This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perldelta: restate the changes to ${x} versus $x
[perl5.git] / pod / perldelta.pod
CommitLineData
44691e6f
AB
1=encoding utf8
2
3=head1 NAME
4
5ed58cbd 5perldelta - what is new for perl v5.18.0
e128ab2c 6
4eabcf70 7=head1 DESCRIPTION
6db9054f 8
e612b5a0 9This document describes differences between the v5.16.0 release and the v5.18.0
e08634c5 10release.
6db9054f 11
e612b5a0
RS
12If you are upgrading from an earlier release such as v5.14.0, first read
13L<perl5160delta>, which describes differences between v5.14.0 and v5.16.0.
3f01b192 14
5ed58cbd 15=head1 Core Enhancements
3f01b192 16
5ed58cbd 17=head2 New mechanism for experimental features
82d98f72 18
5ed58cbd 19Newly-added experimental features will now require this incantation:
82d98f72 20
5ed58cbd
RS
21 no warnings "experimental::feature_name";
22 use feature "feature_name"; # would warn without the prev line
23
24There is a new warnings category, called "experimental", containing
25warnings that the L<feature> pragma emits when enabling experimental
26features.
27
28Newly-added experimental features will also be given special warning IDs,
29which consist of "experimental::" followed by the name of the feature. (The
30plan is to extend this mechanism eventually to all warnings, to allow them
31to be enabled or disabled individually, and not just by category.)
32
33By saying
34
35 no warnings "experimental::feature_name";
36
37you are taking responsibility for any breakage that future changes to, or
38removal of, the feature may cause.
39
1ea0ff56
RS
40Since some features (like C<~~> or C<my $_>) now emit experimental warnings,
41and you may want to disable them in code that is also run on perls that do not
42recognize these warning categories, consider using the C<if> pragma like this:
43
5c46a73f 44 no if $] >= 5.018, 'warnings', "experimental::feature_name";
1ea0ff56 45
5ed58cbd
RS
46Existing experimental features may begin emitting these warnings, too. Please
47consult L<perlexperiment> for information on which features are considered
48experimental.
49
50=head2 Hash overhaul
51
e612b5a0 52Changes to the implementation of hashes in perl v5.18.0 will be one of the most
484d8ba1
DG
53visible changes to the behavior of existing code.
54
8e74b2ed 55By default, two distinct hash variables with identical keys and values may now
484d8ba1
DG
56provide their contents in a different order where it was previously identical.
57
58When encountering these changes, the key to cleaning up from them is to accept
59that B<hashes are unordered collections> and to act accordingly.
5ed58cbd
RS
60
61=head3 Hash randomization
62
63The seed used by Perl's hash function is now random. This means that the
64order which keys/values will be returned from functions like C<keys()>,
65C<values()>, and C<each()> will differ from run to run.
66
67This change was introduced to make Perl's hashes more robust to algorithmic
68complexity attacks, and also because we discovered that it exposes hash
69ordering dependency bugs and makes them easier to track down.
70
71Toolchain maintainers might want to invest in additional infrastructure to
72test for things like this. Running tests several times in a row and then
73comparing results will make it easier to spot hash order dependencies in
74code. Authors are strongly encouraged not to expose the key order of
75Perl's hashes to insecure audiences.
76
77Further, every hash has its own iteration order, which should make it much
78more difficult to determine what the current hash seed is.
79
0f2be12f
RS
80=head3 New hash functions
81
e612b5a0 82Perl v5.18 includes support for multiple hash functions, and changed
0f2be12f
RS
83the default (to ONE_AT_A_TIME_HARD), you can choose a different
84algorithm by defining a symbol at compile time. For a current list,
e612b5a0 85consult the F<INSTALL> document. Note that as of Perl v5.18 we can
0f2be12f
RS
86only recommend use of the default or SIPHASH. All the others are
87known to have security issues and are for research purposes only.
5ed58cbd 88
f105b7be 89=head3 PERL_HASH_SEED environment variable now takes a hex value
5ed58cbd 90
484d8ba1
DG
91C<PERL_HASH_SEED> no longer accepts an integer as a parameter;
92instead the value is expected to be a binary value encoded in a hex
93string, such as "0xf5867c55039dc724". This is to make the
94infrastructure support hash seeds of arbitrary lengths, which might
ad94908e 95exceed that of an integer. (SipHash uses a 16 byte seed.)
5ed58cbd 96
c40c48bb
RS
97=head3 PERL_PERTURB_KEYS environment variable added
98
f105b7be 99The C<PERL_PERTURB_KEYS> environment variable allows one to control the level of
c40c48bb
RS
100randomization applied to C<keys> and friends.
101
f105b7be 102When C<PERL_PERTURB_KEYS> is 0, perl will not randomize the key order at all. The
c40c48bb
RS
103chance that C<keys> changes due to an insert will be the same as in previous
104perls, basically only when the bucket size is changed.
105
f105b7be 106When C<PERL_PERTURB_KEYS> is 1, perl will randomize keys in a non-repeatable
c40c48bb
RS
107way. The chance that C<keys> changes due to an insert will be very high. This
108is the most secure and default mode.
109
f105b7be
KE
110When C<PERL_PERTURB_KEYS> is 2, perl will randomize keys in a repeatable way.
111Repeated runs of the same program should produce the same output every time.
c40c48bb 112
f105b7be
KE
113C<PERL_HASH_SEED> implies a non-default C<PERL_PERTURB_KEYS> setting. Setting
114C<PERL_HASH_SEED=0> (exactly one 0) implies C<PERL_PERTURB_KEYS=0> (hash key
115randomization disabled); settng C<PERL_HASH_SEED> to any other value implies
116C<PERL_PERTURB_KEYS=2> (deterministic and repeatable hash key randomization).
117Specifying C<PERL_PERTURB_KEYS> explicitly to a different level overrides this
c40c48bb
RS
118behavior.
119
5ed58cbd
RS
120=head3 Hash::Util::hash_seed() now returns a string
121
122Hash::Util::hash_seed() now returns a string instead of an integer. This
123is to make the infrastructure support hash seeds of arbitrary lengths
ad94908e 124which might exceed that of an integer. (SipHash uses a 16 byte seed.)
5ed58cbd
RS
125
126=head3 Output of PERL_HASH_SEED_DEBUG has been changed
127
128The environment variable PERL_HASH_SEED_DEBUG now makes perl show both the
f105b7be 129hash function perl was built with, I<and> the seed, in hex, in use for that
5ed58cbd
RS
130process. Code parsing this output, should it exist, must change to accommodate
131the new format. Example of the new format:
132
133 $ PERL_HASH_SEED_DEBUG=1 ./perl -e1
134 HASH_FUNCTION = MURMUR3 HASH_SEED = 0x1476bb9f
135
136=head2 Upgrade to Unicode 6.2
137
2e7bc647 138Perl now supports Unicode 6.2. A list of changes from Unicode
5ed58cbd
RS
1396.1 is at L<http://www.unicode.org/versions/Unicode6.2.0>.
140
141=head2 Character name aliases may now include non-Latin1-range characters
142
143It is possible to define your own names for characters for use in
144C<\N{...}>, C<charnames::vianame()>, etc. These names can now be
145comprised of characters from the whole Unicode range. This allows for
146names to be in your native language, and not just English. Certain
147restrictions apply to the characters that may be used (you can't define
148a name that has punctuation in it, for example). See L<charnames/CUSTOM
149ALIASES>.
150
151=head2 New DTrace probes
152
153The following new DTrace probes have been added:
14731ad1 154
337fb649 155=over 4
14731ad1 156
82d98f72 157=item *
14731ad1 158
5ed58cbd
RS
159C<op-entry>
160
161=item *
162
163C<loading-file>
164
165=item *
166
167C<loaded-file>
168
169=back
170
171=head2 C<${^LAST_FH}>
172
173This new variable provides access to the filehandle that was last read.
174This is the handle used by C<$.> and by C<tell> and C<eof> without
175arguments.
176
177=head2 Regular Expression Set Operations
178
179This is an B<experimental> feature to allow matching against the union,
180intersection, etc., of sets of code points, similar to
181L<Unicode::Regex::Set>. It can also be used to extend C</x> processing
182to [bracketed] character classes, and as a replacement of user-defined
183properties, allowing more complex expressions than they do. See
184L<perlrecharclass/Extended Bracketed Character Classes>.
185
186=head2 Lexical subroutines
187
188This new feature is still considered B<experimental>. To enable it:
189
190 use 5.018;
191 no warnings "experimental::lexical_subs";
192 use feature "lexical_subs";
193
194You can now declare subroutines with C<state sub foo>, C<my sub foo>, and
195C<our sub foo>. (C<state sub> requires that the "state" feature be
196enabled, unless you write it as C<CORE::state sub foo>.)
197
198C<state sub> creates a subroutine visible within the lexical scope in which
199it is declared. The subroutine is shared between calls to the outer sub.
200
201C<my sub> declares a lexical subroutine that is created each time the
202enclosing block is entered. C<state sub> is generally slightly faster than
203C<my sub>.
204
205C<our sub> declares a lexical alias to the package subroutine of the same
206name.
207
208For more information, see L<perlsub/Lexical Subroutines>.
209
210=head2 Computed Labels
211
212The loop controls C<next>, C<last> and C<redo>, and the special C<dump>
213operator, now allow arbitrary expressions to be used to compute labels at run
214time. Previously, any argument that was not a constant was treated as the
215empty string.
216
217=head2 More CORE:: subs
218
219Several more built-in functions have been added as subroutines to the
220CORE:: namespace - namely, those non-overridable keywords that can be
221implemented without custom parsers: C<defined>, C<delete>, C<exists>,
222C<glob>, C<pos>, C<protoytpe>, C<scalar>, C<split>, C<study>, and C<undef>.
223
224As some of these have prototypes, C<prototype('CORE::...')> has been
225changed to not make a distinction between overridable and non-overridable
226keywords. This is to make C<prototype('CORE::pos')> consistent with
227C<prototype(&CORE::pos)>.
228
229=head2 C<kill> with negative signal names
230
231C<kill> has always allowed a negative signal number, which kills the
232process group instead of a single process. It has also allowed signal
233names. But it did not behave consistently, because negative signal names
234were treated as 0. Now negative signals names like C<-INT> are supported
235and treated the same way as -2 [perl #112990].
236
237=head1 Security
238
32affb18
RS
239=head2 See also: hash overhaul
240
241Some of the changes in the L<hash overhaul|/"Hash overhaul"> were made to
242enhance security. Please read that section.
243
5ed58cbd
RS
244=head2 C<Storable> security warning in documentation
245
246The documentation for C<Storable> now includes a section which warns readers
247of the danger of accepting Storable documents from untrusted sources. The
248short version is that deserializing certain types of data can lead to loading
249modules and other code execution. This is documented behavior and wanted
250behavior, but this opens an attack vector for malicious entities.
251
252=head2 C<Locale::Maketext> allowed code injection via a malicious template
253
254If users could provide a translation string to Locale::Maketext, this could be
255used to invoke arbitrary Perl subroutines available in the current process.
256
257This has been fixed, but it is still possible to invoke any method provided by
258C<Locale::Maketext> itself or a subclass that you are using. One of these
259methods in turn will invoke the Perl core's C<sprintf> subroutine.
260
261In summary, allowing users to provide translation strings without auditing
262them is a bad idea.
263
264This vulnerability is documented in CVE-2012-6329.
265
266=head2 Avoid calling memset with a negative count
267
268Poorly written perl code that allows an attacker to specify the count to perl's
269C<x> string repeat operator can already cause a memory exhaustion
e612b5a0 270denial-of-service attack. A flaw in versions of perl before v5.15.5 can escalate
5ed58cbd
RS
271that into a heap buffer overrun; coupled with versions of glibc before 2.16, it
272possibly allows the execution of arbitrary code.
273
274The flaw addressed to this commit has been assigned identifier CVE-2012-5195
275and was researched by Tim Brown.
276
277=head1 Incompatible Changes
278
279=head2 See also: hash overhaul
280
281Some of the changes in the L<hash overhaul|/"Hash overhaul"> are not fully
282compatible with previous versions of perl. Please read that section.
283
284=head2 An unknown character name in C<\N{...}> is now a syntax error
285
286Previously, it warned, and the Unicode REPLACEMENT CHARACTER was
287substituted. Unicode now recommends that this situation be a syntax
288error. Also, the previous behavior led to some confusing warnings and
289behaviors, and since the REPLACEMENT CHARACTER has no use other than as
290a stand-in for some unknown character, any code that has this problem is
291buggy.
292
293=head2 Formerly deprecated characters in C<\N{}> character name aliases are now errors.
294
295Since v5.12.0, it has been deprecated to use certain characters in
296user-defined C<\N{...}> character names. These now cause a syntax
297error. For example, it is now an error to begin a name with a digit,
298such as in
299
300 my $undraftable = "\N{4F}"; # Syntax error!
301
237b9c6d 302or to have commas anywhere in the name. See L<charnames/CUSTOM ALIASES>.
5ed58cbd
RS
303
304=head2 C<\N{BELL}> now refers to U+1F514 instead of U+0007
305
306Unicode 6.0 reused the name "BELL" for a different code point than it
307traditionally had meant. Since Perl v5.14, use of this name still
308referred to U+0007, but would raise a deprecation warning. Now, "BELL"
309refers to U+1F514, and the name for U+0007 is "ALERT". All the
310functions in L<charnames> have been correspondingly updated.
311
312=head2 New Restrictions in Multi-Character Case-Insensitive Matching in Regular Expression Bracketed Character Classes
313
314Unicode has now withdrawn their previous recommendation for regular
315expressions to automatically handle cases where a single character can
237b9c6d 316match multiple characters case-insensitively, for example, the letter
5ed58cbd
RS
317LATIN SMALL LETTER SHARP S and the sequence C<ss>. This is because
318it turns out to be impracticable to do this correctly in all
319circumstances. Because Perl has tried to do this as best it can, it
320will continue to do so. (We are considering an option to turn it off.)
321However, a new restriction is being added on such matches when they
322occur in [bracketed] character classes. People were specifying
323things such as C</[\0-\xff]/i>, and being surprised that it matches the
324two character sequence C<ss> (since LATIN SMALL LETTER SHARP S occurs in
325this range). This behavior is also inconsistent with using a
326property instead of a range: C<\p{Block=Latin1}> also includes LATIN
327SMALL LETTER SHARP S, but C</[\p{Block=Latin1}]/i> does not match C<ss>.
328The new rule is that for there to be a multi-character case-insensitive
329match within a bracketed character class, the character must be
330explicitly listed, and not as an end point of a range. This more
331closely obeys the Principle of Least Astonishment. See
332L<perlrecharclass/Bracketed Character Classes>. Note that a bug [perl
333#89774], now fixed as part of this change, prevented the previous
334behavior from working fully.
335
336=head2 Explicit rules for variable names and identifiers
337
c6c6975e
RS
338Due to an oversight, single character variable names in v5.16 were
339completely unrestricted. This opened the door to several kinds of
340insanity. As of v5.18, these now follow the rules of other identifiers,
341in addition to accepting characters that match the C<\p{POSIX_Punct}>
342property.
5ed58cbd 343
31d20749
RS
344There is no longer any difference in the parsing of identifiers
345specified by using braces versus without braces. For instance, perl
346used to allow C<${foo:bar}> (with a single colon) but not C<$foo:bar>.
347Now that both are handled by a single code path, they are both treated
348the same way: both are forbidden. Note that this change is about the
349range of permissible literal identifiers, not other expressions.
5ed58cbd 350
9be53992 351=head2 Vertical tabs are now whitespace
5ed58cbd
RS
352
353No one could recall why C<\s> didn't match C<\cK>, the vertical tab.
354Now it does. Given the extreme rarity of that character, very little
9be53992
RS
355breakage is expected. That said, here's what it means:
356
357C<\s> in a regex now matches a vertical tab in all circumstances.
358
359Literal vertical tabs in a regex literal are ignored when the C</x>
360modifier is used.
361
362Leading vertical tabs, alone or mixed with other whitespace, are now
363ignored when interpreting a string as a number. For example:
364
365 $dec = " \cK \t 123";
366 $hex = " \cK \t 0xF";
367
368 say 0 + $dec; # was 0 with warning, now 123
369 say int $dec; # was 0, now 123
370 say oct $hex; # was 0, now 15
5ed58cbd
RS
371
372=head2 C</(?{})/> and C</(??{})/> have been heavily reworked
373
374The implementation of this feature has been almost completely rewritten.
