
2210
2211Account for lack of C<inf>, C<nan>, and C<-0.0> support.
2212
2213=back
2214
2215=back
2216
2217=head2 Platform-Specific Notes
2218
2219=over 4
2220
2221=item Darwin
2222
2223=over 4
2224
2225=item *
2226
2227Don't treat C<-Dprefix=/usr> as special: instead require an extra option
2228C<-Ddarwin_distribution> to produce the same results.
2229
2230=item *
2231
2232OS X El Capitan doesn't implement the C<clock_gettime()> or
2233C<clock_getres()> APIs; emulate them as necessary.
2234
2235=item *
2236
2237Deprecated C<syscall(2)> on macOS 10.12.
2238
2239=back
2240
2241=item EBCDIC
2242
2243Several tests have been updated to work (or be skipped) on EBCDIC platforms.
2244
2245=item HP-UX
2246
2247The L<Net::Ping> UDP test is now skipped on HP-UX.
2248
2249=item Hurd
2250
2251The hints for Hurd have been improved, enabling malloc wrap and reporting the
2252GNU libc used (previously it was an empty string when reported).
2253
2254=item VAX
2255
2256VAX floating point formats are now supported on NetBSD.
2257
2258=item VMS
2259
2260=over 4
2261
2262=item *
2263
2264The path separator for the C<PERL5LIB> and C<PERLLIB> environment entries is
2265now a colon (C<":">) when running under a Unix shell. There is no change when
2266running under DCL (it's still C<"|">).
2267
2268=item *
2269
2270F<configure.com> now recognizes the VSI-branded C compiler and no longer
2271recognizes the "DEC"-branded C compiler (as there hasn't been such a thing for
227215 or more years).
2273
2274=back
2275
2276=item Windows
2277
2278=over 4
2279
2280=item *
2281
2282Support for compiling perl on Windows using Microsoft Visual Studio 2015
2283(containing Visual C++ 14.0) has been added.
2284
2285This version of VC++ includes a completely rewritten C run-time library, some
2286of the changes in which mean that work done to resolve a socket
2287C<close()> bug in
2288perl #120091 and perl #118059 is not workable in its current state with this
2289version of VC++. Therefore, we have effectively reverted that bug fix for
2290VS2015 onwards on the basis that being able to build with VS2015 onwards is
2291more important than keeping the bug fix. We may revisit this in the future to
2292attempt to fix the bug again in a way that is compatible with VS2015.
2293
2294These changes do not affect compilation with GCC or with Visual Studio versions
2295up to and including VS2013, I<i.e.>, the bug fix is retained (unchanged) for those
2296compilers.
2297
2298Note that you may experience compatibility problems if you mix a perl built
2299with GCC or VS E<lt>= VS2013 with XS modules built with VS2015, or if you mix a
2300perl built with VS2015 with XS modules built with GCC or VS E<lt>= VS2013.
2301Some incompatibility may arise because of the bug fix that has been reverted
2302for VS2015 builds of perl, but there may well be incompatibility anyway because
2303of the rewritten CRT in VS2015 (I<e.g.>, see discussion at
2304L<http://stackoverflow.com/questions/30412951>).
2305
2306=item *
2307
2308It now automatically detects GCC versus Visual C and sets the VC version
2309number on Win32.
2310
2311=back
2312
2313=item Linux
2314
2315Drop support for Linux F<a.out> executable format. Linux has used ELF for
2316over twenty years.
2317
2318=item OpenBSD 6
2319
2320OpenBSD 6 still does not support returning C<pid>, C<gid>, or C<uid> with
2321C<SA_SIGINFO>. Make sure to account for it.
2322
2323=item FreeBSD
2324
2325F<t/uni/overload.t>: Skip hanging test on FreeBSD.
2326
2327=item DragonFly BSD
2328
2329DragonFly BSD now has support for C<setproctitle()>.
2330L<[perl #130068]|https://rt.perl.org/Public/Bug/Display.html?id=130068>.
2331
2332=back
2333
2334=head1 Internal Changes
2335
2336=over 4
2337
2338=item *
2339
2340A new API function L<C<sv_setpv_bufsize()>|perlapi/sv_setpv_bufsize>
2341allows simultaneously setting the
2342length and the allocated size of the buffer in an C<SV>, growing the
2343buffer if necessary.
2344
2345=item *
2346
2347A new API macro L<C<SvPVCLEAR()>|perlapi/SvPVCLEAR> sets its C<SV>
2348argument to an empty string,
2349like Perl-space C<$x = ''>, but with several optimisations.
2350
2351=item *
2352
2353Several new macros and functions for dealing with Unicode and
2354UTF-8-encoded strings have been added to the API, as well as some
2355changes in the
2356functionality of existing functions (see L<perlapi/Unicode Support> for
2357more details):
2358
2359=over
2360
2361=item *
2362
2363New versions of the API macros like C<isALPHA_utf8> and C<toLOWER_utf8>
2364have been added, each with the suffix C<_safe>, like
2365L<C<isSPACE_utf8_safe>|perlapi/isSPACE>. These take an extra
2366parameter, giving an upper
2367limit of how far into the string it is safe to read. Using the old
2368versions could cause attempts to read beyond the end of the input buffer
2369if the UTF-8 is not well-formed, and their use now raises a deprecation
2370warning. Details are at L<perlapi/Character classification>.
2371
2372=item *
2373
2374Macros like L<C<isALPHA_utf8>|perlapi/isALPHA> and
2375L<C<toLOWER_utf8>|perlapi/toLOWER_utf8> now die if they detect
2376that their input UTF-8 is malformed. A deprecation warning had been
2377issued since Perl 5.18.