375Although its main intent is to fix bugs, some behaviors, especially
376related to the scope of lexical variables, will have changed. This is
377described more fully in the L</Selected Bug Fixes> section.
378
379=head2 Stricter parsing of substitution replacement
380
381It is no longer possible to abuse the way the parser parses C<s///e> like
382this:
383
384 %_=(_,"Just another ");
385 $_="Perl hacker,\n";
386 s//_}->{_/e;print
387
388=head2 C<given> now aliases the global C<$_>
389
390Instead of assigning to an implicit lexical C<$_>, C<given> now makes the
391global C<$_> an alias for its argument, just like C<foreach>. However, it
392still uses lexical C<$_> if there is lexical C<$_> in scope (again, just like
393C<foreach>) [perl #114020].
394
64a74324
RS
395=head2 The smartmatch family of features are now experimental
396
397Smart match, added in v5.10.0 and significantly revised in v5.10.1, has been
398a regular point of complaint. Although there are a number of ways in which
399it is useful, it has also proven problematic and confusing for both users and
400implementors of Perl. There have been a number of proposals on how to best
401address the problem. It is clear that smartmatch is almost certainly either
402going to change or go away in the future. Relying on its current behavior
403is not recommended.
404
405Warnings will now be issued when the parser sees C<~~>, C<given>, or C<when>.
406To disable these warnings, you can add this line to the appropriate scope:
407
408 no if $] >= 5.018, "experimental::smartmatch";
409
410Consider, though, replacing the use of these features, as they may change
411behavior again before becoming stable.
412
5ed58cbd
RS
413=head2 Lexical C<$_> is now experimental
414
e612b5a0 415Since it was introduced in Perl v5.10, it has caused much confusion with no
5ed58cbd
RS
416obvious solution:
417
418=over
829397b9
TC
419
420=item *
421
5ed58cbd
RS
422Various modules (e.g., List::Util) expect callback routines to use the
423global C<$_>. C<use List::Util 'first'; my $_; first { $_ == 1 } @list>
424does not work as one would expect.
71e6aba6
RS
425
426=item *
427
5ed58cbd
RS
428A C<my $_> declaration earlier in the same file can cause confusing closure
429warnings.
71e6aba6
RS
430
431=item *
432
5ed58cbd
RS
433The "_" subroutine prototype character allows called subroutines to access
434your lexical C<$_>, so it is not really private after all.
2426c394 435
71e6aba6
RS
436=item *
437
5ed58cbd
RS
438Nevertheless, subroutines with a "(@)" prototype and methods cannot access
439the caller's lexical C<$_>, unless they are written in XS.
71e6aba6
RS
440
441=item *
442
5ed58cbd
RS
443But even XS routines cannot access a lexical C<$_> declared, not in the
444calling subroutine, but in an outer scope, iff that subroutine happened not
445to mention C<$_> or use any operators that default to C<$_>.
446
447=back
448
449It is our hope that lexical C<$_> can be rehabilitated, but this may
450cause changes in its behavior. Please use it with caution until it
451becomes stable.
452
453=head2 readline() with C<$/ = \N> now reads N characters, not N bytes
454
455Previously, when reading from a stream with I/O layers such as
456C<encoding>, the readline() function, otherwise known as the C<< <> >>
457operator, would read I<N> bytes from the top-most layer. [perl #79960]
458
459Now, I<N> characters are read instead.
460
461There is no change in behaviour when reading from streams with no
462extra layers, since bytes map exactly to characters.
463
464=head2 Overridden C<glob> is now passed one argument
465
466C<glob> overrides used to be passed a magical undocumented second argument
467that identified the caller. Nothing on CPAN was using this, and it got in
468the way of a bug fix, so it was removed. If you really need to identify
469the caller, see L<Devel::Callsite> on CPAN.
470
ef4b82dd 471=head2 Here doc parsing
5ed58cbd 472
ef4b82dd 473The body of a here document inside a quote-like operator now always begins
5ed58cbd
RS
474on the line after the "<<foo" marker. Previously, it was documented to
475begin on the line following the containing quote-like operator, but that
476was only sometimes the case [perl #114040].
477
478=head2 Alphanumeric operators must now be separated from the closing
479delimiter of regular expressions
480
481You may no longer write something like:
482
483 m/a/and 1
484
485Instead you must write
486
487 m/a/ and 1
488
489with whitespace separating the operator from the closing delimiter of
490the regular expression. Not having whitespace has resulted in a
491deprecation warning since Perl v5.14.0.
492
493=head2 qw(...) can no longer be used as parentheses
494
495C<qw> lists used to fool the parser into thinking they were always
496surrounded by parentheses. This permitted some surprising constructions
497such as C<foreach $x qw(a b c) {...}>, which should really be written
498C<foreach $x (qw(a b c)) {...}>. These would sometimes get the lexer into
499the wrong state, so they didn't fully work, and the similar C<foreach qw(a
500b c) {...}> that one might expect to be permitted never worked at all.
501
502This side effect of C<qw> has now been abolished. It has been deprecated
e612b5a0 503since Perl v5.13.11. It is now necessary to use real parentheses
5ed58cbd
RS
504everywhere that the grammar calls for them.
505
506=head2 Interaction of lexical and default warnings
507
508Turning on any lexical warnings used first to disable all default warnings
509if lexical warnings were not already enabled:
510
511 $*; # deprecation warning
512 use warnings "void";
513 $#; # void warning; no deprecation warning
514
f105b7be 515Now, the C<debugging>, C<deprecated>, C<glob>, C<inplace> and C<malloc> warnings
5ed58cbd
RS
516categories are left on when turning on lexical warnings (unless they are
517turned off by C<no warnings>, of course).
518
519This may cause deprecation warnings to occur in code that used to be free
520of warnings.
521
522Those are the only categories consisting only of default warnings. Default
f105b7be
KE
523warnings in other categories are still disabled by C<< use warnings "category" >>,
524as we do not yet have the infrastructure for controlling
5ed58cbd
RS
525individual warnings.
526
527=head2 C<state sub> and C<our sub>
528
529Due to an accident of history, C<state sub> and C<our sub> were equivalent
530to a plain C<sub>, so one could even create an anonymous sub with
531C<our sub { ... }>. These are now disallowed outside of the "lexical_subs"
532feature. Under the "lexical_subs" feature they have new meanings described
533in L<perlsub/Lexical Subroutines>.
534
535=head2 Defined values stored in environment are forced to byte strings
536
537A value stored in an environment variable has always been stringified. In this
4f5e76eb 538release, it is converted to be only a byte string. First, it is forced to be
5ed58cbd
RS
539only a string. Then if the string is utf8 and the equivalent of
540C<utf8::downgrade()> works, that result is used; otherwise, the equivalent of
541C<utf8::encode()> is used, and a warning is issued about wide characters
542(L</Diagnostics>).
543
544=head2 C<require> dies for unreadable files
545
546When C<require> encounters an unreadable file, it now dies. It used to
547ignore the file and continue searching the directories in C<@INC>
548[perl #113422].
549
550=head2 C<gv_fetchmeth_*> and SUPER
551
552The various C<gv_fetchmeth_*> XS functions used to treat a package whose
f105b7be
KE
553named ended with C<::SUPER> specially. A method lookup on the C<Foo::SUPER>
554package would be treated as a C<SUPER> method lookup on the C<Foo> package. This
555is no longer the case. To do a C<SUPER> lookup, pass the C<Foo> stash and the
556C<GV_SUPER> flag.
5ed58cbd
RS
557
558=head2 C<split>'s first argument is more consistently interpreted
559
e612b5a0 560After some changes earlier in v5.17, C<split>'s behavior has been
3853dabd 561simplified: if the PATTERN argument evaluates to a string
5ed58cbd
RS
562containing one space, it is treated the way that a I<literal> string
563containing one space once was.
564
565=head1 Deprecations
566
c68b7fe2 567=head2 Module removals
5ed58cbd 568
c68b7fe2
AP
569The following modules will be removed from the core distribution in a future
570release, and will at that time need to be installed from CPAN. Distributions
571on CPAN which require these modules will need to list them as prerequisites.
572
573The core versions of these modules will now issue C<"deprecated">-category
574warnings to alert you to this fact. To silence these deprecation warnings,
575install the modules in question from CPAN.
5ed58cbd 576
c68b7fe2
AP
577Note that these are (with rare exceptions) fine modules that you are encouraged
578to continue to use. Their disinclusion from core primarily hinges on their
579necessity to bootstrapping a fully functional, CPAN-capable Perl installation,
580not usually on concerns over their design.
5ed58cbd
RS
581
582=over
583
c68b7fe2
AP
584=item L<encoding>
585
586The use of this pragma is now strongly discouraged. It conflates the encoding
587of source text with the encoding of I/O data, reinterprets escape sequences in
588source text (a questionable choice), and introduces the UTF-8 bug to all runtime
589handling of character strings. It is broken as designed and beyond repair.
590
591For using non-ASCII literal characters in source text, please refer to L<utf8>.
592For dealing with textual I/O data, please refer to L<Encode> and L<open>.
593
5ed58cbd
RS
594=item L<Archive::Extract>
595
596=item L<B::Lint>
597
598=item L<B::Lint::Debug>
599
600=item L<CPANPLUS> and all included C<CPANPLUS::*> modules
601
602=item L<Devel::InnerPackage>
603
5ed58cbd
RS
604=item L<Log::Message>
605
606=item L<Log::Message::Config>
607
608=item L<Log::Message::Handlers>
609
610=item L<Log::Message::Item>
611
612=item L<Log::Message::Simple>
613
614=item L<Module::Pluggable>
615
616=item L<Module::Pluggable::Object>
617
618=item L<Object::Accessor>
619
620=item L<Pod::LaTeX>
621
622=item L<Term::UI>
623
624=item L<Term::UI::History>
625
626=back
627
628=head2 Deprecated Utilities
629
630The following utilities will be removed from the core distribution in a
631future release as their associated modules have been deprecated. They
632will remain available with the applicable CPAN distribution.
633
634=over
635
636=item L<cpanp>
637
638=item C<cpanp-run-perl>
639
640=item L<cpan2dist>
641
642These items are part of the C<CPANPLUS> distribution.
643
644=item L<pod2latex>
645
646This item is part of the C<Pod::LaTeX> distribution.
647
648=back
649
650=head2 PL_sv_objcount
651
652This interpreter-global variable used to track the total number of
653Perl objects in the interpreter. It is no longer maintained and will
e612b5a0 654be removed altogether in Perl v5.20.
5ed58cbd
RS
655
656=head2 Five additional characters should be escaped in patterns with C</x>
657
658When a regular expression pattern is compiled with C</x>, Perl treats 6
659characters as white space to ignore, such as SPACE and TAB. However,
660Unicode recommends 11 characters be treated thusly. We will conform
661with this in a future Perl version. In the meantime, use of any of the
662missing characters will raise a deprecation warning, unless turned off.
663The five characters are:
f105b7be 664
c163876c
RS
665 U+0085 NEXT LINE
666 U+200E LEFT-TO-RIGHT MARK
667 U+200F RIGHT-TO-LEFT MARK
668 U+2028 LINE SEPARATOR
669 U+2029 PARAGRAPH SEPARATOR
5ed58cbd
RS
670
671=head2 User-defined charnames with surprising whitespace
672
673A user-defined character name with trailing or multiple spaces in a row is
674likely a typo. This now generates a warning when defined, on the assumption
675that uses of it will be unlikely to include the excess whitespace.
676
677=head2 Various XS-callable functions are now deprecated
678
679All the functions used to classify characters will be removed from a
680future version of Perl, and should not be used. With participating C
681compilers (e.g., gcc), compiling any file that uses any of these will
682generate a warning. These were not intended for public use; there are
683equivalent, faster, macros for most of them.
e612b5a0 684
2e7bc647 685See L<perlapi/Character classes>. The complete list is:
e612b5a0 686
5ed58cbd
RS
687C<is_uni_alnum>, C<is_uni_alnumc>, C<is_uni_alnumc_lc>,
688C<is_uni_alnum_lc>, C<is_uni_alpha>, C<is_uni_alpha_lc>,
689C<is_uni_ascii>, C<is_uni_ascii_lc>, C<is_uni_blank>,
690C<is_uni_blank_lc>, C<is_uni_cntrl>, C<is_uni_cntrl_lc>,
691C<is_uni_digit>, C<is_uni_digit_lc>, C<is_uni_graph>,
692C<is_uni_graph_lc>, C<is_uni_idfirst>, C<is_uni_idfirst_lc>,
693C<is_uni_lower>, C<is_uni_lower_lc>, C<is_uni_print>,
694C<is_uni_print_lc>, C<is_uni_punct>, C<is_uni_punct_lc>,
695C<is_uni_space>, C<is_uni_space_lc>, C<is_uni_upper>,
696C<is_uni_upper_lc>, C<is_uni_xdigit>, C<is_uni_xdigit_lc>,
697C<is_utf8_alnum>, C<is_utf8_alnumc>, C<is_utf8_alpha>,
698C<is_utf8_ascii>, C<is_utf8_blank>, C<is_utf8_char>,
699C<is_utf8_cntrl>, C<is_utf8_digit>, C<is_utf8_graph>,
700C<is_utf8_idcont>, C<is_utf8_idfirst>, C<is_utf8_lower>,
701C<is_utf8_mark>, C<is_utf8_perl_space>, C<is_utf8_perl_word>,
702C<is_utf8_posix_digit>, C<is_utf8_print>, C<is_utf8_punct>,
703C<is_utf8_space>, C<is_utf8_upper>, C<is_utf8_xdigit>,
704C<is_utf8_xidcont>, C<is_utf8_xidfirst>.
705
706In addition these three functions that have never worked properly are
707deprecated:
708C<to_uni_lower_lc>, C<to_uni_title_lc>, and C<to_uni_upper_lc>.
709
f105b7be 710=head2 Certain rare uses of backslashes within regexes are now deprecated
5ed58cbd
RS
711
712There are three pairs of characters that Perl recognizes as
713metacharacters in regular expression patterns: C<{}>, C<[]>, and C<()>.
714These can be used as well to delimit patterns, as in:
715
716 m{foo}
717 s(foo)(bar)
718
719Since they are metacharacters, they have special meaning to regular
720expression patterns, and it turns out that you can't turn off that
721special meaning by the normal means of preceding them with a backslash,
f105b7be 722if you use them, paired, within a pattern delimited by them. For
5ed58cbd
RS
723example, in
724
725 m{foo\{1,3\}}
726
727the backslashes do not change the behavior, and this matches
728S<C<"f o">> followed by one to three more occurrences of C<"o">.
729
730Usages like this, where they are interpreted as metacharacters, are
731exceedingly rare; we think there are none, for example, in all of CPAN.
732Hence, this deprecation should affect very little code. It does give
733notice, however, that any such code needs to change, which will in turn
734allow us to change the behavior in future Perl versions so that the
735backslashes do have an effect, and without fear that we are silently
736breaking any existing code.
737
d5f315e8
KW
738=head2 Splitting the tokens C<(?> and C<(*> in regular expressions
739
740A deprecation warning is now raised if the C<(> and C<?> are separated
741by white space or comments in C<(?...)> regular expression constructs.
742Similarly, if the C<(> and C<*> are separated in C<(*VERB...)>
743constructs.
744
e0a1dec5
LT
745=head2 Pre-PerlIO IO implementations
746
f3068e1b
RS
747In theory, you can currently build perl without PerlIO. Instead, you'd use a
748wrapper around stdio or sfio. In practice, this isn't very useful. It's not
749well tested, and without any support for IO layers or (thus) Unicode, it's not
750much of a perl. Building without PerlIO will most likely be removed in the
751next version of perl.
752
753PerlIO supports a C<stdio> layer if stdio use is desired. Similarly a
754sfio layer could be produced in the future, if needed.
e0a1dec5 755
5ed58cbd
RS
756=head1 Future Deprecations
757
758=over
71e6aba6
RS
759
760=item *
761
4263dd11 762Platforms without support infrastructure
5ed58cbd
RS
763
764Both Windows CE and z/OS have been historically under-maintained, and are
765currently neither successfully building nor regularly being smoke tested.