2378
2379=item *
2380
2381Several new macros for analysing the validity of utf8 sequences. These
2382are:
2383
2384L<C<UTF8_GOT_ABOVE_31_BIT>|perlapi/UTF8_GOT_ABOVE_31_BIT>
2385L<C<UTF8_GOT_CONTINUATION>|perlapi/UTF8_GOT_CONTINUATION>
2386L<C<UTF8_GOT_EMPTY>|perlapi/UTF8_GOT_EMPTY>
2387L<C<UTF8_GOT_LONG>|perlapi/UTF8_GOT_LONG>
2388L<C<UTF8_GOT_NONCHAR>|perlapi/UTF8_GOT_NONCHAR>
2389L<C<UTF8_GOT_NON_CONTINUATION>|perlapi/UTF8_GOT_NON_CONTINUATION>
2390L<C<UTF8_GOT_OVERFLOW>|perlapi/UTF8_GOT_OVERFLOW>
2391L<C<UTF8_GOT_SHORT>|perlapi/UTF8_GOT_SHORT>
2392L<C<UTF8_GOT_SUPER>|perlapi/UTF8_GOT_SUPER>
2393L<C<UTF8_GOT_SURROGATE>|perlapi/UTF8_GOT_SURROGATE>
2394L<C<UTF8_IS_INVARIANT>|perlapi/UTF8_IS_INVARIANT>
2395L<C<UTF8_IS_NONCHAR>|perlapi/UTF8_IS_NONCHAR>
2396L<C<UTF8_IS_SUPER>|perlapi/UTF8_IS_SUPER>
2397L<C<UTF8_IS_SURROGATE>|perlapi/UTF8_IS_SURROGATE>
2398L<C<UVCHR_IS_INVARIANT>|perlapi/UVCHR_IS_INVARIANT>
2399L<C<isUTF8_CHAR_flags>|perlapi/isUTF8_CHAR_flags>
2400L<C<isSTRICT_UTF8_CHAR>|perlapi/isSTRICT_UTF8_CHAR>
2401L<C<isC9_STRICT_UTF8_CHAR>|perlapi/isC9_STRICT_UTF8_CHAR>
2402
2403=item *
2404
2405Functions that are all extensions of the C<is_utf8_string_I<*>()> functions,
2406that apply various restrictions to the UTF-8 recognized as valid:
2407
2408L<C<is_strict_utf8_string>|perlapi/is_strict_utf8_string>,
2409L<C<is_strict_utf8_string_loc>|perlapi/is_strict_utf8_string_loc>,
2410L<C<is_strict_utf8_string_loclen>|perlapi/is_strict_utf8_string_loclen>,
2411
2412L<C<is_c9strict_utf8_string>|perlapi/is_c9strict_utf8_string>,
2413L<C<is_c9strict_utf8_string_loc>|perlapi/is_c9strict_utf8_string_loc>,
2414L<C<is_c9strict_utf8_string_loclen>|perlapi/is_c9strict_utf8_string_loclen>,
2415
2416L<C<is_utf8_string_flags>|perlapi/is_utf8_string_flags>,
2417L<C<is_utf8_string_loc_flags>|perlapi/is_utf8_string_loc_flags>,
2418L<C<is_utf8_string_loclen_flags>|perlapi/is_utf8_string_loclen_flags>,
2419
2420L<C<is_utf8_fixed_width_buf_flags>|perlapi/is_utf8_fixed_width_buf_flags>,
2421L<C<is_utf8_fixed_width_buf_loc_flags>|perlapi/is_utf8_fixed_width_buf_loc_flags>,
2422L<C<is_utf8_fixed_width_buf_loclen_flags>|perlapi/is_utf8_fixed_width_buf_loclen_flags>.
2423
2424L<C<is_utf8_invariant_string>|perlapi/is_utf8_invariant_string>.
2425L<C<is_utf8_valid_partial_char>|perlapi/is_utf8_valid_partial_char>.
2426L<C<is_utf8_valid_partial_char_flags>|perlapi/is_utf8_valid_partial_char_flags>.
2427
2428=item *
2429
2430The functions L<C<utf8n_to_uvchr>|perlapi/utf8n_to_uvchr> and its
2431derivatives have had several changes of behaviour.
2432
2433Calling them, while passing a string length of 0 is now asserted against
2434in DEBUGGING builds, and otherwise, returns the Unicode REPLACEMENT
2435CHARACTER. If you have nothing to decode, you shouldn't call the decode
2436function.
2437
2438They now return the Unicode REPLACEMENT CHARACTER if called with UTF-8
2439that has the overlong malformation and that malformation is allowed by
2440the input parameters. This malformation is where the UTF-8 looks valid
2441syntactically, but there is a shorter sequence that yields the same code
2442point. This has been forbidden since Unicode version 3.1.
2443
2444They now accept an input
2445flag to allow the overflow malformation. This malformation is when the
2446UTF-8 may be syntactically valid, but the code point it represents is
2447not capable of being represented in the word length on the platform.
2448What "allowed" means, in this case, is that the function doesn't return an
2449error, and it advances the parse pointer to beyond the UTF-8 in
2450question, but it returns the Unicode REPLACEMENT CHARACTER as the value
2451of the code point (since the real value is not representable).
2452
2453They no longer abandon searching for other malformations when the first
2454one is encountered. A call to one of these functions thus can generate
2455multiple diagnostics, instead of just one.
2456
2457=item *
2458
2459L<C<valid_utf8_to_uvchr()>|perlapi/valid_utf8_to_uvchr> has been added
2460to the API (although it was
2461present in core earlier). Like C<utf8_to_uvchr_buf()>, but assumes that
2462the next character is well-formed. Use with caution.
2463
2464=item *
2465
2466A new function, L<C<utf8n_to_uvchr_error>|perlapi/utf8n_to_uvchr_error>,
2467has been added for
2468use by modules that need to know the details of UTF-8 malformations
2469beyond pass/fail. Previously, the only ways to know why a sequence was
2470ill-formed was to capture and parse the generated diagnostics or to do
2471your own analysis.
2472
2473=item *
2474
2475There is now a safer version of utf8_hop(), called
2476L<C<utf8_hop_safe()>|perlapi/utf8_hop_safe>.
2477Unlike utf8_hop(), utf8_hop_safe() won't navigate before the beginning or
2478after the end of the supplied buffer.
2479
2480=item *
2481
2482Two new functions, L<C<utf8_hop_forward()>|perlapi/utf8_hop_forward> and
2483L<C<utf8_hop_back()>|perlapi/utf8_hop_back> are
2484similar to C<utf8_hop_safe()> but are for when you know which direction
2485you wish to travel.
2486
2487=item *
2488
2489Two new macros which return useful utf8 byte sequences:
2490
2491L<C<BOM_UTF8>|perlapi/BOM_UTF8>
2492
2493L<C<REPLACEMENT_CHARACTER_UTF8>|perlapi/REPLACEMENT_CHARACTER_UTF8>
2494
2495=back
2496
2497=item *
2498
2499Perl is now built with the C<PERL_OP_PARENT> compiler define enabled by
2500default. To disable it, use the C<PERL_NO_OP_PARENT> compiler define.
2501This flag alters how the C<op_sibling> field is used in C<OP> structures,
2502and has been available optionally since perl 5.22.
2503
2504See L<perl5220delta/"Internal Changes"> for more details of what this
2505build option does.
2506
2507=item *
2508
2509Three new ops, C<OP_ARGELEM>, C<OP_ARGDEFELEM>, and C<OP_ARGCHECK> have
2510been added. These are intended principally to implement the individual
2511elements of a subroutine signature, plus any overall checking required.
2512
2513=item *
2514
2515The C<OP_PUSHRE> op has been eliminated and the C<OP_SPLIT> op has been
2516changed from class C<LISTOP> to C<PMOP>.
2517
2518Formerly the first child of a split would be a C<pushre>, which would have the
2519C<split>'s regex attached to it. Now the regex is attached directly to the
2520C<split> op, and the C<pushre> has been eliminated.
2521
2522=item *
2523
2524The L<C<op_class()>|perlapi/op_class> API function has been added. This
2525is like the existing
2526C<OP_CLASS()> macro, but can more accurately determine what struct an op
2527has been allocated as. For example C<OP_CLASS()> might return
2528C<OA_BASEOP_OR_UNOP> indicating that ops of this type are usually
2529allocated as an C<OP> or C<UNOP>; while C<op_class()> will return
2530C<OPclass_BASEOP> or C<OPclass_UNOP> as appropriate.
2531
2532=item *
2533
2534All parts of the internals now agree that the C<sassign> op is a C<BINOP>;
2535previously it was listed as a C<BASEOP> in F<regen/opcodes>, which meant
2536that several parts of the internals had to be special-cased to accommodate
2537it. This oddity's original motivation was to handle code like C<$x ||= 1>;
2538that is now handled in a simpler way.
2539
2540=item *
2541
2542The output format of the L<C<op_dump()>|perlapi/op_dump> function (as
2543used by C<perl -Dx>)
2544has changed: it now displays an "ASCII-art" tree structure, and shows more
2545low-level details about each op, such as its address and class.
2546
2547=item *
2548
2549The C<PADOFFSET> type has changed from being unsigned to signed, and
2550several pad-related variables such as C<PL_padix> have changed from being
2551of type C<I32> to type C<PADOFFSET>.
2552
2553=item *
2554
2555The C<DEBUGGING>-mode output for regex compilation and execution has been
2556enhanced.
2557
2558=item *
2559
2560Several obscure SV flags have been eliminated, sometimes along with the
2561macros which manipulate them: C<SVpbm_VALID>, C<SVpbm_TAIL>, C<SvTAIL_on>,
2562C<SvTAIL_off>, C<SVrepl_EVAL>, C<SvEVALED>.
2563
2564=item *
2565
2566An OP C<op_private> flag has been eliminated: C<OPpRUNTIME>. This used to
2567often get set on C<PMOP> ops, but had become meaningless over time.
2568
2569=back
2570
2571=head1 Selected Bug Fixes
2572
2573=over 4
2574
2575=item *
2576
2577Perl no longer panics when switching into some locales on machines with
2578buggy C<strxfrm()> implementations in their F<libc>.
2579L<[perl #121734]|https://rt.perl.org/Public/Bug/Display.html?id=121734>
2580
2581=item *
2582
2583C< $-{$name} > would leak an C<AV> on each access if the regular
2584expression had no named captures. The same applies to access to any
2585hash tied with L<Tie::Hash::NamedCapture> and C<< all =E<gt> 1 >>.
2586L<[perl #130822]|https://rt.perl.org/Public/Bug/Display.html?id=130822>
2587
2588=item *
2589
2590Attempting to use the deprecated variable C<$#> as the object in an
2591indirect object method call could cause a heap use after free or
2592buffer overflow.
2593L<[perl #129274]|https://rt.perl.org/Public/Bug/Display.html?id=129274>
2594
2595=item *
2596
2597When checking for an indirect object method call, in some rare cases
2598the parser could reallocate the line buffer but then continue to use
2599pointers to the old buffer.
2600L<[perl #129190]|https://rt.perl.org/Public/Bug/Display.html?id=129190>
2601
2602=item *
2603
2604Supplying a glob as the format argument to
2605L<C<formline>|perlfunc/formline> would
2606cause an assertion failure.
2607L<[perl #130722]|https://rt.perl.org/Public/Bug/Display.html?id=130722>
2608
2609=item *
2610
2611Code like C< $value1 =~ qr/.../ ~~ $value2 > would have the match
2612converted into a C<qr//> operator, leaving extra elements on the stack to
2613confuse any surrounding expression.
2614L<[perl #130705]|https://rt.perl.org/Public/Bug/Display.html?id=130705>
2615
2616=item *
2617
2618Since v5.24 in some obscure cases, a regex which included code blocks
2619from multiple sources (I<e.g.>, via embedded via C<qr//> objects) could end up
2620with the wrong current pad and crash or give weird results.
2621L<[perl #129881]|https://rt.perl.org/Public/Bug/Display.html?id=129881>
2622
2623=item *
2624
2625Occasionally C<local()>s in a code block within a patterns weren't being
2626undone when the pattern matching backtracked over the code block.
2627L<[perl #126697]|https://rt.perl.org/Public/Bug/Display.html?id=126697>
2628
2629=item *
2630
2631Using C<substr()> to modify a magic variable could access freed memory
2632in some cases.