766Efforts are underway to change this situation, but it should not be taken for
767granted that the platforms are safe and supported. If they do not become
768buildable and regularly smoked, support for them may be actively removed in
769future releases. If you have an interest in these platforms and you can lend
770your time, expertise, or hardware to help support these platforms, please let
771the perl development effort know by emailing C<perl5-porters@perl.org>.
772
773Some platforms that appear otherwise entirely dead are also on the short list
e612b5a0 774for removal between now and v5.20.0:
5ed58cbd
RS
775
776=over
777
778=item DG/UX
779
780=item NeXT
781
782=back
1993add8 783
ec985017
RS
784We also think it likely that current versions of Perl will no longer
785build AmigaOS, DJGPP, NetWare (natively), OS/2 and Plan 9. If you
786are using Perl on such a platform and have an interest in ensuring
787Perl's future on them, please contact us.
788
789We believe that Perl has long been unable to build on mixed endian
ad94908e 790architectures (such as PDP-11s), and intend to remove any remaining
ec985017
RS
791support code. Similarly, code supporting the long umaintained GNU
792dld will be removed soon if no-one makes themselves known as an
793active user.
794
1993add8
RS
795=item *
796
5ed58cbd
RS
797Swapping of $< and $>
798
064fb797
LT
799Perl has supported the idiom of swapping $< and $> (and likewise $( and
800$)) to temporarily drop permissions since 5.0, like this:
801
802 ($<, $>) = ($>, $<);
803
804However, this idiom modifies the real user/group id, which can have
805undesirable side-effects, is no longer useful on any platform perl
806supports and complicates the implementation of these variables and list
807assignment in general.
808
809As an alternative, assignment only to C<< $> >> is recommended:
810
811 local $> = $<;
812
813See also: L<Setuid Demystified|http://www.cs.berkeley.edu/~daw/papers/setuid-usenix02.pdf>.
71e6aba6
RS
814
815=item *
816
5ed58cbd 817C<microperl>, long broken and of unclear present purpose, will be removed.
71e6aba6
RS
818
819=item *
820
5ed58cbd
RS
821Revamping C<< "\Q" >> semantics in double-quotish strings when combined with
822other escapes.
823
824There are several bugs and inconsistencies involving combinations
825of C<\Q> and escapes like C<\x>, C<\L>, etc., within a C<\Q...\E> pair.
826These need to be fixed, and doing so will necessarily change current
827behavior. The changes have not yet been settled.
71e6aba6 828
d5f315e8
KW
829=item *
830
9c34b375
ZA
831Use of C<$x>, where C<x> stands for any actual (non-printing) C0 control
832character will be disallowed in a future Perl version. Use C<${x}>
833instead (where again C<x> stands for a control character),
834or better, C<$^A> , where C<^> is a caret (CIRCUMFLEX ACCENT),
d5f315e8
KW
835and C<A> stands for any of the characters listed at the end of
836L<perlebcdic/OPERATOR DIFFERENCES>.
837
337fb649 838=back
2426c394 839
5ed58cbd
RS
840=head1 Performance Enhancements
841
842=over 4
2426c394 843
5ed58cbd 844=item *
2426c394 845
5ed58cbd
RS
846Lists of lexical variable declarations (C<my($x, $y)>) are now optimised
847down to a single op and are hence faster than before.
2426c394 848
5ed58cbd 849=item *
2426c394 850
5ed58cbd
RS
851A new C preprocessor define C<NO_TAINT_SUPPORT> was added that, if set,
852disables Perl's taint support altogether. Using the -T or -t command
853line flags will cause a fatal error. Beware that both core tests as
854well as many a CPAN distribution's tests will fail with this change. On
855the upside, it provides a small performance benefit due to reduced
856branching.
2426c394 857
5ed58cbd
RS
858B<Do not enable this unless you know exactly what you are getting yourself
859into.>
860
861=item *
862
863C<pack> with constant arguments is now constant folded in most cases
864[perl #113470].
865
866=item *
867
868Speed up in regular expression matching against Unicode properties. The
869largest gain is for C<\X>, the Unicode "extended grapheme cluster." The
870gain for it is about 35% - 40%. Bracketed character classes, e.g.,
871C<[0-9\x{100}]> containing code points above 255 are also now faster.
872
873=item *
874
875On platforms supporting it, several former macros are now implemented as static
876inline functions. This should speed things up slightly on non-GCC platforms.
877
878=item *
879
66f62cf6
RS
880The optimisation of hashes in boolean context has been extended to
881affect C<scalar(%hash)>, C<%hash ? ... : ...>, and C<sub { %hash || ... }>.
5ed58cbd
RS
882
883=item *
884
f105b7be 885Filetest operators manage the stack in a fractionally more efficient manner.
5ed58cbd
RS
886
887=item *
888
889Globs used in a numeric context are now numified directly in most cases,
f105b7be 890rather than being numified via stringification.
5ed58cbd
RS
891
892=item *
893
894The C<x> repetition operator is now folded to a single constant at compile
895time if called in scalar context with constant operands and no parentheses
896around the left operand.
897
898=back
899
900=head1 Modules and Pragmata
901
902=head2 New Modules and Pragmata
2426c394 903
337fb649 904=over 4
982110e0 905
82d98f72 906=item *
2426c394 907
5ed58cbd
RS
908L<Config::Perl::V> version 0.16 has been added as a dual-lifed module.
909It provides structured data retrieval of C<perl -V> output including
910information only known to the C<perl> binary and not available via L<Config>.
911
912=back
913
914=head2 Updated Modules and Pragmata
915
9863976f 916For a complete list of updates, run:
5ed58cbd 917
f105b7be 918 $ corelist --diff 5.16.0 5.18.0
5ed58cbd 919
e612b5a0 920You can substitute your favorite version in place of C<5.16.0>, too.
5ed58cbd 921
357b01df
AAS
922=over
923
924=item *
925
606e21ea 926L<Archive::Extract> has been upgraded to 0.68.
357b01df
AAS
927
928Work around an edge case on Linux with Busybox's unzip.
929
930=item *
931
606e21ea 932L<Archive::Tar> has been upgraded to 1.90.
357b01df
AAS
933
934ptar now supports the -T option as well as dashless options
935[rt.cpan.org #75473], [rt.cpan.org #75475].
936
937Auto-encode filenames marked as UTF-8 [rt.cpan.org #75474].
938
939Don't use C<tell> on L<IO::Zlib> handles [rt.cpan.org #64339].
940
941Don't try to C<chown> on symlinks.
942
943=item *
944
606e21ea 945L<autodie> has been upgraded to 2.13.
357b01df
AAS
946
947C<autodie> now plays nicely with the 'open' pragma.
948
949=item *
950
606e21ea 951L<B> has been upgraded to 1.42.
357b01df
AAS
952
953The C<stashoff> method of COPs has been added. This provides access to an
954internal field added in perl 5.16 under threaded builds [perl #113034].
955
956C<B::COP::stashpv> now supports UTF-8 package names and embedded NULs.
957
958All C<CVf_*> and C<GVf_*>
959and more SV-related flag values are now provided as constants in the C<B::>
960namespace and available for export. The default export list has not changed.
961
962This makes the module work with the new pad API.
963
964=item *
965
606e21ea 966L<B::Concise> has been upgraded to 0.95.
357b01df
AAS
967
968The C<-nobanner> option has been fixed, and C<format>s can now be dumped.
969When passed a sub name to dump, it will check also to see whether it
970is the name of a format. If a sub and a format share the same name,
971it will dump both.
972
973This adds support for the new C<OpMAYBE_TRUEBOOL> and C<OPpTRUEBOOL> flags.
974
975=item *
976
606e21ea 977L<B::Debug> has been upgraded to 1.18.
357b01df
AAS
978
979This adds support (experimentally) for C<B::PADLIST>, which will be
980added in Perl 5.17.4.
981
982=item *
983
606e21ea 984L<B::Deparse> has been upgraded to 1.20.
357b01df
AAS
985
986Avoid warning when run under C<perl -w>.
987
988It now deparses
989loop controls with the correct precedence, and multiple statements in a
990C<format> line are also now deparsed correctly.
991
992This release suppresses trailing semicolons in formats.
993
994This release adds stub deparsing for lexical subroutines.
995
996It no longer dies when deparsing C<sort> without arguments. It now
997correctly omits the comma for C<system $prog @args> and C<exec $prog
998@args>.
999
1000=item *
1001
606e21ea 1002L<bignum>, L<bigint> and L<bigrat> have been upgraded to 0.33.
357b01df
AAS
1003
1004The overrides for C<hex> and C<oct> have been rewritten, eliminating
1005several problems, and making one incompatible change:
1006
1007=over
1008
1009=item *
1010
1011Formerly, whichever of C<use bigint> or C<use bigrat> was compiled later
1012would take precedence over the other, causing C<hex> and C<oct> not to
1013respect the other pragma when in scope.
1014
1015=item *
1016
1017Using any of these three pragmata would cause C<hex> and C<oct> anywhere
1018else in the program to evalute their arguments in list context and prevent
1019them from inferring $_ when called without arguments.
1020
1021=item *
1022
1023Using any of these three pragmata would make C<oct("1234")> return 1234
1024(for any number not beginning with 0) anywhere in the program. Now "1234"
1025is translated from octal to decimal, whether within the pragma's scope or
1026not.
1027
1028=item *
1029
1030The global overrides that facilitate lexical use of C<hex> and C<oct> now
1031respect any existing overrides that were in place before the new overrides
1032were installed, falling back to them outside of the scope of C<use bignum>.
1033
1034=item *
1035
1036C<use bignum "hex">, C<use bignum "oct"> and similar invocations for bigint
1037and bigrat now export a C<hex> or C<oct> function, instead of providing a
1038global override.
1039
1040=back
1041
1042=item *
1043
a0932e7e 1044L<Carp> has been upgraded to 1.29.
357b01df
AAS
1045
1046Carp is no longer confused when C<caller> returns undef for a package that
1047has been deleted.
1048
1049The C<longmess()> and C<shortmess()> functions are now documented.
1050
1051=item *
1052
606e21ea 1053L<CGI> has been upgraded to 3.63.
357b01df
AAS
1054
1055Unrecognized HTML escape sequences are now handled better, problematic
1056trailing newlines are no longer inserted after E<lt>formE<gt> tags
1057by C<startform()> or C<start_form()>, and bogus "Insecure Dependency"
1058warnings appearing with some versions of perl are now worked around.
1059
1060=item *
1061
606e21ea 1062L<Class::Struct> has been upgraded to 0.64.
357b01df
AAS
1063
1064The constructor now respects overridden accessor methods [perl #29230].
1065
1066=item *
1067
606e21ea 1068L<Compress::Raw::Bzip2> has been upgraded to 2.060.
357b01df
AAS
1069
1070The misuse of Perl's "magic" API has been fixed.
1071
1072=item *
1073
606e21ea 1074L<Compress::Raw::Zlib> has been upgraded to 2.060.
357b01df
AAS
1075
1076Upgrade bundled zlib to version 1.2.7.
1077
1078Fix build failures on Irix, Solaris, and Win32, and also when building as C++
1079[rt.cpan.org #69985], [rt.cpan.org #77030], [rt.cpan.org #75222].
1080
1081The misuse of Perl's "magic" API has been fixed.
1082
1083C<compress()>, C<uncompress()>, C<memGzip()> and C<memGunzip()> have
1084been speeded up by making parameter validation more efficient.
1085
1086=item *
1087
606e21ea 1088L<CPAN::Meta::Requirements> has been upgraded to 2.122.
357b01df
AAS
1089
1090Treat undef requirements to C<from_string_hash> as 0 (with a warning).
1091
1092Added C<requirements_for_module> method.
1093
1094=item *
1095
606e21ea 1096L<CPANPLUS> has been upgraded to 0.9135.
357b01df
AAS
1097
1098Allow adding F<blib/script> to PATH.
1099
1100Save the history between invocations of the shell.
1101
1102Handle multiple C<makemakerargs> and C<makeflags> arguments better.
1103
1104This resolves issues with the SQLite source engine.
1105
1106=item *
1107
606e21ea 1108L<Data::Dumper> has been upgraded to 2.145.
357b01df
AAS
1109
1110It has been optimized to only build a seen-scalar hash as necessary,
1111thereby speeding up serialization drastically.
1112
1113Additional tests were added in order to improve statement, branch, condition
1114and subroutine coverage. On the basis of the coverage analysis, some of the
1115internals of Dumper.pm were refactored. Almost all methods are now
1116documented.
1117
1118=item *
1119
606e21ea 1120L<DB_File> has been upgraded to 1.827.
357b01df
AAS
1121
1122The main Perl module no longer uses the C<"@_"> construct.
1123
1124=item *
1125
606e21ea 1126L<Devel::Peek> has been upgraded to 1.11.
357b01df
AAS
1127
1128This fixes compilation with C++ compilers and makes the module work with
1129the new pad API.
1130
1131=item *
1132
606e21ea 1133L<Digest::MD5> has been upgraded to 2.52.
357b01df
AAS
1134
1135Fix C<Digest::Perl::MD5> OO fallback [rt.cpan.org #66634].
1136
1137=item *
1138
606e21ea 1139L<Digest::SHA> has been upgraded to 5.84.
357b01df
AAS
1140
1141This fixes a double-free bug, which might have caused vulnerabilities
1142in some cases.
1143
1144=item *
1145
606e21ea 1146L<DynaLoader> has been upgraded to 1.18.
357b01df
AAS
1147
1148This is due to a minor code change in the XS for the VMS implementation.
1149
1150This fixes warnings about using C<CODE> sections without an C<OUTPUT>
1151section.
1152
1153=item *
1154
606e21ea 1155L<Encode> has been upgraded to 2.49.
357b01df
AAS
1156
1157The Mac alias x-mac-ce has been added, and various bugs have been fixed
1158in Encode::Unicode, Encode::UTF7 and Encode::GSM0338.
1159
1160=item *
1161
606e21ea 1162L<Env> has been upgraded to 1.04.
357b01df
AAS
1163
1164Its SPLICE implementation no longer misbehaves in list context.
1165
1166=item *
1167
606e21ea 1168L<ExtUtils::CBuilder> has been upgraded to 0.280210.
357b01df
AAS
1169
1170Manifest files are now correctly embedded for those versions of VC++ which
1171make use of them. [perl #111782, #111798].
1172
1173A list of symbols to export can now be passed to C<link()> when on
1174Windows, as on other OSes [perl #115100].
1175
1176=item *
1177
606e21ea 1178L<ExtUtils::ParseXS> has been upgraded to 3.18.
357b01df
AAS
1179
1180The generated C code now avoids unnecessarily incrementing
1181C<PL_amagic_generation> on Perl versions where it's done automatically
1182(or on current Perl where the variable no longer exists).
1183
1184This avoids a bogus warning for initialised XSUB non-parameters [perl
1185#112776].
1186
1187=item *
1188
606e21ea 1189L<File::Copy> has been upgraded to 2.26.
357b01df
AAS
1190
1191C<copy()> no longer zeros files when copying into the same directory,
1192and also now fails (as it has long been documented to do) when attempting
1193to copy a file over itself.
1194
1195=item *
1196
606e21ea 1197L<File::DosGlob> has been upgraded to 1.10.
357b01df
AAS
1198
1199The internal cache of file names that it keeps for each caller is now
1200freed when that caller is freed. This means
1201C<< use File::DosGlob 'glob'; eval 'scalar <*>' >> no longer leaks memory.
1202
1203=item *
1204
606e21ea 1205L<File::Fetch> has been upgraded to 0.38.
357b01df
AAS
1206
1207Added the 'file_default' option for URLs that do not have a file
1208component.
1209
1210Use C<File::HomeDir> when available, and provide C<PERL5_CPANPLUS_HOME> to
1211override the autodetection.
1212
1213Always re-fetch F<CHECKSUMS> if C<fetchdir> is set.
1214
1215=item *
1216
606e21ea 1217L<File::Find> has been upgraded to 1.23.
357b01df
AAS
1218
1219This fixes inconsistent unixy path handling on VMS.
1220
1221Individual files may now appear in list of directories to be searched
1222[perl #59750].
1223
1224=item *
1225
606e21ea 1226L<File::Glob> has been upgraded to 1.20.
357b01df
AAS
1227
1228File::Glob has had exactly the same fix as File::DosGlob. Since it is
1229what Perl's own C<glob> operator itself uses (except on VMS), this means
1230C<< eval 'scalar <*>' >> no longer leaks.