61e2e049 2633L<[perl #130766]|https://rt.perl.org/Public/Bug/Display.html?id=130766>
b5cbfe35
S
2634
2635=item *
2636
2637Under C<use utf8>, the entire source code is now checked for being UTF-8
2638well formed, not just quoted strings as before.
2639L<[perl #126310]|https://rt.perl.org/Public/Bug/Display.html?id=126310>.
2640
2641=item *
2642
2643The range operator C<".."> on strings now handles its arguments correctly when in
2644the scope of the L<< C<unicode_strings>|feature/"The 'unicode_strings' feature" >>
2645feature. The previous behaviour was sufficiently unexpected that we believe no
2646correct program could have made use of it.
2647
2648=item *
2649
2650The C<split> operator did not ensure enough space was allocated for
2651its return value in scalar context. It could then write a single
2652pointer immediately beyond the end of the memory block allocated for
2653the stack.
2654L<[perl #130262]|https://rt.perl.org/Public/Bug/Display.html?id=130262>
2655
2656=item *
2657
2658Using a large code point with the C<"W"> pack template character with
2659the current output position aligned at just the right point could
2660cause a write of a single zero byte immediately beyond the end of an
2661allocated buffer.
2662L<[perl #129149]|https://rt.perl.org/Public/Bug/Display.html?id=129149>
2663
2664=item *
2665
2666Supplying a format's picture argument as part of the format argument list
2667where the picture specifies modifying the argument could cause an
95918867 2668access to the new freed compiled format.
b5cbfe35
S
2669L<[perl #129125]|https://rt.perl.org/Public/Bug/Display.html?id=129125>
2670
2671=item *
2672
2673The L<sort()|perlfunc/sort> operator's built-in numeric comparison
2674function didn't handle large integers that weren't exactly
2675representable by a double. This now uses the same code used to
2676implement the C<< E<lt>=E<gt> >> operator.
2677L<[perl #130335]|https://rt.perl.org/Public/Bug/Display.html?id=130335>
2678
2679=item *
2680
2681Fix issues with C</(?{ ... E<lt>E<lt>EOF })/> that broke
2682L<Method::Signatures>.
2683L<[perl #130398]|https://rt.perl.org/Public/Bug/Display.html?id=130398>
2684
2685=item *
2686
2687Fixed an assertion failure with C<chop> and C<chomp>, which
2688could be triggered by C<chop(@x =~ tr/1/1/)>.
2689L<[perl #130198]|https://rt.perl.org/Public/Bug/Display.html?id=130198>.
2690
2691=item *
2692
2693Fixed a comment skipping error in patterns under C</x>; it could stop
2694skipping a byte early, which could be in the middle of a UTF-8
2695character.
2696L<[perl #130495]|https://rt.perl.org/Public/Bug/Display.html?id=130495>.
2697
2698=item *
2699
2700F<perldb> now ignores F</dev/tty> on non-Unix systems.
2701L<[perl #113960]|https://rt.perl.org/Public/Bug/Display.html?id=113960>;
2702
2703=item *
2704
2705Fix assertion failure for C<{}-E<gt>$x> when C<$x> isn't defined.
2706L<[perl #130496]|https://rt.perl.org/Public/Bug/Display.html?id=130496>.
2707
2708=item *
2709
2710Fix an assertion error which could be triggered when a lookahead string
2711in patterns exceeded a minimum length.
2712L<[perl #130522]|https://rt.perl.org/Public/Bug/Display.html?id=130522>.
2713
2714=item *
2715
2716Only warn once per literal number about a misplaced C<"_">.
2717L<[perl #70878]|https://rt.perl.org/Public/Bug/Display.html?id=70878>.
2718
2719=item *
2720
2721The C<tr///> parse code could be looking at uninitialized data after a
2722perse error.
2723L<[perl #129342]|https://rt.perl.org/Public/Bug/Display.html?id=129342>.
2724
2725=item *
2726
2727In a pattern match, a back-reference (C<\1>) to an unmatched capture could
2728read back beyond the start of the string being matched.
2729L<[perl #129377]|https://rt.perl.org/Public/Bug/Display.html?id=129377>.
2730
2731=item *
2732
2733C<use re 'strict'> is supposed to warn if you use a range (such as
2734C</(?[ [ X-Y ] ])/>) whose start and end digit aren't from the same group
2735of 10. It didn't do that for five groups of mathematical digits starting
2736at C<U+1D7E>.
2737
2738=item *
2739
2740A sub containing a "forward" declaration with the same name (I<e.g.>,
2741C<sub c { sub c; }>) could sometimes crash or loop infinitely.
2742L<[perl #129090]|https://rt.perl.org/Public/Bug/Display.html?id=129090>
2743
2744=item *
2745
2746A crash in executing a regex with a non-anchored UTF-8 substring against a
2747target string that also used UTF-8 has been fixed.
2748L<[perl #129350]|https://rt.perl.org/Public/Bug/Display.html?id=129350>
2749
2750=item *
2751
2752Previously, a shebang line like C<#!perl -i u> could be erroneously
2753interpreted as requesting the C<-u> option. This has been fixed.
2754L<[perl #129336]|https://rt.perl.org/Public/Bug/Display.html?id=129336>
2755
2756=item *
2757
2758The regex engine was previously producing incorrect results in some rare
2759situations when backtracking past an alternation that matches only one
2760thing; this
2761showed up as capture buffers (C<$1>, C<$2>, I<etc.>) erroneously containing data
2762from regex execution paths that weren't actually executed for the final
2763match.
2764L<[perl #129897]|https://rt.perl.org/Public/Bug/Display.html?id=129897>
2765
2766=item *
2767
2768Certain regexes making use of the experimental C<regex_sets> feature could
2769trigger an assertion failure. This has been fixed.
2770L<[perl #129322]|https://rt.perl.org/Public/Bug/Display.html?id=129322>
2771
2772=item *
2773
2774Invalid assignments to a reference constructor (I<e.g.>, C<\eval=time>) could
2775sometimes crash in addition to giving a syntax error.
2776L<[perl #125679]|https://rt.perl.org/Public/Bug/Display.html?id=125679>
2777
2778=item *
2779
2780The parser could sometimes crash if a bareword came after C<evalbytes>.
2781L<[perl #129196]|https://rt.perl.org/Public/Bug/Display.html?id=129196>
2782
2783=item *
2784
2785Autoloading via a method call would warn erroneously ("Use of inherited
2786AUTOLOAD for non-method") if there was a stub present in the package into
2787which the invocant had been blessed. The warning is no longer emitted in
2788such circumstances.
2789L<[perl #47047]|https://rt.perl.org/Public/Bug/Display.html?id=47047>
2790
2791=item *
2792
2793The use of C<splice> on arrays with non-existent elements could cause other
2794operators to crash.
2795L<[perl #129164]|https://rt.perl.org/Public/Bug/Display.html?id=129164>
2796
2797=item *
2798
2799A possible buffer overrun when a pattern contains a fixed utf8 substring.
2800L<[perl #129012]|https://rt.perl.org/Public/Bug/Display.html?id=129012>
2801
2802=item *
2803
2804Fixed two possible use-after-free bugs in perl's lexer.
2805L<[perl #129069]|https://rt.perl.org/Public/Bug/Display.html?id=129069>
2806
2807=item *
2808
2809Fixed a crash with C<s///l> where it thought it was dealing with UTF-8
2810when it wasn't.
2811L<[perl #129038]|https://rt.perl.org/Public/Bug/Display.html?id=129038>
2812
2813=item *
2814
2815Fixed a place where the regex parser was not setting the syntax error
2816correctly on a syntactically incorrect pattern.
2817L<[perl #129122]|https://rt.perl.org/Public/Bug/Display.html?id=129122>
2818
2819=item *
2820
2821The C<&.> operator (and the C<"&"> operator, when it treats its arguments as
2822strings) were failing to append a trailing null byte if at least one string
2823was marked as utf8 internally. Many code paths (system calls, regexp
2824compilation) still expect there to be a null byte in the string buffer
2825just past the end of the logical string. An assertion failure was the
2826result.
2827L<[perl #129287]|https://rt.perl.org/Public/Bug/Display.html?id=129287>
2828
2829=item *
2830
2831Avoid a heap-after-use error in the parser when creating an error messge
2832for a syntactically invalid heredoc.
2833L<[perl #128988]|https://rt.perl.org/Public/Bug/Display.html?id=128988>
2834
2835=item *
2836
2837Fix a segfault when run with C<-DC> options on DEBUGGING builds.
2838L<[perl #129106]|https://rt.perl.org/Public/Bug/Display.html?id=129106>
2839
2840=item *
2841
2842Fixed the parser error handling in subroutine attributes for an
2843'C<:attr(foo>' that does not have an ending 'C<")">'.
2844
2845=item *
2846
2847Fix the perl lexer to correctly handle a backslash as the last char in
2848quoted-string context. This actually fixed two bugs,
2849L<[perl #129064]|https://rt.perl.org/Public/Bug/Display.html?id=129064> and
2850L<[perl #129176]|https://rt.perl.org/Public/Bug/Display.html?id=129176>.
2851
2852=item *
2853
2854In the API function C<gv_fetchmethod_pvn_flags>, rework separator parsing
2855to prevent possible string overrun with an invalid C<len> argument.
2856L<[perl #129267]|https://rt.perl.org/Public/Bug/Display.html?id=129267>
2857
2858=item *
2859
2860Problems with in-place array sorts: code like C<@a = sort { ... } @a>,
2861where the source and destination of the sort are the same plain array, are
2862optimised to do less copying around. Two side-effects of this optimisation
2863were that the contents of C<@a> as seen by sort routines were
2864partially sorted; and under some circumstances accessing C<@a> during the
2865sort could crash the interpreter. Both these issues have been fixed, and
2866Sort functions see the original value of C<@a>.
2867L<[perl #128340]|https://rt.perl.org/Public/Bug/Display.html?id=128340>
2868
2869=item *
2870
2871Non-ASCII string delimiters are now reported correctly in error messages
2872for unterminated strings.
2873L<[perl #128701]|https://rt.perl.org/Public/Bug/Display.html?id=128701>
2874
2875=item *
2876
2877C<pack("p", ...)> used to emit its warning ("Attempt to pack pointer to
2878temporary value") erroneously in some cases, but has been fixed.
2879
2880=item *
2881
2882C<@DB::args> is now exempt from "used once" warnings. The warnings only
2883occurred under B<-w>, because F<warnings.pm> itself uses C<@DB::args>
2884multiple times.
2885
2886=item *
2887
2888The use of built-in arrays or hash slices in a double-quoted string no
2889longer issues a warning ("Possible unintended interpolation...") if the
2890variable has not been mentioned before. This affected code like
2891C<qq|@DB::args|> and C<qq|@SIG{'CHLD', 'HUP'}|>. (The special variables
2892C<@-> and C<@+> were already exempt from the warning.)
2893
2894=item *
2895
2896C<gethostent> and similar functions now perform a null check internally, to
2897avoid crashing with the torsocks library. This was a regression from v5.22.
2898L<[perl #128740]|https://rt.perl.org/Public/Bug/Display.html?id=128740>
2899
2900=item *
2901
2902C<defined *{'!'}>, C<defined *{'['}>, and C<defined *{'-'}> no longer leak
2903memory if the typeglob in question has never been accessed before.
2904
2905=item *
2906
2907Mentioning the same constant twice in a row (which is a syntax error) no
2908longer fails an assertion under debugging builds. This was a regression
2909from v5.20.
2910L<[perl #126482]|https://rt.perl.org/Public/Bug/Display.html?id=126482>
2911
2912=item *
2913
2914Many issues relating to C<printf "%a"> of hexadecimal floating point
2915were fixed. In addition, the "subnormals" (formerly known as "denormals")
2916floating point numbers are now supported both with the plain IEEE 754
2917floating point numbers (64-bit or 128-bit) and the x86 80-bit
2918"extended precision". Note that subnormal hexadecimal floating
2919point literals will give a warning about "exponent underflow".