1231
1232A space-separated list of patterns return long lists of results no longer
1233results in memory corruption or crashes. This bug was introduced in
1234Perl 5.16.0. [perl #114984]
1235
1236=item *
1237
606e21ea 1238L<File::Spec::Unix> has been upgraded to 3.40.
357b01df
AAS
1239
1240C<abs2rel> could produce incorrect results when given two relative paths or
1241the root directory twice [perl #111510].
1242
1243=item *
1244
606e21ea 1245L<File::stat> has been upgraded to 1.07.
357b01df
AAS
1246
1247C<File::stat> ignores the L<filetest> pragma, and warns when used in
1248combination therewith. But it was not warning for C<-r>. This has been
1249fixed [perl #111640].
1250
1251C<-p> now works, and does not return false for pipes [perl #111638].
1252
1253Previously C<File::stat>'s overloaded C<-x> and C<-X> operators did not give
1254the correct results for directories or executable files when running as
1255root. They had been treating executable permissions for root just like for
1256any other user, performing group membership tests I<etc> for files not owned
1257by root. They now follow the correct Unix behaviour - for a directory they
1258are always true, and for a file if any of the three execute permission bits
1259are set then they report that root can execute the file. Perl's builtin
1260C<-x> and C<-X> operators have always been correct.
1261
1262=item *
1263
606e21ea 1264L<File::Temp> has been upgraded to 0.23
357b01df
AAS
1265
1266Fixes various bugs involving directory removal. Defers unlinking tempfiles if
1267the initial unlink fails, which fixes problems on NFS.
1268
1269=item *
1270
606e21ea 1271L<GDBM_File> has been upgraded to 1.15.
357b01df
AAS
1272
1273The undocumented optional fifth parameter to C<TIEHASH> has been
1274removed. This was intended to provide control of the callback used by
1275C<gdbm*> functions in case of fatal errors (such as filesystem problems),
1276but did not work (and could never have worked). No code on CPAN even
1277attempted to use it. The callback is now always the previous default,
1278C<croak>. Problems on some platforms with how the C<C> C<croak> function
1279is called have also been resolved.
1280
1281=item *
1282
606e21ea 1283L<Hash::Util> has been upgraded to 0.15.
357b01df
AAS
1284
1285C<hash_unlocked> and C<hashref_unlocked> now returns true if the hash is
1286unlocked, instead of always returning false [perl #112126].
1287
1288C<hash_unlocked>, C<hashref_unlocked>, C<lock_hash_recurse> and
1289C<unlock_hash_recurse> are now exportable [perl #112126].
1290
1291Two new functions, C<hash_locked> and C<hashref_locked>, have been added.
1292Oddly enough, these two functions were already exported, even though they
1293did not exist [perl #112126].
1294
1295=item *
1296
606e21ea 1297L<HTTP::Tiny> has been upgraded to 0.025.
357b01df
AAS
1298
1299Add SSL verification features [github #6], [github #9].
1300
1301Include the final URL in the response hashref.
1302
1303Add C<local_address> option.
1304
1305This improves SSL support.
1306
1307=item *
1308
606e21ea 1309L<IO> has been upgraded to 1.28.
357b01df
AAS
1310
1311C<sync()> can now be called on read-only file handles [perl #64772].
1312
1313L<IO::Socket> tries harder to cache or otherwise fetch socket
1314information.
1315
1316=item *
1317
606e21ea 1318L<IPC::Cmd> has been upgraded to 0.80.
357b01df
AAS
1319
1320Use C<POSIX::_exit> instead of C<exit> in C<run_forked> [rt.cpan.org #76901].
1321
1322=item *
1323
606e21ea 1324L<IPC::Open3> has been upgraded to 1.13.
357b01df
AAS
1325
1326The C<open3()> function no longer uses C<POSIX::close()> to close file
1327descriptors since that breaks the ref-counting of file descriptors done by
1328PerlIO in cases where the file descriptors are shared by PerlIO streams,
1329leading to attempts to close the file descriptors a second time when
1330any such PerlIO streams are closed later on.
1331
1332=item *
1333
606e21ea 1334L<Locale::Codes> has been upgraded to 3.25.
357b01df
AAS
1335
1336It includes some new codes.
1337
1338=item *
1339
606e21ea 1340L<Memoize> has been upgraded to 1.03.
357b01df
AAS
1341
1342Fix the C<MERGE> cache option.
1343
1344=item *
1345
606e21ea 1346L<Module::Build> has been upgraded to 0.4003.
357b01df
AAS
1347
1348Fixed bug where modules without C<$VERSION> might have a version of '0' listed
1349in 'provides' metadata, which will be rejected by PAUSE.
1350
1351Fixed bug in PodParser to allow numerals in module names.
1352
1353Fixed bug where giving arguments twice led to them becoming arrays, resulting
1354in install paths like F<ARRAY(0xdeadbeef)/lib/Foo.pm>.
1355
1356A minor bug fix allows markup to be used around the leading "Name" in
1357a POD "abstract" line, and some documentation improvements have been made.
1358
1359=item *
1360
606e21ea 1361L<Module::CoreList> has been upgraded to 2.90
357b01df
AAS
1362
1363Version information is now stored as a delta, which greatly reduces the
1364size of the F<CoreList.pm> file.
1365
1366This restores compatibility with older versions of perl and cleans up
1367the corelist data for various modules.
1368
1369=item *
1370
606e21ea 1371L<Module::Load::Conditional> has been upgraded to 0.54.
357b01df
AAS
1372
1373Fix use of C<requires> on perls installed to a path with spaces.
1374
1375Various enhancements include the new use of Module::Metadata.
1376
1377=item *
1378
606e21ea 1379L<Module::Metadata> has been upgraded to 1.000011.
357b01df
AAS
1380
1381The creation of a Module::Metadata object for a typical module file has
1382been sped up by about 40%, and some spurious warnings about C<$VERSION>s
1383have been suppressed.
1384
1385=item *
1386
606e21ea 1387L<Module::Pluggable> has been upgraded to 4.7.
357b01df
AAS
1388
1389Amongst other changes, triggers are now allowed on events, which gives
1390a powerful way to modify behaviour.
1391
1392=item *
1393
606e21ea 1394L<Net::Ping> has been upgraded to 2.41.
357b01df
AAS
1395
1396This fixes some test failures on Windows.
1397
1398=item *
1399
606e21ea 1400L<Opcode> has been upgraded to 1.25.
357b01df
AAS
1401
1402Reflect the removal of the boolkeys opcode and the addition of the
1403clonecv, introcv and padcv opcodes.
1404
1405=item *
1406
606e21ea 1407L<overload> has been upgraded to 1.22.
357b01df
AAS
1408
1409C<no overload> now warns for invalid arguments, just like C<use overload>.
1410
1411=item *
1412
606e21ea 1413L<PerlIO::encoding> has been upgraded to 0.16.
357b01df
AAS
1414
1415This is the module implementing the ":encoding(...)" I/O layer. It no
1416longer corrupts memory or crashes when the encoding back-end reallocates
1417the buffer or gives it a typeglob or shared hash key scalar.
1418
1419=item *
1420
606e21ea 1421L<PerlIO::scalar> has been upgraded to 0.16.
357b01df
AAS
1422
1423The buffer scalar supplied may now only contain code pounts 0xFF or
1424lower. [perl #109828]
1425
1426=item *
1427
606e21ea 1428L<Perl::OSType> has been upgraded to 1.003.
357b01df
AAS
1429
1430This fixes a bug detecting the VOS operating system.
1431
1432=item *
1433
606e21ea 1434L<Pod::Html> has been upgraded to 1.18.
357b01df
AAS
1435
1436The option C<--libpods> has been reinstated. It is deprecated, and its use
1437does nothing other than issue a warning that it is no longer supported.
1438
1439Since the HTML files generated by pod2html claim to have a UTF-8 charset,
1440actually write the files out using UTF-8 [perl #111446].
1441
1442=item *
1443
606e21ea 1444L<Pod::Simple> has been upgraded to 3.28.
357b01df
AAS
1445
1446Numerous improvements have been made, mostly to Pod::Simple::XHTML,
1447which also has a compatibility change: the C<codes_in_verbatim> option
1448is now disabled by default. See F<cpan/Pod-Simple/ChangeLog> for the
1449full details.
1450
1451=item *
1452
606e21ea 1453L<re> has been upgraded to 0.23
357b01df
AAS
1454
1455Single character [class]es like C</[s]/> or C</[s]/i> are now optimized
1456as if they did not have the brackets, i.e. C</s/> or C</s/i>.
1457
1458See note about C<op_comp> in the L</Internal Changes> section below.
1459
1460=item *
1461
606e21ea 1462L<Safe> has been upgraded to 2.35.
357b01df
AAS
1463
1464Fix interactions with C<Devel::Cover>.
1465
1466Don't eval code under C<no strict>.
1467
1468=item *
1469
606e21ea 1470L<Scalar::Util> has been upgraded to version 1.27.
357b01df
AAS
1471
1472Fix an overloading issue with C<sum>.
1473
1474C<first> and C<reduce> now check the callback first (so C<&first(1)> is
1475disallowed).
1476
1477Fix C<tainted> on magical values [rt.cpan.org #55763].
1478
1479Fix C<sum> on previously magical values [rt.cpan.org #61118].
1480
1481Fix reading past the end of a fixed buffer [rt.cpan.org #72700].
1482
1483=item *
1484
606e21ea 1485L<Search::Dict> has been upgraded to 1.07.
357b01df
AAS
1486
1487No longer require C<stat> on filehandles.
1488
1489Use C<fc> for casefolding.
1490
1491=item *
1492
606e21ea 1493L<Socket> has been upgraded to 2.009.
357b01df
AAS
1494
1495Constants and functions required for IP multicast source group membership
1496have been added.
1497
1498C<unpack_sockaddr_in()> and C<unpack_sockaddr_in6()> now return just the IP
1499address in scalar context, and C<inet_ntop()> now guards against incorrect
1500length scalars being passed in.
1501
1502This fixes an uninitialized memory read.
1503
1504=item *
1505
606e21ea 1506L<Storable> has been upgraded to 2.41.
357b01df
AAS
1507
1508Modifying C<$_[0]> within C<STORABLE_freeze> no longer results in crashes
1509[perl #112358].
1510
1511An object whose class implements C<STORABLE_attach> is now thawed only once
1512when there are multiple references to it in the structure being thawed
1513[perl #111918].
1514
1515Restricted hashes were not always thawed correctly [perl #73972].
1516
1517Storable would croak when freezing a blessed REF object with a
1518C<STORABLE_freeze()> method [perl #113880].
1519
1520It can now freeze and thaw vstrings correctly. This causes a slight
1521incompatible change in the storage format, so the format version has
1522increased to 2.9.
1523
1524This contains various bugfixes, including compatibility fixes for older
1525versions of Perl and vstring handling.
1526
1527=item *
1528
606e21ea 1529L<Sys::Syslog> has been upgraded to 0.32.
357b01df
AAS
1530
1531This contains several bug fixes relating to C<getservbyname()>,
1532C<setlogsock()>and log levels in C<syslog()>, together with fixes for
1533Windows, Haiku-OS and GNU/kFreeBSD. See F<cpan/Sys-Syslog/Changes>
1534for the full details.
1535
1536=item *
1537
606e21ea 1538L<Term::ANSIColor> has been upgraded to 4.02.
357b01df
AAS
1539
1540Add support for italics.
1541
1542Improve error handling.
1543
1544=item *
1545
606e21ea 1546L<Term::ReadLine> has been upgraded to 1.10. This fixes the
357b01df
AAS
1547use of the B<cpan> and B<cpanp> shells on Windows in the event that the current
1548drive happens to contain a F<\dev\tty> file.
1549
1550=item *
1551
606e21ea 1552L<Test::Harness> has been upgraded to 3.26.
357b01df
AAS
1553
1554Fix glob semantics on Win32 [rt.cpan.org #49732].
1555
1556Don't use C<Win32::GetShortPathName> when calling perl [rt.cpan.org #47890].
1557
1558Ignore -T when reading shebang [rt.cpan.org #64404].
1559
1560Handle the case where we don't know the wait status of the test more
1561gracefully.
1562
1563Make the test summary 'ok' line overridable so that it can be changed to a
1564plugin to make the output of prove idempotent.
1565
1566Don't run world-writable files.
1567
1568=item *
1569
606e21ea 1570L<Text::Tabs> and L<Text::Wrap> have been upgraded to
357b01df
AAS
15712012.0818. Support for Unicode combining characters has been added to them
1572both.
1573
1574=item *
1575
606e21ea 1576L<threads::shared> has been upgraded to 1.31.
357b01df
AAS
1577
1578This adds the option to warn about or ignore attempts to clone structures
1579that can't be cloned, as opposed to just unconditionally dying in
1580that case.
1581
1582This adds support for dual-valued values as created by
1583L<Scalar::Util::dualvar|Scalar::Util/"dualvar NUM, STRING">.
1584
1585=item *
1586
606e21ea 1587L<Tie::StdHandle> has been upgraded to 4.3.
357b01df
AAS
1588
1589C<READ> now respects the offset argument to C<read> [perl #112826].
1590
1591=item *
1592
606e21ea 1593L<Time::Local> has been upgraded to 1.2300.
357b01df
AAS
1594
1595Seconds values greater than 59 but less than 60 no longer cause
1596C<timegm()> and C<timelocal()> to croak.
1597
1598=item *
1599
606e21ea 1600L<Unicode::UCD> has been upgraded to 0.53.
357b01df
AAS
1601
1602This adds a function L<all_casefolds()|Unicode::UCD/all_casefolds()>
1603that returns all the casefolds.
1604
1605=item *
1606
606e21ea 1607L<Win32> has been upgraded to 0.47.
357b01df
AAS
1608
1609New APIs have been added for getting and setting the current code page.
1610
1611=back
1612
1613
5ed58cbd
RS
1614=head2 Removed Modules and Pragmata
1615
32b79602 1616=over
33392251
BF
1617
1618=item *
1619
5ed58cbd
RS
1620L<Version::Requirements> has been removed from the core distribution. It is
1621available under a different name: L<CPAN::Meta::Requirements>.
2426c394 1622
337fb649 1623=back
2426c394 1624
5ed58cbd 1625=head1 Documentation
19718730 1626
5ed58cbd
RS
1627=head2 Changes to Existing Documentation
1628
1629=head3 L<perlcheat>
82d98f72 1630
5a6a30f4 1631=over 4
b7c7d786 1632
5ed58cbd
RS
1633=item *
1634
1635L<perlcheat> has been reorganized, and a few new sections were added.
1636
1637=back
1638
1639=head3 L<perldata>
1640
1641=over 4
82d98f72 1642
5ed58cbd 1643=item *
d2d1e842 1644
5ed58cbd
RS
1645Now explicitly documents the behaviour of hash initializer lists that
1646contain duplicate keys.
f355e93d 1647
5a6a30f4 1648=back
f355e93d 1649
5ed58cbd 1650=head3 L<perldiag>
19718730 1651
19718730 1652=over 4
e14ac59b 1653
5ed58cbd
RS
1654=item *
1655
1656The explanation of symbolic references being prevented by "strict refs"
1657now doesn't assume that the reader knows what symbolic references are.
1658
1659=back
9f351b45 1660
5ed58cbd 1661=head3 L<perlfaq>
9f351b45 1662
5ed58cbd 1663=over 4
9f351b45 1664
5ed58cbd 1665=item *
7cf3104f 1666
5ed58cbd 1667L<perlfaq> has been synchronized with version 5.0150040 from CPAN.
12719193 1668
6253ee75 1669=back
216cf7fc 1670
5ed58cbd 1671=head3 L<perlfunc>
f5b73711 1672
5ed58cbd
RS
1673=over 4
1674
1675=item *
a75569c0 1676
5ed58cbd 1677The return value of C<pipe> is now documented.
a75569c0 1678
5ed58cbd 1679=item *
a75569c0 1680
5ed58cbd
RS
1681Clarified documentation of C<our>.
1682
1683=back
1684
1685=head3 L<perlop>
1686
1687=over 4
1688
1689=item *
1690
1691Loop control verbs (C<dump>, C<goto>, C<next>, C<last> and C<redo>) have always
1692had the same precedence as assignment operators, but this was not documented
1693until now.
1694
1695=back
1696
1697=head3 Diagnostics
1698
1699The following additions or changes have been made to diagnostic output,
1700including warnings and fatal error messages. For the complete list of
1701diagnostic messages, see L<perldiag>.