2920L<[perl #128843]|https://rt.perl.org/Public/Bug/Display.html?id=128843>
2921L<[perl #128889]|https://rt.perl.org/Public/Bug/Display.html?id=128889>
2922L<[perl #128890]|https://rt.perl.org/Public/Bug/Display.html?id=128890>
2923L<[perl #128893]|https://rt.perl.org/Public/Bug/Display.html?id=128893>
2924L<[perl #128909]|https://rt.perl.org/Public/Bug/Display.html?id=128909>
2925L<[perl #128919]|https://rt.perl.org/Public/Bug/Display.html?id=128919>
2926
2927=item *
2928
2929A regression in v5.24 with C<tr/\N{U+...}/foo/> when the code point was between
2930128 and 255 has been fixed.
2931L<[perl #128734]|https://rt.perl.org/Public/Bug/Display.html?id=128734>.
2932
2933=item *
2934
2935Use of a string delimiter whose code point is above 2**31 now works
2936correctly on platforms that allow this. Previously, certain characters,
2937due to truncation, would be confused with other delimiter characters
2938with special meaning (such as C<"?"> in C<m?...?>), resulting
2939in inconsistent behaviour. Note that this is non-portable,
2940and is based on Perl's extension to UTF-8, and is probably not
2941displayable nor enterable by any editor.
2942L<[perl #128738]|https://rt.perl.org/Public/Bug/Display.html?id=128738>
2943
2944=item *
2945
2946C<@{x> followed by a newline where C<"x"> represents a control or non-ASCII
2947character no longer produces a garbled syntax error message or a crash.
2948L<[perl #128951]|https://rt.perl.org/Public/Bug/Display.html?id=128951>
2949
2950=item *
2951
2952An assertion failure with C<%: = 0> has been fixed.
2953L<[perl #128238]|https://rt.perl.org/Public/Bug/Display.html?id=128238>
2954
2955=item *
2956
2957In Perl 5.18, the parsing of C<"$foo::$bar"> was accidentally changed, such
2958that it would be treated as C<$foo."::".$bar>. The previous behavior, which
2959was to parse it as C<$foo:: . $bar>, has been restored.
2960L<[perl #128478]|https://rt.perl.org/Public/Bug/Display.html?id=128478>
2961
2962=item *
2963
2964Since Perl 5.20, line numbers have been off by one when perl is invoked with
2965the B<-x> switch. This has been fixed.
2966L<[perl #128508]|https://rt.perl.org/Public/Bug/Display.html?id=128508>
2967
2968=item *
2969
2970Vivifying a subroutine stub in a deleted stash (I<e.g.>,
2971C<delete $My::{"Foo::"}; \&My::Foo::foo>) no longer crashes. It had begun
2972crashing in Perl 5.18.
2973L<[perl #128532]|https://rt.perl.org/Public/Bug/Display.html?id=128532>
2974
2975=item *
2976
2977Some obscure cases of subroutines and file handles being freed at the same time
2978could result in crashes, but have been fixed. The crash was introduced in Perl
29795.22.
2980L<[perl #128597]|https://rt.perl.org/Public/Bug/Display.html?id=128597>
2981
2982=item *
2983
2984Code that looks for a variable name associated with an uninitialized value
2985could cause an assertion failure in cases where magic is involved, such as
2986C<$ISA[0][0]>. This has now been fixed.
2987L<[perl #128253]|https://rt.perl.org/Public/Bug/Display.html?id=128253>
2988
2989=item *
2990
2991A crash caused by code generating the warning "Subroutine STASH::NAME
2992redefined" in cases such as C<sub P::f{} undef *P::; *P::f =sub{};> has been
2993fixed. In these cases, where the STASH is missing, the warning will now appear
2994as "Subroutine NAME redefined".
2995L<[perl #128257]|https://rt.perl.org/Public/Bug/Display.html?id=128257>
2996
2997=item *
2998
2999Fixed an assertion triggered by some code that handles deprecated behavior in
3000formats, I<e.g.>, in cases like this:
3001
3002 format STDOUT =
3003 @
3004 0"$x"
3005
3006L<[perl #128255]|https://rt.perl.org/Public/Bug/Display.html?id=128255>
3007
3008=item *
3009
3010A possible divide by zero in string transformation code on Windows has been
3011avoided, fixing a crash when collating an empty string.
3012L<[perl #128618]|https://rt.perl.org/Public/Bug/Display.html?id=128618>
3013
3014=item *
3015
3016Some regular expression parsing glitches could lead to assertion failures with
3017regular expressions such as C</(?E<lt>=/> and C</(?E<lt>!/>. This has now been fixed.
3018L<[perl #128170]|https://rt.perl.org/Public/Bug/Display.html?id=128170>
3019
3020=item *
3021
3022C< until ($x = 1) { ... } > and C< ... until $x = 1 > now properly
3023warn when syntax warnings are enabled.
3024L<[perl #127333]|https://rt.perl.org/Public/Bug/Display.html?id=127333>
3025
3026=item *
3027
3028socket() now leaves the error code returned by the system in C<$!> on
3029failure.
3030L<[perl #128316]|https://rt.perl.org/Public/Bug/Display.html?id=128316>
3031
3032=item *
3033
3034Assignment variants of any bitwise ops under the C<bitwise> feature would
3035crash if the left-hand side was an array or hash.
3036L<[perl #128204]|https://rt.perl.org/Public/Bug/Display.html?id=128204>
3037
3038=item *
3039
3040C<require> followed by a single colon (as in C<foo() ? require : ...> is
3041now parsed correctly as C<require> with implicit C<$_>, rather than
3042C<require "">.
3043L<[perl #128307]|https://rt.perl.org/Public/Bug/Display.html?id=128307>
3044
3045=item *
3046
3047Scalar C<keys %hash> can now be assigned to consistently in all scalar
3048lvalue contexts. Previously it worked for some contexts but not others.
3049
3050=item *
3051
3052List assignment to C<vec> or C<substr> with an array or hash for its first
3053argument used to result in crashes or "Can't coerce" error messages at run
3054time, unlike scalar assignment, which would give an error at compile time.
3055List assignment now gives a compile-time error, too.
3056L<[perl #128260]|https://rt.perl.org/Public/Bug/Display.html?id=128260>
3057
3058=item *
3059
3060Expressions containing an C<&&> or C<||> operator (or their synonyms C<and>
3061and C<or>) were being compiled incorrectly in some cases. If the left-hand
3062side consisted of either a negated bareword constant or a negated C<do {}>
3063block containing a constant expression, and the right-hand side consisted of
3064a negated non-foldable expression, one of the negations was effectively
3065ignored. The same was true of C<if> and C<unless> statement modifiers,
3066though with the left-hand and right-hand sides swapped. This long-standing
3067bug has now been fixed.
3068L<[perl #127952]|https://rt.perl.org/Public/Bug/Display.html?id=127952>
3069
3070=item *
3071
3072C<reset> with an argument no longer crashes when encountering stash entries
3073other than globs.
3074L<[perl #128106]|https://rt.perl.org/Public/Bug/Display.html?id=128106>
3075
3076=item *
3077
3078Assignment of hashes to, and deletion of, typeglobs named C<*::::::> no
3079longer causes crashes.
3080L<[perl #128086]|https://rt.perl.org/Public/Bug/Display.html?id=128086>
3081
3082=item *
3083
3084Perl wasn't correctly handling true/false values in the LHS of a list
3085assign; specifically the truth values returned by boolean operators.
3086This could trigger an assertion failure in something like the following:
3087
3088 for ($x > $y) {
3089 ($_, ...) = (...); # here $_ is aliased to a truth value
3090 }
3091
3092This was a regression from v5.24.
3093L<[perl #129991]|https://rt.perl.org/Public/Bug/Display.html?id=129991>
3094
3095=item *
3096
3097Assertion failure with user-defined Unicode-like properties.
3098L<[perl #130010]|https://rt.perl.org/Public/Bug/Display.html?id=130010>
3099
3100=item *
3101
3102Fix error message for unclosed C<\N{> in a regex. An unclosed C<\N{>
3103could give the wrong error message:
3104C<"\N{NAME} must be resolved by the lexer">.
3105
3106=item *
3107
3108List assignment in list context where the LHS contained aggregates and
3109where there were not enough RHS elements, used to skip scalar lvalues.
3110Previously, C<(($a,$b,@c,$d) = (1))> in list context returned C<($a)>; now
3111it returns C<($a,$b,$d)>. C<(($a,$b,$c) = (1))> is unchanged: it still
3112returns C<($a,$b,$c)>. This can be seen in the following:
3113
3114 sub inc { $_++ for @_ }
3115 inc(($a,$b,@c,$d) = (10))
3116
3117Formerly, the values of C<($a,$b,$d)> would be left as C<(11,undef,undef)>;
3118now they are C<(11,1,1)>.
3119
3120=item *
3121
3122Code like this: C</(?{ s!!! })/> could trigger infinite recursion on the C
3123stack (not the normal perl stack) when the last successful pattern in
3124scope is itself. We avoid the segfault by simply forbidding the use of
3125the empty pattern when it would resolve to the currently executing
3126pattern.
3127L<[perl #129903]|https://rt.perl.org/Public/Bug/Display.html?id=129903>
3128
3129=item *
3130
3131Avoid reading beyond the end of the line buffer in perl's lexer when
3132there's a short UTF-8 character at the end.
3133L<[perl #128997]|https://rt.perl.org/Public/Bug/Display.html?id=128997>
3134
3135=item *
3136
3137Alternations in regular expressions were sometimes failing to match
3138a utf8 string against a utf8 alternate.
3139L<[perl #129950]|https://rt.perl.org/Public/Bug/Display.html?id=129950>
3140
3141=item *
3142
3143Make C<do "a\0b"> fail silently (and return C<undef> and set C<$!>)
3144instead of throwing an error.
3145L<[perl #129928]|https://rt.perl.org/Public/Bug/Display.html?id=129928>
3146
3147=item *
3148
3149C<chdir> with no argument didn't ensure that there was stack space
3150available for returning its result.
3151L<[perl #129130]|https://rt.perl.org/Public/Bug/Display.html?id=129130>
3152
3153=item *
3154
3155All error messages related to C<do> now refer to C<do>; some formerly
3156claimed to be from C<require> instead.
3157
3158=item *
3159
3160Executing C<undef $x> where C<$x> is tied or magical no longer incorrectly
3161blames the variable for an uninitialized-value warning encountered by the
3162tied/magical code.
3163
3164=item *
3165
3166Code like C<$x = $x . "a"> was incorrectly failing to yield a
3167L<use of uninitialized value|perldiag/"Use of uninitialized value%s">
3168warning when C<$x> was a lexical variable with an undefined value. That has
3169now been fixed.
3170L<[perl #127877]|https://rt.perl.org/Public/Bug/Display.html?id=127877>
3171
3172=item *
3173
3174C<undef *_; shift> or C<undef *_; pop> inside a subroutine, with no
3175argument to C<shift> or C<pop>, began crashing in Perl 5.14, but has now
3176been fixed.
3177
3178=item *
3179
3180C<< "string$scalar-E<gt>$*" >> now correctly prefers concatenation
3181overloading to string overloading if C<< $scalar-E<gt>$* >> returns an
3182overloaded object, bringing it into consistency with C<$$scalar>.
3183
3184=item *
3185
3186C<< /@0{0*-E<gt>@*/*0 >> and similar contortions used to crash, but no longer
3187do, but merely produce a syntax error.
3188L<[perl #128171]|https://rt.perl.org/Public/Bug/Display.html?id=128171>
3189
3190=item *
3191
3192C<do> or C<require> with an argument which is a reference or typeglob
3193which, when stringified,
3194contains a null character, started crashing in Perl 5.20, but has now been
3195fixed.
3196L<[perl #128182]|https://rt.perl.org/Public/Bug/Display.html?id=128182>
3197
3198=item *
3199
3200Improve the error message for a missing C<tie()> package/method. This
3201brings the error messages in line with the ones used for normal method
3202calls.
3203
3204=item *
3205
3206Parsing bad POSIX charclasses no longer leaks memory.