1702
5ed58cbd
RS
1703=head2 New Diagnostics
1704
5ed58cbd
RS
1705=head3 New Errors
1706
1707=over 4
1708
1709=item *
1710
1711L<Unterminated delimiter for here document|perldiag/"Unterminated delimiter for here document">
1712
1713This message now occurs when a here document label has an initial quotation
1714mark but the final quotation mark is missing.
1715
1716This replaces a bogus and misleading error message about not finding the label
1717itself [perl #114104].
1718
1719=item *
1720
1721L<panic: child pseudo-process was never scheduled|perldiag/"panic: child pseudo-process was never scheduled">
1722
1723This error is thrown when a child pseudo-process in the ithreads implementation
1724on Windows was not scheduled within the time period allowed and therefore was
1725not able to initialize properly [perl #88840].
1726
1727=item *
1728
1729L<Group name must start with a non-digit word character in regex; marked by <-- HERE in mE<sol>%sE<sol>|perldiag/"Group name must start with a non-digit word character in regex; marked by <-- HERE in m/%s/">
1730
1731This error has been added for C<(?&0)>, which is invalid. It used to
1732produce an incomprehensible error message [perl #101666].
1733
1734=item *
1735
1736L<Can't use an undefined value as a subroutine reference|perldiag/"Can't use an undefined value as %s reference">
1737
1738Calling an undefined value as a subroutine now produces this error message.
1739It used to, but was accidentally disabled, first in Perl 5.004 for
e612b5a0 1740non-magical variables, and then in Perl v5.14 for magical (e.g., tied)
5ed58cbd
RS
1741variables. It has now been restored. In the mean time, undef was treated
1742as an empty string [perl #113576].
1743
1744=item *
1745
1746L<Experimental "%s" subs not enabled|perldiag/"Experimental "%s" subs not enabled">
1747
1748To use lexical subs, you must first enable them:
1749
1750 no warnings 'experimental::lexical_subs';
1751 use feature 'lexical_subs';
1752 my sub foo { ... }
1753
1754=back
1755
1756=head3 New Warnings
1757
1758=over 4
1759
1760=item *
1761
5cd341ec 1762L<'Strings with code points over 0xFF may not be mapped into in-memory file handles'|perldiag/"Strings with code points over 0xFF may not be mapped into in-memory file handles">
5ed58cbd
RS
1763
1764=item *
1765
1766L<'%s' resolved to '\o{%s}%d'|perldiag/"'%s' resolved to '\o{%s}%d'">
1767
1768=item *
1769
1770L<'Trailing white-space in a charnames alias definition is deprecated'|perldiag/"Trailing white-space in a charnames alias definition is deprecated">
1771
1772=item *
1773
1774L<'A sequence of multiple spaces in a charnames alias definition is deprecated'|perldiag/"A sequence of multiple spaces in a charnames alias definition is deprecated">
1775
1776=item *
1777
1778L<'Passing malformed UTF-8 to "%s" is deprecated'|perldiag/"Passing malformed UTF-8 to "%s" is deprecated">
1779
1780=item *
1781
1782L<Subroutine "&%s" is not available|perldiag/"Subroutine "&%s" is not available">
1783
1784(W closure) During compilation, an inner named subroutine or eval is
1785attempting to capture an outer lexical subroutine that is not currently
1786available. This can happen for one of two reasons. First, the lexical
1787subroutine may be declared in an outer anonymous subroutine that has not
1788yet been created. (Remember that named subs are created at compile time,
1789while anonymous subs are created at run-time.) For example,
1790
1791 sub { my sub a {...} sub f { \&a } }
1792
1793At the time that f is created, it can't capture the current the "a" sub,
1794since the anonymous subroutine hasn't been created yet. Conversely, the
1795following won't give a warning since the anonymous subroutine has by now
1796been created and is live:
1797
1798 sub { my sub a {...} eval 'sub f { \&a }' }->();
1799
1800The second situation is caused by an eval accessing a variable that has
1801gone out of scope, for example,
1802
1803 sub f {
1804 my sub a {...}
1805 sub { eval '\&a' }
1806 }
1807 f()->();
1808
1809Here, when the '\&a' in the eval is being compiled, f() is not currently
1810being executed, so its &a is not available for capture.
1811
1812=item *
1813
1814L<"%s" subroutine &%s masks earlier declaration in same %s|perldiag/"%s" subroutine &%s masks earlier declaration in same %s>
1815
1816(W misc) A "my" or "state" subroutine has been redeclared in the
1817current scope or statement, effectively eliminating all access to
1818the previous instance. This is almost always a typographical error.
1819Note that the earlier subroutine will still exist until the end of
1820the scope or until all closure references to it are destroyed.
1821
1822=item *
1823
1824L<The %s feature is experimental|perldiag/"The %s feature is experimental">
1825
1826(S experimental) This warning is emitted if you enable an experimental
1827feature via C<use feature>. Simply suppress the warning if you want
1828to use the feature, but know that in doing so you are taking the risk
1829of using an experimental feature which may change or be removed in a
1830future Perl version:
1831
1832 no warnings "experimental::lexical_subs";
1833 use feature "lexical_subs";
1834
1835=item *
1836
1837L<sleep(%u) too large|perldiag/"sleep(%u) too large">
1838
1839(W overflow) You called C<sleep> with a number that was larger than it can
1840reliably handle and C<sleep> probably slept for less time than requested.
1841
1842=item *
1843
1844L<Wide character in setenv|perldiag/"Wide character in %s">
1845
1846Attempts to put wide characters into environment variables via C<%ENV> now
1847provoke this warning.
1848
1849=item *
1850
1851"L<Invalid negative number (%s) in chr|perldiag/"Invalid negative number (%s) in chr">"
1852
1853C<chr()> now warns when passed a negative value [perl #83048].
1854
1855=item *
1856
1857"L<Integer overflow in srand|perldiag/"Integer overflow in srand">"
1858
1859C<srand()> now warns when passed a value that doesn't fit in a C<UV> (since the
1860value will be truncated rather than overflowing) [perl #40605].
1861
1862=item *
1863
1864"L<-i used with no filenames on the command line, reading from STDIN|perldiag/"-i used with no filenames on the command line, reading from STDIN">"
1865
1866Running perl with the C<-i> flag now warns if no input files are provided on
1867the command line [perl #113410].
1868
1869=back
1870
1871=head2 Changes to Existing Diagnostics
1872
1873=over 4
1874
1875=item *
1876
1877L<$* is no longer supported|perldiag/"$* is no longer supported">
1878
1879The warning that use of C<$*> and C<$#> is no longer supported is now
1880generated for every location that references them. Previously it would fail
1881to be generated if another variable using the same typeglob was seen first
1882(e.g. C<@*> before C<$*>), and would not be generated for the second and
1883subsequent uses. (It's hard to fix the failure to generate warnings at all
1884without also generating them every time, and warning every time is
1885consistent with the warnings that C<$[> used to generate.)
1886
1887=item *
1888
1889The warnings for C<\b{> and C<\B{> were added. They are a deprecation
1890warning which should be turned off by that category. One should not
1891have to turn off regular regexp warnings as well to get rid of these.
1892
1893=item *
1894
1895L<Constant(%s): Call to &{$^H{%s}} did not return a defined value|perldiag/Constant(%s): Call to &{$^H{%s}} did not return a defined value>
1896
1897Constant overloading that returns C<undef> results in this error message.
1898For numeric constants, it used to say "Constant(undef)". "undef" has been
1899replaced with the number itself.
1900
1901=item *
1902
1903The error produced when a module cannot be loaded now includes a hint that
1904the module may need to be installed: "Can't locate hopping.pm in @INC (you
1905may need to install the hopping module) (@INC contains: ...)"
1906
1907=item *
1908
1909L<vector argument not supported with alpha versions|perldiag/vector argument not supported with alpha versions>
1910
1911This warning was not suppressable, even with C<no warnings>. Now it is
1912suppressible, and has been moved from the "internal" category to the
1913"printf" category.
1914
1915=item *
1916
1917C<< Can't do {n,m} with n > m in regex; marked by <-- HERE in m/%s/ >>
1918
1919This fatal error has been turned into a warning that reads:
1920
1921L<< Quantifier {n,m} with n > m can't match in regex | perldiag/Quantifier {n,m} with n > m can't match in regex >>
1922
1923(W regexp) Minima should be less than or equal to maxima. If you really want
1924your regexp to match something 0 times, just put {0}.
1925
1926=item *
1927
1928The "Runaway prototype" warning that occurs in bizarre cases has been
1929removed as being unhelpful and inconsistent.
1930
1931=item *
1932
1933The "Not a format reference" error has been removed, as the only case in
1934which it could be triggered was a bug.
1935
1936=item *
1937
1938The "Unable to create sub named %s" error has been removed for the same
1939reason.
1940
1941=item *
1942
1943The 'Can't use "my %s" in sort comparison' error has been downgraded to a
1944warning, '"my %s" used in sort comparison' (with 'state' instead of 'my'
1945for state variables). In addition, the heuristics for guessing whether
1946lexical $a or $b has been misused have been improved to generate fewer
1947false positives. Lexical $a and $b are no longer disallowed if they are
1948outside the sort block. Also, a named unary or list operator inside the
1949sort block no longer causes the $a or $b to be ignored [perl #86136].
1950
1951=back
1952
1953=head1 Utility Changes
1954
1955=head3 L<h2xs>
1956
1957=over 4
1958
1959=item *
1960
1961F<h2xs> no longer produces invalid code for empty defines. [perl #20636]
1962
1963=back
1964
1965=head1 Configuration and Compilation
1966
1967=over 4
1968
1969=item *
1970
1971Added C<useversionedarchname> option to Configure
1972
1973When set, it includes 'api_versionstring' in 'archname'. E.g.
1974x86_64-linux-5.13.6-thread-multi. It is unset by default.
1975
1976This feature was requested by Tim Bunce, who observed that
f105b7be 1977C<INSTALL_BASE> creates a library structure that does not
5ed58cbd
RS
1978differentiate by perl version. Instead, it places architecture
1979specific files in "$install_base/lib/perl5/$archname". This makes
f105b7be 1980it difficult to use a common C<INSTALL_BASE> library path with
5ed58cbd
RS
1981multiple versions of perl.
1982
f105b7be 1983By setting C<-Duseversionedarchname>, the $archname will be
c2959982 1984distinct for architecture I<and> API version, allowing mixed use of
f105b7be 1985C<INSTALL_BASE>.
5ed58cbd
RS
1986
1987=item *
1988
ff772877
RS
1989Add a C<PERL_NO_INLINE_FUNCTIONS> option
1990
f105b7be 1991If C<PERL_NO_INLINE_FUNCTIONS> is defined, don't include "inline.h"
ff772877
RS
1992
1993This permits test code to include the perl headers for definitions without
1994creating a link dependency on the perl library (which may not exist yet).
1995
1996=item *
1997
5ed58cbd
RS
1998Configure will honour the external C<MAILDOMAIN> environment variable, if set.
1999
2000=item *
2001
2002C<installman> no longer ignores the silent option
2003
2004=item *
2005
2006Both C<META.yml> and C<META.json> files are now included in the distribution.
2007
2008=item *
2009
2010F<Configure> will now correctly detect C<isblank()> when compiling with a C++
2011compiler.
2012
2013=item *
2014
2015The pager detection in F<Configure> has been improved to allow responses which
2016specify options after the program name, e.g. B</usr/bin/less -R>, if the user
2017accepts the default value. This helps B<perldoc> when handling ANSI escapes
2018[perl #72156].
2019
2020=back
2021
2022=head1 Testing
2023
2024=over 4
2025
2026=item *
2027
2028The test suite now has a section for tests that require very large amounts
2029of memory. These tests won't run by default; they can be enabled by
2030setting the C<PERL_TEST_MEMORY> environment variable to the number of
2031gibibytes of memory that may be safely used.
2032
2033=back
2034
2035=head1 Platform Support
2036
2037=head2 Discontinued Platforms
2038
2039=over 4
2040
2041=item BeOS
2042
2043BeOS was an operating system for personal computers developed by Be Inc,
2044initially for their BeBox hardware. The OS Haiku was written as an open
2045source replacement for/continuation of BeOS, and its perl port is current and
2046actively maintained.
2047
2048=item UTS Global
2049
2050Support code relating to UTS global has been removed. UTS was a mainframe
2051version of System V created by Amdahl, subsequently sold to UTS Global. The
e612b5a0 2052port has not been touched since before Perl v5.8.0, and UTS Global is now
5ed58cbd
RS
2053defunct.
2054
2055=item VM/ESA
2056
2057Support for VM/ESA has been removed. The port was tested on 2.3.0, which
2058IBM ended service on in March 2002. 2.4.0 ended service in June 2003, and
2059was superseded by Z/VM. The current version of Z/VM is V6.2.0, and scheduled
2060for end of service on 2015/04/30.
2061
2062=item MPE/IX
2063
2064Support for MPE/IX has been removed.
2065
2066=item EPOC
2067
2068Support code relating to EPOC has been removed. EPOC was a family of
2069operating systems developed by Psion for mobile devices. It was the
2070predecessor of Symbian. The port was last updated in April 2002.
2071
2072=item Rhapsody
2073
2074Support for Rhapsody has been removed.
2075
2076=back
2077
2078=head2 Platform-Specific Notes
2079
2080=head3 AIX
2081
2082Configure now always adds C<-qlanglvl=extc99> to the CC flags on AIX when
2083using xlC. This will make it easier to compile a number of XS-based modules
2084that assume C99 [perl #113778].
2085
2086=head3 clang++
2087
2088There is now a workaround for a compiler bug that prevented compiling
e612b5a0 2089with clang++ since Perl v5.15.7 [perl #112786].
5ed58cbd
RS
2090
2091=head3 C++
2092
2093When compiling the Perl core as C++ (which is only semi-supported), the
2094mathom functions are now compiled as C<extern "C">, to ensure proper
2095binary compatibility. (However, binary compatibility isn't generally
2096guaranteed anyway in the situations where this would matter.)
2097
2098=head3 Darwin
2099
2100Stop hardcoding an alignment on 8 byte boundaries to fix builds using
2101-Dusemorebits.
2102
2103=head3 Haiku
2104
2105Perl should now work out of the box on Haiku R1 Alpha 4.
2106
2107=head3 MidnightBSD
2108
2109C<libc_r> was removed from recent versions of MidnightBSD and older versions
2110work better with C<pthread>. Threading is now enabled using C<pthread> which
2111corrects build errors with threading enabled on 0.4-CURRENT.
2112
2113=head3 Solaris
2114
2115In Configure, avoid running sed commands with flags not supported on Solaris.
2116
2117=head3 VMS
2118
2119=over
2120
2121=item *
2122
2123Where possible, the case of filenames and command-line arguments is now
2124preserved by enabling the CRTL features C<DECC$EFS_CASE_PRESERVE> and
2125C<DECC$ARGV_PARSE_STYLE> at start-up time. The latter only takes effect
2126when extended parse is enabled in the process from which Perl is run.
2127
2128=item *
2129
2130The character set for Extended Filename Syntax (EFS) is now enabled by default
2131on VMS. Among other things, this provides better handling of dots in directory
05f5908f 2132names, multiple dots in filenames, and spaces in filenames. To obtain the old
5ed58cbd
RS
2133behavior, set the logical name C<DECC$EFS_CHARSET> to C<DISABLE>.
2134
2135=item *
2136
05f5908f 2137Fixed linking on builds configured with C<-Dusemymalloc=y>.
5ed58cbd
RS
2138
2139=item *
2140
05f5908f
CB
2141Experimental support for building Perl with the HP C++ compiler is available
2142by configuring with C<-Dusecxx>.
5ed58cbd
RS
2143
2144=item *
2145
2146All C header files from the top-level directory of the distribution are now
2147installed on VMS, providing consistency with a long-standing practice on other
2148platforms. Previously only a subset were installed, which broke non-core
2149extension builds for extensions that depended on the missing include files.
2150
2151=item *
2152
2153Quotes are now removed from the command verb (but not the parameters) for
2154commands spawned via C<system>, backticks, or a piped C<open>. Previously,
2155quotes on the verb were passed through to DCL, which would fail to recognize
2156the command. Also, if the verb is actually a path to an image or command
2157procedure on an ODS-5 volume, quoting it now allows the path to contain spaces.