3207L<[perl #128313]|https://rt.perl.org/Public/Bug/Display.html?id=128313>
3208
3209=back
3210
3211=head1 Known Problems
3212
3213=over 4
3214
3215=item *
3216
3217G++ 6 handles subnormal (denormal) floating point values differently
3218than gcc 6 or g++ 5 resulting in "flush-to-zero". The end result is
3219that if you specify very small values using the hexadecimal floating
3220point format, like C<0x1.fffffffffffffp-1022>, they become zeros.
3221L<[perl #131388]|https://rt.perl.org/Ticket/Display.html?id=131388>
3222
3223=back
3224
3225=head1 Errata From Previous Releases
3226
3227=over 4
3228
3229=item *
3230
3231Fixed issues with recursive regexes. The behavior was fixed in Perl 5.24.
3232L<[perl #126182]|https://rt.perl.org/Public/Bug/Display.html?id=126182>
3233
3234=back
3235
3236=head1 Obituary
3237
3238Jon Portnoy (AVENJ), a prolific Perl author and admired Gentoo community
3239member, has passed away on August 10, 2016. He will be remembered and
3240missed by all those who he came in contact with, and enriched with his
3241intellect, wit, and spirit.
3242
3243It is with great sadness that we also note Kip Hampton's passing. Probably
3244best known as the author of the Perl & XML column on XML.com, he was a
3245core contributor to AxKit, an XML server platform that became an Apache
3246Foundation project. He was a frequent speaker in the early days at
3247OSCON, and most recently at YAPC::NA in Madison. He was frequently on
3248irc.perl.org as ubu, generally in the #axkit-dahut community, the
3249group responsible for YAPC::NA Asheville in 2011.
3250
3251Kip and his constant contributions to the community will be greatly
3252missed.
3253
3254=head1 Acknowledgements
3255
3256Perl 5.26.0 represents approximately 13 months of development since Perl 5.24.0
3257and contains approximately 360,000 lines of changes across 2,600 files from 86
3258authors.
3259
3260Excluding auto-generated files, documentation and release tools, there were
3261approximately 230,000 lines of changes to 1,800 .pm, .t, .c and .h files.
3262
3263Perl continues to flourish into its third decade thanks to a vibrant community
3264of users and developers. The following people are known to have contributed the
3265improvements that became Perl 5.26.0:
3266
3267Aaron Crane, Abigail, Ævar Arnfjörð Bjarmason, Alex Vandiver, Andreas
3268König, Andreas Voegele, Andrew Fresh, Andy Lester, Aristotle Pagaltzis, Chad
3269Granum, Chase Whitener, Chris 'BinGOs' Williams, Chris Lamb, Christian Hansen,
3270Christian Millour, Colin Newell, Craig A. Berry, Dagfinn Ilmari Mannsåker, Dan
3271Collins, Daniel Dragan, Dave Cross, Dave Rolsky, David Golden, David H.
3272Gutteridge, David Mitchell, Dominic Hargreaves, Doug Bell, E. Choroba, Ed Avis,
3273Father Chrysostomos, François Perrad, Hauke D, H.Merijn Brand, Hugo van der
3274Sanden, Ivan Pozdeev, James E Keenan, James Raspass, Jarkko Hietaniemi, Jerry
3275D. Hedden, Jim Cromie, J. Nick Koston, John Lightsey, Karen Etheridge, Karl
3276Williamson, Leon Timmermans, Lukas Mai, Matthew Horsfall, Maxwell Carey, Misty
3277De Meo, Neil Bowers, Nicholas Clark, Nicolas R., Niko Tyni, Pali, Paul
3278Marquess, Peter Avalos, Petr Písař, Pino Toscano, Rafael Garcia-Suarez, Reini
3279Urban, Renee Baecker, Ricardo Signes, Richard Levitte, Rick Delaney, Salvador
3280Fandiño, Samuel Thibault, Sawyer X, Sébastien Aperghis-Tramoni, Sergey
3281Aleynikov, Shlomi Fish, Smylers, Stefan Seifert, Steffen Müller, Stevan
3282Little, Steve Hay, Steven Humphrey, Sullivan Beck, Theo Buehler, Thomas Sibley,
3283Todd Rinaldo, Tomasz Konojacki, Tony Cook, Unicode Consortium, Yaroslav Kuzmin,
3284Yves Orton, Zefram.
3285
3286The list above is almost certainly incomplete as it is automatically generated
3287from version control history. In particular, it does not include the names of
3288the (very much appreciated) contributors who reported issues to the Perl bug
3289tracker.
3290
3291Many of the changes included in this version originated in the CPAN modules
3292included in Perl's core. We're grateful to the entire CPAN community for
3293helping Perl to flourish.
3294
3295For a more complete list of all of Perl's historical contributors, please see
3296the F<AUTHORS> file in the Perl source distribution.
3297
3298=head1 Reporting Bugs
3299
3300If you find what you think is a bug, you might check the perl bug database at
3301L<https://rt.perl.org/>. There may also be information at
3302L<http://www.perl.org/>, the Perl Home Page.
3303
3304If you believe you have an unreported bug, please run the L<perlbug> program
3305included with your release. Be sure to trim your bug down to a tiny but
3306sufficient test case. Your bug report, along with the output of C<perl -V>,
3307will be sent off to C<perlbug@perl.org> to be analysed by the Perl porting team.
3308
3309If the bug you are reporting has security implications which make it
3310inappropriate to send to a publicly archived mailing list, then see
3311L<perlsec/SECURITY VULNERABILITY CONTACT INFORMATION>
3312for details of how to report the issue.
3313
3314=head1 Give Thanks
3315
3316If you wish to thank the Perl 5 Porters for the work we had done in Perl 5,
3317you can do so by running the C<perlthanks> program:
3318
3319 perlthanks
3320
3321This will send an email to the Perl 5 Porters list with your show of thanks.
3322
3323=head1 SEE ALSO
3324
3325The F<Changes> file for an explanation of how to view exhaustive details on
3326what changed.
3327
3328The F<INSTALL> file for how to build Perl.
3329
3330The F<README> file for general stuff.
3331
3332The F<Artistic> and F<Copying> files for copyright information.
3333
3334=cut