2158
2159=item *
2160
2161The B<a2p> build has been fixed for the HP C++ compiler on OpenVMS.
2162
2163=back
2164
2165=head3 Win32
2166
2167=over
2168
2169=item *
2170
2171Perl can now be built using Microsoft's Visual C++ 2012 compiler by specifying
2172CCTYPE=MSVC110 (or MSVC110FREE if you are using the free Express edition for
2173Windows Desktop) in F<win32/Makefile>.
2174
2175=item *
2176
f105b7be 2177The option to build without C<USE_SOCKETS_AS_HANDLES> has been removed.
5ed58cbd
RS
2178
2179=item *
2180
2181Fixed a problem where perl could crash while cleaning up threads (including the
2182main thread) in threaded debugging builds on Win32 and possibly other platforms
2183[perl #114496].
2184
2185=item *
2186
2187A rare race condition that would lead to L<sleep|perlfunc/sleep> taking more
2188time than requested, and possibly even hanging, has been fixed [perl #33096].
2189
2190=item *
2191
2192C<link> on Win32 now attempts to set C<$!> to more appropriate values
2193based on the Win32 API error code. [perl #112272]
2194
2195Perl no longer mangles the environment block, e.g. when launching a new
2196sub-process, when the environment contains non-ASCII characters. Known
2197problems still remain, however, when the environment contains characters
2198outside of the current ANSI codepage (e.g. see the item about Unicode in
2199C<%ENV> in L<http://perl5.git.perl.org/perl.git/blob/HEAD:/Porting/todo.pod>).
2200[perl #113536]
2201
2202=item *
2203
2204Building perl with some Windows compilers used to fail due to a problem
2205with miniperl's C<glob> operator (which uses the C<perlglob> program)
2206deleting the PATH environment variable [perl #113798].
2207
2208=item *
2209
f105b7be 2210A new makefile option, C<USE_64_BIT_INT>, has been added to the Windows
5ed58cbd
RS
2211makefiles. Set this to "define" when building a 32-bit perl if you want
2212it to use 64-bit integers.
2213
2214Machine code size reductions, already made to the DLLs of XS modules in
e612b5a0 2215Perl v5.17.2, have now been extended to the perl DLL itself.
5ed58cbd 2216
e612b5a0 2217Building with VC++ 6.0 was inadvertently broken in Perl v5.17.2 but has
5ed58cbd
RS
2218now been fixed again.
2219
2220=back
2221
2222=head3 WinCE
2223
2224Building on WinCE is now possible once again, although more work is required
2225to fully restore a clean build.
2226
2227=head1 Internal Changes
2228
2229=over
2230
2231=item *
2232
4263dd11 2233Synonyms for the misleadingly named C<av_len()> have been created:
5ed58cbd
RS
2234C<av_top_index()> and C<av_tindex>. All three of these return the
2235number of the highest index in the array, not the number of elements it
2236contains.
2237
2238=item *
2239
2240SvUPGRADE() is no longer an expression. Originally this macro (and its
2241underlying function, sv_upgrade()) were documented as boolean, although
2242in reality they always croaked on error and never returned false. In 2005
2243the documentation was updated to specify a void return value, but
2244SvUPGRADE() was left always returning 1 for backwards compatibility. This
2245has now been removed, and SvUPGRADE() is now a statement with no return
2246value.
2247
2248So this is now a syntax error:
2249
2250 if (!SvUPGRADE(sv)) { croak(...); }
2251
2252If you have code like that, simply replace it with
2253
2254 SvUPGRADE(sv);
2255
237b9c6d 2256or to avoid compiler warnings with older perls, possibly
5ed58cbd
RS
2257
2258 (void)SvUPGRADE(sv);
2259
2260=item *
2261
2262Perl has a new copy-on-write mechanism that allows any SvPOK scalar to be
2263upgraded to a copy-on-write scalar. A reference count on the string buffer
d16360cf
RS
2264is stored in the string buffer itself. This feature is B<not enabled by
2265default>.
5ed58cbd 2266
d16360cf
RS
2267It can be enabled in a perl build by running F<Configure> with
2268B<-Accflags=-DPERL_NEW_COPY_ON_WRITE>, and we would encourage XS authors
2269to try their code with such an enabled perl, and provide feedback.
2270Unfortunately, there is not yet a good guide to updating XS code to cope
2271with COW. Until such a document is available, consult the perl5-porters
2272mailing list.
5ed58cbd 2273
d16360cf
RS
2274It breaks a few XS modules by allowing copy-on-write scalars to go
2275through code paths that never encountered them before.
5ed58cbd
RS
2276
2277=item *
2278
2279Copy-on-write no longer uses the SvFAKE and SvREADONLY flags. Hence,
2280SvREADONLY indicates a true read-only SV.
2281
2282Use the SvIsCOW macro (as before) to identify a copy-on-write scalar.
2283
2284=item *
2285
f105b7be 2286C<PL_glob_index> is gone.
5ed58cbd
RS
2287
2288=item *
2289
2290The private Perl_croak_no_modify has had its context parameter removed. It is
2291now has a void prototype. Users of the public API croak_no_modify remain
2292unaffected.
2293
2294=item *
2295
2296Copy-on-write (shared hash key) scalars are no longer marked read-only.
2297C<SvREADONLY> returns false on such an SV, but C<SvIsCOW> still returns
2298true.
2299
2300=item *
2301
2302A new op type, C<OP_PADRANGE> has been introduced. The perl peephole
2303optimiser will, where possible, substitute a single padrange op for a
2304pushmark followed by one or more pad ops, and possibly also skipping list
2305and nextstate ops. In addition, the op can carry out the tasks associated
f105b7be 2306with the RHS of a C<< my(...) = @_ >> assignment, so those ops may be optimised
5ed58cbd
RS
2307away too.
2308
2309=item *
2310
2311Case-insensitive matching inside a [bracketed] character class with a
2312multi-character fold no longer excludes one of the possibilities in the
2313circumstances that it used to. [perl #89774].
2314
2315=item *
2316
2317C<PL_formfeed> has been removed.
2318
2319=item *
2320
2321The regular expression engine no longer reads one byte past the end of the
2322target string. While for all internally well-formed scalars this should
2323never have been a problem, this change facilitates clever tricks with
2324string buffers in CPAN modules. [perl #73542]
2325
2326=item *
2327
2328Inside a BEGIN block, C<PL_compcv> now points to the currently-compiling
2329subroutine, rather than the BEGIN block itself.
2330
2331=item *
2332
2333C<mg_length> has been deprecated.
2334
2335=item *
2336
2337C<sv_len> now always returns a byte count and C<sv_len_utf8> a character
2338count. Previously, C<sv_len> and C<sv_len_utf8> were both buggy and would
2339sometimes returns bytes and sometimes characters. C<sv_len_utf8> no longer
be12dd22 2340assumes that its argument is in UTF-8. Neither of these creates UTF-8 caches
5ed58cbd
RS
2341for tied or overloaded values or for non-PVs any more.
2342
2343=item *
2344
2345C<sv_mortalcopy> now copies string buffers of shared hash key scalars when
2346called from XS modules [perl #79824].
2347
2348=item *
2349
2350C<RXf_SPLIT> and C<RXf_SKIPWHITE> are no longer used. They are now
2351#defined as 0.
2352
2353=item *
2354
2355The new C<RXf_MODIFIES_VARS> flag can be set by custom regular expression
2356engines to indicate that the execution of the regular expression may cause
2357variables to be modified. This lets C<s///> know to skip certain
2358optimisations. Perl's own regular expression engine sets this flag for the
2359special backtracking verbs that set $REGMARK and $REGERROR.
2360
2361=item *
2362
2363The APIs for accessing lexical pads have changed considerably.
2364
2365C<PADLIST>s are now longer C<AV>s, but their own type instead.
2366C<PADLIST>s now contain a C<PAD> and a C<PADNAMELIST> of C<PADNAME>s,
2367rather than C<AV>s for the pad and the list of pad names. C<PAD>s,
2368C<PADNAMELIST>s, and C<PADNAME>s are to be accessed as such through the
2369newly added pad API instead of the plain C<AV> and C<SV> APIs. See
2370L<perlapi> for details.
2371
2372=item *
2373
2374In the regex API, the numbered capture callbacks are passed an index
2375indicating what match variable is being accessed. There are special
2376index values for the C<$`, $&, $&> variables. Previously the same three
2377values were used to retrieve C<${^PREMATCH}, ${^MATCH}, ${^POSTMATCH}>
2378too, but these have now been assigned three separate values. See
2379L<perlreapi/Numbered capture callbacks>.
2380
2381=item *
2382
2383C<PL_sawampersand> was previously a boolean indicating that any of
2384C<$`, $&, $&> had been seen; it now contains three one-bit flags
2385indicating the presence of each of the variables individually.
2386
2387=item *
2388
2389The C<CV *> typemap entry now supports C<&{}> overloading and typeglobs,
2390just like C<&{...}> [perl #96872].
2391
2392=item *
2393
2394The C<SVf_AMAGIC> flag to indicate overloading is now on the stash, not the
2395object. It is now set automatically whenever a method or @ISA changes, so
2396its meaning has changed, too. It now means "potentially overloaded". When
2397the overload table is calculated, the flag is automatically turned off if
2398there is no overloading, so there should be no noticeable slowdown.
2399
2400The staleness of the overload tables is now checked when overload methods
2401are invoked, rather than during C<bless>.
2402
2403"A" magic is gone. The changes to the handling of the C<SVf_AMAGIC> flag
2404eliminate the need for it.
2405
2406C<PL_amagic_generation> has been removed as no longer necessary. For XS
2407modules, it is now a macro alias to C<PL_na>.
2408
2409The fallback overload setting is now stored in a stash entry separate from
2410overloadedness itself.
2411
2412=item *
2413
2414The character-processing code has been cleaned up in places. The changes
2415should be operationally invisible.
2416
2417=item *
2418
e612b5a0 2419The C<study> function was made a no-op in v5.16. It was simply disabled via
5ed58cbd
RS
2420a C<return> statement; the code was left in place. Now the code supporting
2421what C<study> used to do has been removed.
2422
2423=item *
2424
2425Under threaded perls, there is no longer a separate PV allocated for every
2426COP to store its package name (C<< cop->stashpv >>). Instead, there is an
2427offset (C<< cop->stashoff >>) into the new C<PL_stashpad> array, which
2428holds stash pointers.
2429
2430=item *
2431
2432In the pluggable regex API, the C<regexp_engine> struct has acquired a new
2433field C<op_comp>, which is currently just for perl's internal use, and
f105b7be 2434should be initialized to NULL by other regex plugin modules.
5ed58cbd
RS
2435
2436=item *
2437
7779650e 2438A new function C<alloccopstash> has been added to the API, but is considered
5ed58cbd
RS
2439experimental. See L<perlapi>.
2440
2441=item *
2442
2443Perl used to implement get magic in a way that would sometimes hide bugs in
4263dd11 2444code that could call mg_get() too many times on magical values. This hiding of
5ed58cbd
RS
2445errors no longer occurs, so long-standing bugs may become visible now. If
2446you see magic-related errors in XS code, check to make sure it, together
2447with the Perl API functions it uses, calls mg_get() only once on SvGMAGICAL()
2448values.
2449
2450=item *
2451
2452OP allocation for CVs now uses a slab allocator. This simplifies
2453memory management for OPs allocated to a CV, so cleaning up after a
2454compilation error is simpler and safer [perl #111462][perl #112312].
2455
2456=item *
2457
f105b7be 2458C<PERL_DEBUG_READONLY_OPS> has been rewritten to work with the new slab
5ed58cbd
RS
2459allocator, allowing it to catch more violations than before.
2460
2461=item *
2462
f105b7be
KE
2463The old slab allocator for ops, which was only enabled for C<PERL_IMPLICIT_SYS>
2464and C<PERL_DEBUG_READONLY_OPS>, has been retired.
5ed58cbd
RS
2465
2466=back
2467
2468=head1 Selected Bug Fixes
2469
2470=over 4
2471
2472=item *
2473
ef4b82dd 2474Here document terminators no longer require a terminating newline character when
5ed58cbd
RS
2475they occur at the end of a file. This was already the case at the end of a
2476string eval [perl #65838].
2477
2478=item *
2479
f105b7be 2480C<-DPERL_GLOBAL_STRUCT> builds now free the global struct B<after>
5ed58cbd
RS
2481they've finished using it.
2482
2483=item *
2484
2485A trailing '/' on a path in @INC will no longer have an additional '/'
2486appended.
2487
2488=item *
2489
2490The C<:crlf> layer now works when unread data doesn't fit into its own
2491buffer. [perl #112244].
2492
2493=item *
2494
2495C<ungetc()> now handles UTF-8 encoded data. [perl #116322].
2496
2497=item *
2498
2499A bug in the core typemap caused any C types that map to the T_BOOL core
2500typemap entry to not be set, updated, or modified when the T_BOOL variable was
2501used in an OUTPUT: section with an exception for RETVAL. T_BOOL in an INPUT:
2502section was not affected. Using a T_BOOL return type for an XSUB (RETVAL)
2503was not affected. A side effect of fixing this bug is, if a T_BOOL is specified
2504in the OUTPUT: section (which previous did nothing to the SV), and a read only
2505SV (literal) is passed to the XSUB, croaks like "Modification of a read-only
2506value attempted" will happen. [perl #115796]
2507
2508=item *
2509
2510On many platforms, providing a directory name as the script name caused perl
2511to do nothing and report success. It should now universally report an error
2512and exit nonzero. [perl #61362]
2513
2514=item *
2515
2516C<sort {undef} ...> under fatal warnings no longer crashes. It had
e612b5a0 2517begun crashing in Perl v5.16.
5ed58cbd
RS
2518
2519=item *
2520
2521Stashes blessed into each other
2522(C<bless \%Foo::, 'Bar'; bless \%Bar::, 'Foo'>) no longer result in double
e612b5a0 2523frees. This bug started happening in Perl v5.16.
5ed58cbd
RS
2524
2525=item *
2526
2527Numerous memory leaks have been fixed, mostly involving fatal warnings and
2528syntax errors.
2529
2530=item *
2531
2532Some failed regular expression matches such as C<'f' =~ /../g> were not
2533resetting C<pos>. Also, "match-once" patterns (C<m?...?g>) failed to reset
2534it, too, when invoked a second time [perl #23180].
2535
2536=item *
2537
5ed58cbd
RS
2538Several bugs involving C<local *ISA> and C<local *Foo::> causing stale
2539MRO caches have been fixed.
2540
2541=item *
2542
2543Defining a subroutine when its typeglob has been aliased no longer results
e612b5a0 2544in stale method caches. This bug was introduced in Perl v5.10.
5ed58cbd
RS
2545
2546=item *
2547
2548Localising a typeglob containing a subroutine when the typeglob's package
2549has been deleted from its parent stash no longer produces an error. This
e612b5a0 2550bug was introduced in Perl v5.14.
5ed58cbd
RS
2551
2552=item *
2553
2554Under some circumstances, C<local *method=...> would fail to reset method
2555caches upon scope exit.
2556
2557=item *
2558
2559C</[.foo.]/> is no longer an error, but produces a warning (as before) and
2560is treated as C</[.fo]/> [perl #115818].
2561
2562=item *
2563
2564C<goto $tied_var> now calls FETCH before deciding what type of goto
2565(subroutine or label) this is.
2566
2567=item *
2568
2569Renaming packages through glob assignment
2570(C<*Foo:: = *Bar::; *Bar:: = *Baz::>) in combination with C<m?...?> and
2571C<reset> no longer makes threaded builds crash.
2572
2573=item *
2574
2575A number of bugs related to assigning a list to hash have been fixed. Many of
2576these involve lists with repeated keys like C<(1, 1, 1, 1)>.
2577
2578=over 4
2579
2580=item *
2581
2582The expression C<scalar(%h = (1, 1, 1, 1))> now returns C<4>, not C<2>.
2583
2584=item *
2585
2586The return value of C<%h = (1, 1, 1)> in list context was wrong. Previously
2587this would return C<(1, undef, 1)>, now it returns C<(1, undef)>.
2588
2589=item *
2590
2591Perl now issues the same warning on C<($s, %h) = (1, {})> as it does for
2592C<(%h) = ({})>, "Reference found where even-sized list expected".
2593
2594=item *
2595
2596A number of additional edge cases in list assignment to hashes were
2597corrected. For more details see commit 23b7025ebc.
2598
2599=back
2600
2601=item *
2602
2603Attributes applied to lexical variables no longer leak memory.
2604[perl #114764]
2605
2606=item *
2607
2608C<dump>, C<goto>, C<last>, C<next>, C<redo> or C<require> followed by a
2609bareword (or version) and then an infix operator is no longer a syntax
2610error. It used to be for those infix operators (like C<+>) that have a
2611different meaning where a term is expected. [perl #105924]
2612
2613=item *
2614
2615C<require a::b . 1> and C<require a::b + 1> no longer produce erroneous
2616ambiguity warnings. [perl #107002]
2617
2618=item *
2619
2620Class method calls are now allowed on any string, and not just strings
2621beginning with an alphanumeric character. [perl #105922]
2622
2623=item *
2624
2625An empty pattern created with C<qr//> used in C<m///> no longer triggers
2626the "empty pattern reuses last pattern" behaviour. [perl #96230]
2627
2628=item *
2629
2630Tying a hash during iteration no longer results in a memory leak.
2631
2632=item *
2633
2634Freeing a tied hash during iteration no longer results in a memory leak.
2635
2636=item *
2637
2638List assignment to a tied array or hash that dies on STORE no longer
2639results in a memory leak.
2640
2641=item *
2642
2643If the hint hash (C<%^H>) is tied, compile-time scope entry (which copies
2644the hint hash) no longer leaks memory if FETCH dies. [perl #107000]
2645
2646=item *
2647
2648Constant folding no longer inappropriately triggers the special
2649C<split " "> behaviour. [perl #94490]
2650
2651=item *
2652
2653C<defined scalar(@array)>, C<defined do { &foo }>, and similar constructs
2654now treat the argument to C<defined> as a simple scalar. [perl #97466]
2655
2656=item *
2657
2658Running a custom debugging that defines no C<*DB::DB> glob or provides a
2659subroutine stub for C<&DB::DB> no longer results in a crash, but an error
2660instead. [perl #114990]
2661
2662=item *
2663
2664C<reset ""> now matches its documentation. C<reset> only resets C<m?...?>
2665patterns when called with no argument. An empty string for an argument now
2666does nothing. (It used to be treated as no argument.) [perl #97958]
2667
2668=item *
2669
2670C<printf> with an argument returning an empty list no longer reads past the
2671end of the stack, resulting in erratic behaviour. [perl #77094]
2672
2673=item *
2674
2675C<--subname> no longer produces erroneous ambiguity warnings.
2676[perl #77240]
2677
2678=item *
2679
2680C<v10> is now allowed as a label or package name. This was inadvertently
e612b5a0 2681broken when v-strings were added in Perl v5.6. [perl #56880]
5ed58cbd
RS
2682
2683=item *
2684
2685C<length>, C<pos>, C<substr> and C<sprintf> could be confused by ties,
2686overloading, references and typeglobs if the stringification of such
be12dd22 2687changed the internal representation to or from UTF-8. [perl #114410]
5ed58cbd
RS
2688
2689=item *
2690
2691utf8::encode now calls FETCH and STORE on tied variables. utf8::decode now
2692calls STORE (it was already calling FETCH).
2693
2694=item *
2695
2696C<$tied =~ s/$non_utf8/$utf8/> no longer loops infinitely if the tied
2697variable returns a Latin-1 string, shared hash key scalar, or reference or
2ae351f8 2698typeglob that stringifies as ASCII or Latin-1. This was a regression from
e612b5a0 2699v5.12.
5ed58cbd
RS
2700
2701=item *
2702
2703C<s///> without /e is now better at detecting when it needs to forego
2704certain optimisations, fixing some buggy cases:
2705
2706=over
2707
2708=item *
2709
2710Match variables in certain constructs (C<&&>, C<||>, C<..> and others) in
2711the replacement part; e.g., C<s/(.)/$l{$a||$1}/g>. [perl #26986]
2712
2713=item *
2714
2715Aliases to match variables in the replacement.
2716
2717=item *
2718
2719C<$REGERROR> or C<$REGMARK> in the replacement. [perl #49190]
2720
2721=item *
2722
2723An empty pattern (C<s//$foo/>) that causes the last-successful pattern to
2724be used, when that pattern contains code blocks that modify the variables
2725in the replacement.
2726
2727=back
2728
2729=item *
2730
2731The taintedness of the replacement string no longer affects the taintedness
2732of the return value of C<s///e>.
2733
2734=item *
2735
2736The C<$|> autoflush variable is created on-the-fly when needed. If this
2737happened (e.g., if it was mentioned in a module or eval) when the
2738currently-selected filehandle was a typeglob with an empty IO slot, it used
2739to crash. [perl #115206]
2740
2741=item *
2742
2743Line numbers at the end of a string eval are no longer off by one.
2744[perl #114658]
2745
2746=item *
2747
2748@INC filters (subroutines returned by subroutines in @INC) that set $_ to a
2749copy-on-write scalar no longer cause the parser to modify that string
2750buffer in place.
2751
2752=item *
2753
2754C<length($object)> no longer returns the undefined value if the object has
2755string overloading that returns undef. [perl #115260]
2756
2757=item *
2758
2759The use of C<PL_stashcache>, the stash name lookup cache for method calls, has
2760been restored,
2761
2762Commit da6b625f78f5f133 in August 2011 inadvertently broke the code that looks
2763up values in C<PL_stashcache>. As it's a only cache, quite correctly everything
2764carried on working without it.
2765
2766=item *
2767
e612b5a0 2768The error "Can't localize through a reference" had disappeared in v5.16.0
5ed58cbd 2769when C<local %$ref> appeared on the last line of an lvalue subroutine.
e612b5a0 2770This error disappeared for C<\local %$ref> in perl v5.8.1. It has now
5ed58cbd
RS
2771been restored.
2772
2773=item *
2774
2775The parsing of here-docs has been improved significantly, fixing several
2776parsing bugs and crashes and one memory leak, and correcting wrong
2777subsequent line numbers under certain conditions.
2778
2779=item *
2780
2781Inside an eval, the error message for an unterminated here-doc no longer
2782has a newline in the middle of it [perl #70836].
2783
2784=item *
2785
2786A substitution inside a substitution pattern (C<s/${s|||}//>) no longer
2787confuses the parser.
2788
2789=item *
2790
2791It may be an odd place to allow comments, but C<s//"" # hello/e> has
2792always worked, I<unless> there happens to be a null character before the
2793first #. Now it works even in the presence of nulls.
2794
2795=item *
2796
2797An invalid range in C<tr///> or C<y///> no longer results in a memory leak.
2798
2799=item *
2800
2801String eval no longer treats a semicolon-delimited quote-like operator at
2802the very end (C<eval 'q;;'>) as a syntax error.
2803
2804=item *
2805
2806C<< warn {$_ => 1} + 1 >> is no longer a syntax error. The parser used to
2807get confused with certain list operators followed by an anonymous hash and
2808then an infix operator that shares its form with a unary operator.
2809
2810=item *
2811
2812C<(caller $n)[6]> (which gives the text of the eval) used to return the
2813actual parser buffer. Modifying it could result in crashes. Now it always
2814returns a copy. The string returned no longer has "\n;" tacked on to the
2815end. The returned text also includes here-doc bodies, which used to be
2816omitted.
2817
2818=item *
2819
5cd341ec
RS
2820The UTF-8 position cache is now reset when accessing magical variables, to
2821avoid the string buffer and the UTF-8 position cache getting out of sync
5ed58cbd
RS
2822[perl #114410].
2823
2824=item *
2825
be12dd22
RS
2826Various cases of get magic being called twice for magical UTF-8
2827strings have been fixed.
5ed58cbd
RS
2828
2829=item *
2830
2831This code (when not in the presence of C<$&> etc)
2832
2833 $_ = 'x' x 1_000_000;
2834 1 while /(.)/;
2835
2836used to skip the buffer copy for performance reasons, but suffered from C<$1>
2837etc changing if the original string changed. That's now been fixed.
2838
2839=item *
2840
2841Perl doesn't use PerlIO anymore to report out of memory messages, as PerlIO
2842might attempt to allocate more memory.
2843
2844=item *
2845
2846In a regular expression, if something is quantified with C<{n,m}> where
2847C<S<n E<gt> m>>, it can't possibly match. Previously this was a fatal
2848error, but now is merely a warning (and that something won't match).
2849[perl #82954].
2850
2851=item *
2852
2853It used to be possible for formats defined in subroutines that have
2854subsequently been undefined and redefined to close over variables in the
2855wrong pad (the newly-defined enclosing sub), resulting in crashes or
2856"Bizarre copy" errors.
2857
2858=item *
2859
2860Redefinition of XSUBs at run time could produce warnings with the wrong
2861line number.
2862
2863=item *
2864
2865The %vd sprintf format does not support version objects for alpha versions.
2866It used to output the format itself (%vd) when passed an alpha version, and
2867also emit an "Invalid conversion in printf" warning. It no longer does,
2868but produces the empty string in the output. It also no longer leaks
2869memory in this case.
2870
2871=item *
2872
2873C<< $obj->SUPER::method >> calls in the main package could fail if the
2874SUPER package had already been accessed by other means.
2875
2876=item *
2877
f105b7be 2878Stash aliasing (C<< *foo:: = *bar:: >>) no longer causes SUPER calls to ignore
5ed58cbd
RS
2879changes to methods or @ISA or use the wrong package.
2880
2881=item *
2882
2883Method calls on packages whose names end in ::SUPER are no longer treated
2884as SUPER method calls, resulting in failure to find the method.
2885Furthermore, defining subroutines in such packages no longer causes them to
2886be found by SUPER method calls on the containing package [perl #114924].
2887
2888=item *
2889
2890C<\w> now matches the code points U+200C (ZERO WIDTH NON-JOINER) and U+200D
2891(ZERO WIDTH JOINER). C<\W> no longer matches these. This change is because
2892Unicode corrected their definition of what C<\w> should match.
2893
2894=item *
2895
2896C<dump LABEL> no longer leaks its label.
2897
2898=item *
2899
2900Constant folding no longer changes the behaviour of functions like C<stat()>
2901and C<truncate()> that can take either filenames or handles.
2902C<stat 1 ? foo : bar> nows treats its argument as a file name (since it is an
2903arbitrary expression), rather than the handle "foo".
2904
2905=item *
2906
2907C<truncate FOO, $len> no longer falls back to treating "FOO" as a file name if
e612b5a0 2908the filehandle has been deleted. This was broken in Perl v5.16.0.
5ed58cbd
RS
2909
2910=item *
2911
2912Subroutine redefinitions after sub-to-glob and glob-to-glob assignments no
2913longer cause double frees or panic messages.
2914
2915=item *
2916
2917C<s///> now turns vstrings into plain strings when performing a substitution,
2918even if the resulting string is the same (C<s/a/a/>).
2919
2920=item *
2921
2922Prototype mismatch warnings no longer erroneously treat constant subs as having
2923no prototype when they actually have "".
2924
2925=item *
2926
2927Constant subroutines and forward declarations no longer prevent prototype
2928mismatch warnings from omitting the sub name.
2929
2930=item *
2931
2932C<undef> on a subroutine now clears call checkers.
2933
2934=item *
2935
e612b5a0 2936The C<ref> operator started leaking memory on blessed objects in Perl v5.16.0.
5ed58cbd
RS
2937This has been fixed [perl #114340].
2938
2939=item *
2940
2941C<use> no longer tries to parse its arguments as a statement, making
2942C<use constant { () };> a syntax error [perl #114222].
2943
2944=item *
2945
2946On debugging builds, "uninitialized" warnings inside formats no longer cause
2947assertion failures.
2948
2949=item *
2950
2951On debugging builds, subroutines nested inside formats no longer cause
2952assertion failures [perl #78550].
2953
2954=item *
2955
2956Formats and C<use> statements are now permitted inside formats.
2957
2958=item *
2959
2960C<print $x> and C<sub { print $x }-E<gt>()> now always produce the same output.
2961It was possible for the latter to refuse to close over $x if the variable was
2962not active; e.g., if it was defined outside a currently-running named
2963subroutine.
2964
2965=item *
2966
2967Similarly, C<print $x> and C<print eval '$x'> now produce the same output.
2968This also allows "my $x if 0" variables to be seen in the debugger [perl
2969#114018].
2970
2971=item *
2972
2973Formats called recursively no longer stomp on their own lexical variables, but
2974each recursive call has its own set of lexicals.
2975
2976=item *
2977
2978Attempting to free an active format or the handle associated with it no longer
2979results in a crash.
2980
2981=item *
2982
2983Format parsing no longer gets confused by braces, semicolons and low-precedence
2984operators. It used to be possible to use braces as format delimiters (instead
2985of C<=> and C<.>), but only sometimes. Semicolons and low-precedence operators
2986in format argument lines no longer confuse the parser into ignoring the line's
2987return value. In format argument lines, braces can now be used for anonymous
2988hashes, instead of being treated always as C<do> blocks.
2989
2990=item *
2991
2992Formats can now be nested inside code blocks in regular expressions and other
2993quoted constructs (C</(?{...})/> and C<qq/${...}/>) [perl #114040].
2994
2995=item *
2996
2997Formats are no longer created after compilation errors.
2998
2999=item *
3000
3001Under debugging builds, the B<-DA> command line option started crashing in Perl
e612b5a0 3002v5.16.0. It has been fixed [perl #114368].
5ed58cbd
RS
3003
3004=item *
3005
3006A potential deadlock scenario involving the premature termination of a pseudo-
3007forked child in a Windows build with ithreads enabled has been fixed. This
3008resolves the common problem of the F<t/op/fork.t> test hanging on Windows [perl
3009#88840].
3010
3011=item *
3012
5ed58cbd
RS
3013The code which generates errors from C<require()> could potentially read one or
3014two bytes before the start of the filename for filenames less than three bytes
3015long and ending C</\.p?\z/>. This has now been fixed. Note that it could
3016never have happened with module names given to C<use()> or C<require()> anyway.
3017
3018=item *
3019
3020The handling of pathnames of modules given to C<require()> has been made
3021thread-safe on VMS.
3022
3023=item *
3024
d85cd26b
RS
3025Non-blocking sockets have been fixed on VMS.
3026
3027=item *
3028
5ed58cbd
RS
3029Pod can now be nested in code inside a quoted construct outside of a string
3030eval. This used to work only within string evals [perl #114040].
3031
3032=item *
3033
3034C<goto ''> now looks for an empty label, producing the "goto must have
3035label" error message, instead of exiting the program [perl #111794].
3036
3037=item *
3038
3039C<goto "\0"> now dies with "Can't find label" instead of "goto must have
3040label".
3041
3042=item *
3043
3044The C function C<hv_store> used to result in crashes when used on C<%^H>
3045[perl #111000].
3046
3047=item *
3048
3049A call checker attached to a closure prototype via C<cv_set_call_checker>
3050is now copied to closures cloned from it. So C<cv_set_call_checker> now
3051works inside an attribute handler for a closure.
3052
3053=item *
3054
3055Writing to C<$^N> used to have no effect. Now it croaks with "Modification
3056of a read-only value" by default, but that can be overridden by a custom
3057regular expression engine, as with C<$1> [perl #112184].
3058
3059=item *
3060
3061C<undef> on a control character glob (C<undef *^H>) no longer emits an
3062erroneous warning about ambiguity [perl #112456].
3063
3064=item *
3065
3066For efficiency's sake, many operators and built-in functions return the
3067same scalar each time. Lvalue subroutines and subroutines in the CORE::
3068namespace were allowing this implementation detail to leak through.
3069C<print &CORE::uc("a"), &CORE::uc("b")> used to print "BB". The same thing
3070would happen with an lvalue subroutine returning the return value of C<uc>.
3071Now the value is copied in such cases.
3072
3073=item *
3074
3075C<method {}> syntax with an empty block or a block returning an empty list
3076used to crash or use some random value left on the stack as its invocant.
3077Now it produces an error.
3078
3079=item *
3080
3081C<vec> now works with extremely large offsets (E<gt>2 GB) [perl #111730].
3082
3083=item *
3084
3085Changes to overload settings now take effect immediately, as do changes to
3086inheritance that affect overloading. They used to take effect only after
3087C<bless>.
3088
3089Objects that were created before a class had any overloading used to remain
3090non-overloaded even if the class gained overloading through C<use overload>
3091or @ISA changes, and even after C<bless>. This has been fixed
3092[perl #112708].
3093
3094=item *
3095
3096Classes with overloading can now inherit fallback values.
3097
3098=item *
3099
3100Overloading was not respecting a fallback value of 0 if there were
3101overloaded objects on both sides of an assignment operator like C<+=>
3102[perl #111856].
3103
3104=item *
3105
3106C<pos> now croaks with hash and array arguments, instead of producing
3107erroneous warnings.
3108
3109=item *
3110
3111C<while(each %h)> now implies C<while(defined($_ = each %h))>, like
3112C<readline> and C<readdir>.
3113
3114=item *
3115
3116Subs in the CORE:: namespace no longer crash after C<undef *_> when called
3117with no argument list (C<&CORE::time> with no parentheses).
3118
3119=item *
3120
3121C<unpack> no longer produces the "'/' must follow a numeric type in unpack"
3122error when it is the data that are at fault [perl #60204].
3123
3124=item *
3125
3126C<join> and C<"@array"> now call FETCH only once on a tied C<$">
3127[perl #8931].
3128
3129=item *
3130
3131Some subroutine calls generated by compiling core ops affected by a
3132C<CORE::GLOBAL> override had op checking performed twice. The checking
3133is always idempotent for pure Perl code, but the double checking can
3134matter when custom call checkers are involved.
3135
3136=item *
3137
3138A race condition used to exist around fork that could cause a signal sent to
3139the parent to be handled by both parent and child. Signals are now blocked
3140briefly around fork to prevent this from happening [perl #82580].
3141
3142=item *
3143
3144The implementation of code blocks in regular expressions, such as C<(?{})>
3145and C<(??{})>, has been heavily reworked to eliminate a whole slew of bugs.
3146The main user-visible changes are:
3147
3148=over 4
3149
3150=item *
3151
3152Code blocks within patterns are now parsed in the same pass as the
3153surrounding code; in particular it is no longer necessary to have balanced
3154braces: this now works:
3155
3156 /(?{ $x='{' })/
3157
3158This means that this error message is no longer generated:
3159
3160 Sequence (?{...}) not terminated or not {}-balanced in regex
3161
3162but a new error may be seen:
3163
3164 Sequence (?{...}) not terminated with ')'
3165
3166In addition, literal code blocks within run-time patterns are only
3167compiled once, at perl compile-time:
3168
3169 for my $p (...) {
3170 # this 'FOO' block of code is compiled once,
3171 # at the same time as the surrounding 'for' loop
3172 /$p{(?{FOO;})/;
3173 }
3174
3175=item *
3176
3177Lexical variables are now sane as regards scope, recursion and closure
3178behavior. In particular, C</A(?{B})C/> behaves (from a closure viewpoint)
3179exactly like C</A/ && do { B } && /C/>, while C<qr/A(?{B})C/> is like
3180C<sub {/A/ && do { B } && /C/}>. So this code now works how you might
3181expect, creating three regexes that match 0, 1, and 2:
3182
3183 for my $i (0..2) {
3184 push @r, qr/^(??{$i})$/;
3185 }
3186 "1" =~ $r[1]; # matches
3187
3188=item *
3189
3190The C<use re 'eval'> pragma is now only required for code blocks defined
3191at runtime; in particular in the following, the text of the C<$r> pattern is
3192still interpolated into the new pattern and recompiled, but the individual
3193compiled code-blocks within C<$r> are reused rather than being recompiled,
3194and C<use re 'eval'> isn't needed any more:
3195
3196 my $r = qr/abc(?{....})def/;
3197 /xyz$r/;
3198
3199=item *
3200
3201Flow control operators no longer crash. Each code block runs in a new
3202dynamic scope, so C<next> etc. will not see
3203any enclosing loops. C<return> returns a value
3204from the code block, not from any enclosing subroutine.
3205
3206=item *
3207
3208Perl normally caches the compilation of run-time patterns, and doesn't
3209recompile if the pattern hasn't changed, but this is now disabled if
3210required for the correct behavior of closures. For example:
3211
3212 my $code = '(??{$x})';
3213 for my $x (1..3) {
3214 # recompile to see fresh value of $x each time
3215 $x =~ /$code/;
3216 }
3217
3218=item *
3219
3220The C</msix> and C<(?msix)> etc. flags are now propagated into the return
3221value from C<(??{})>; this now works:
3222
3223 "AB" =~ /a(??{'b'})/i;
3224
3225=item *
3226
3227Warnings and errors will appear to come from the surrounding code (or for
3228run-time code blocks, from an eval) rather than from an C<re_eval>:
3229
3230 use re 'eval'; $c = '(?{ warn "foo" })'; /$c/;
3231 /(?{ warn "foo" })/;
3232
3233formerly gave:
3234
3235 foo at (re_eval 1) line 1.
3236 foo at (re_eval 2) line 1.
3237
3238and now gives:
3239
3240 foo at (eval 1) line 1.
3241 foo at /some/prog line 2.
3242
3243=back
3244
3245=item *
3246
2e7bc647
KW
3247Perl now can be recompiled to use any Unicode version. In v5.16, it
3248worked on Unicodes 6.0 and 6.1, but there were various bugs if earlier
3249releases were used; the older the release the more problems.
5ed58cbd
RS
3250
3251=item *
3252
3253C<vec> no longer produces "uninitialized" warnings in lvalue context
3254[perl #9423].
3255
3256=item *
3257
3258An optimization involving fixed strings in regular expressions could cause
3259a severe performance penalty in edge cases. This has been fixed
3260[perl #76546].
3261
3262=item *
3263
3264In certain cases, including empty subpatterns within a regular expression (such
3265as C<(?:)> or C<(?:|)>) could disable some optimizations. This has been fixed.
3266
3267=item *
3268
3269The "Can't find an opnumber" message that C<prototype> produces when passed
3270a string like "CORE::nonexistent_keyword" now passes UTF-8 and embedded
3271NULs through unchanged [perl #97478].
3272
3273=item *
3274
3275C<prototype> now treats magical variables like C<$1> the same way as
3276non-magical variables when checking for the CORE:: prefix, instead of
3277treating them as subroutine names.
3278
3279=item *
3280
3281Under threaded perls, a runtime code block in a regular expression could
3282corrupt the package name stored in the op tree, resulting in bad reads
3283in C<caller>, and possibly crashes [perl #113060].
3284
3285=item *
3286
3287Referencing a closure prototype (C<\&{$_[1]}> in an attribute handler for a
3288closure) no longer results in a copy of the subroutine (or assertion
3289failures on debugging builds).
3290
3291=item *
3292
3293C<eval '__PACKAGE__'> now returns the right answer on threaded builds if
3294the current package has been assigned over (as in
3295C<*ThisPackage:: = *ThatPackage::>) [perl #78742].
3296
3297=item *
3298
3299If a package is deleted by code that it calls, it is possible for C<caller>
3300to see a stack frame belonging to that deleted package. C<caller> could
3301crash if the stash's memory address was reused for a scalar and a
3302substitution was performed on the same scalar [perl #113486].
3303
3304=item *
3305
3306C<UNIVERSAL::can> no longer treats its first argument differently
3307depending on whether it is a string or number internally.
3308
3309=item *
3310
3311C<open> with C<< <& >> for the mode checks to see whether the third argument is
3312a number, in determining whether to treat it as a file descriptor or a handle
3313name. Magical variables like C<$1> were always failing the numeric check and
3314being treated as handle names.
3315
3316=item *
3317
3318C<warn>'s handling of magical variables (C<$1>, ties) has undergone several
3319fixes. C<FETCH> is only called once now on a tied argument or a tied C<$@>
3320[perl #97480]. Tied variables returning objects that stringify as "" are
3321no longer ignored. A tied C<$@> that happened to return a reference the
3322I<previous> time it was used is no longer ignored.
3323
3324=item *
3325
3326C<warn ""> now treats C<$@> with a number in it the same way, regardless of
3327whether it happened via C<$@=3> or C<$@="3">. It used to ignore the
3328former. Now it appends "\t...caught", as it has always done with
3329C<$@="3">.
3330
3331=item *
3332
3333Numeric operators on magical variables (e.g., S<C<$1 + 1>>) used to use
3334floating point operations even where integer operations were more appropriate,
3335resulting in loss of accuracy on 64-bit platforms [perl #109542].
3336
3337=item *
3338
3339Unary negation no longer treats a string as a number if the string happened
3340to be used as a number at some point. So, if C<$x> contains the string "dogs",
3341C<-$x> returns "-dogs" even if C<$y=0+$x> has happened at some point.
3342
3343=item *
3344
e612b5a0 3345In Perl v5.14, C<-'-10'> was fixed to return "10", not "+10". But magical
5ed58cbd
RS
3346variables (C<$1>, ties) were not fixed till now [perl #57706].
3347
3348=item *
3349
3350Unary negation now treats strings consistently, regardless of the internal
3351C<UTF8> flag.
3352
3353=item *
3354
3355A regression introduced in Perl v5.16.0 involving
3356C<tr/I<SEARCHLIST>/I<REPLACEMENTLIST>/> has been fixed. Only the first
3357instance is supposed to be meaningful if a character appears more than
3358once in C<I<SEARCHLIST>>. Under some circumstances, the final instance
3359was overriding all earlier ones. [perl #113584]
3360
3361=item *
3362
3363Regular expressions like C<qr/\87/> previously silently inserted a NUL
3364character, thus matching as if it had been written C<qr/\00087/>. Now it
3365matches as if it had been written as C<qr/87/>, with a message that the
3366sequence C<"\8"> is unrecognized.
3367
3368=item *
3369
3370C<__SUB__> now works in special blocks (C<BEGIN>, C<END>, etc.).
3371
3372=item *
3373
3374Thread creation on Windows could theoretically result in a crash if done
3375inside a C<BEGIN> block. It still does not work properly, but it no longer
3376crashes [perl #111610].
3377
3378=item *
3379
3380C<\&{''}> (with the empty string) now autovivifies a stub like any other
3381sub name, and no longer produces the "Unable to create sub" error
3382[perl #94476].
3383
3384=item *
3385
3386A regression introduced in v5.14.0 has been fixed, in which some calls
3387to the C<re> module would clobber C<$_> [perl #113750].
3388
3389=item *
3390
3391C<do FILE> now always either sets or clears C<$@>, even when the file can't be
3392read. This ensures that testing C<$@> first (as recommended by the
3393documentation) always returns the correct result.
3394
3395=item *
3396
3397The array iterator used for the C<each @array> construct is now correctly
237b9c6d
SK
3398reset when C<@array> is cleared [perl #75596]. This happens, for example, when
3399the array is globally assigned to, as in C<@array = (...)>, but not when its
5ed58cbd
RS
3400B<values> are assigned to. In terms of the XS API, it means that C<av_clear()>
3401will now reset the iterator.
3402
3403This mirrors the behaviour of the hash iterator when the hash is cleared.
3404
3405=item *
3406
3407C<< $class->can >>, C<< $class->isa >>, and C<< $class->DOES >> now return
3408correct results, regardless of whether that package referred to by C<$class>
3409exists [perl #47113].
3410
3411=item *
3412
3413Arriving signals no longer clear C<$@> [perl #45173].
3414
3415=item *
3416
3417Allow C<my ()> declarations with an empty variable list [perl #113554].
3418
3419=item *
3420
3421During parsing, subs declared after errors no longer leave stubs
3422[perl #113712].
3423
3424=item *
3425
3426Closures containing no string evals no longer hang on to their containing
3427subroutines, allowing variables closed over by outer subroutines to be
3428freed when the outer sub is freed, even if the inner sub still exists
3429[perl #89544].
3430
3431=item *
3432
3433Duplication of in-memory filehandles by opening with a "<&=" or ">&=" mode
e612b5a0 3434stopped working properly in v5.16.0. It was causing the new handle to
5ed58cbd
RS
3435reference a different scalar variable. This has been fixed [perl #113764].
3436
3437=item *
3438
3439C<qr//> expressions no longer crash with custom regular expression engines
3440that do not set C<offs> at regular expression compilation time
3441[perl #112962].
3442
3443=item *
3444
3445C<delete local> no longer crashes with certain magical arrays and hashes
3446[perl #112966].
3447
3448=item *
3449
3450C<local> on elements of certain magical arrays and hashes used not to
3451arrange to have the element deleted on scope exit, even if the element did
3452not exist before C<local>.
3453
3454=item *
3455
3456C<scalar(write)> no longer returns multiple items [perl #73690].
3457
3458=item *
3459
3460String to floating point conversions no longer misparse certain strings under
3461C<use locale> [perl #109318].
3462
3463=item *
3464
3465C<@INC> filters that die no longer leak memory [perl #92252].
3466
3467=item *
3468
3469The implementations of overloaded operations are now called in the correct
3470context. This allows, among other things, being able to properly override
3471C<< <> >> [perl #47119].
3472
3473=item *
3474
3475Specifying only the C<fallback> key when calling C<use overload> now behaves
3476properly [perl #113010].
3477
3478=item *
3479
3480C<< sub foo { my $a = 0; while ($a) { ... } } >> and
3481C<< sub foo { while (0) { ... } } >> now return the same thing [perl #73618].
3482
3483=item *
3484
3485String negation now behaves the same under C<use integer;> as it does
3486without [perl #113012].
3487
3488=item *
3489
3490C<chr> now returns the Unicode replacement character (U+FFFD) for -1,
3491regardless of the internal representation. -1 used to wrap if the argument
3492was tied or a string internally.
3493
3494=item *
3495
3496Using a C<format> after its enclosing sub was freed could crash as of
e612b5a0 3497perl v5.12.0, if the format referenced lexical variables from the outer sub.
5ed58cbd
RS
3498
3499=item *
3500
3501Using a C<format> after its enclosing sub was undefined could crash as of
e612b5a0 3502perl v5.10.0, if the format referenced lexical variables from the outer sub.
5ed58cbd
RS
3503
3504=item *
3505
3506Using a C<format> defined inside a closure, which format references
3507lexical variables from outside, never really worked unless the C<write>
e612b5a0 3508call was directly inside the closure. In v5.10.0 it even started crashing.
5ed58cbd
RS
3509Now the copy of that closure nearest the top of the call stack is used to
3510find those variables.
3511
3512=item *
3513
3514Formats that close over variables in special blocks no longer crash if a
3515stub exists with the same name as the special block before the special
3516block is compiled.
3517
3518=item *
3519
3520The parser no longer gets confused, treating C<eval foo ()> as a syntax
3521error if preceded by C<print;> [perl #16249].
3522
3523=item *
3524
3525The return value of C<syscall> is no longer truncated on 64-bit platforms
3526[perl #113980].
3527
3528=item *
3529
3530Constant folding no longer causes C<print 1 ? FOO : BAR> to print to the
3531FOO handle [perl #78064].
3532
3533=item *
3534
3535C<do subname> now calls the named subroutine and uses the file name it
3536returns, instead of opening a file named "subname".
3537
3538=item *
3539
3540Subroutines looked up by rv2cv check hooks (registered by XS modules) are
3541now taken into consideration when determining whether C<foo bar> should be
3542the sub call C<foo(bar)> or the method call C<< "bar"->foo >>.
3543
3544=item *
3545
3546C<CORE::foo::bar> is no longer treated specially, allowing global overrides
3547to be called directly via C<CORE::GLOBAL::uc(...)> [perl #113016].
3548
3549=item *
3550
3551Calling an undefined sub whose typeglob has been undefined now produces the
3552customary "Undefined subroutine called" error, instead of "Not a CODE
3553reference".
3554
3555=item *
3556
3557Two bugs involving @ISA have been fixed. C<*ISA = *glob_without_array> and
3558C<undef *ISA; @{*ISA}> would prevent future modifications to @ISA from
3559updating the internal caches used to look up methods. The
e612b5a0 3560*glob_without_array case was a regression from Perl v5.12.
5ed58cbd
RS
3561
3562=item *
3563
3564Regular expression optimisations sometimes caused C<$> with C</m> to
3565produce failed or incorrect matches [perl #114068].
3566
3567=item *
3568
3569C<__SUB__> now works in a C<sort> block when the enclosing subroutine is
3570predeclared with C<sub foo;> syntax [perl #113710].
3571
3572=item *
3573
3574Unicode properties only apply to Unicode code points, which leads to
3575some subtleties when regular expressions are matched against
3576above-Unicode code points. There is a warning generated to draw your
3577attention to this. However, this warning was being generated
3578inappropriately in some cases, such as when a program was being parsed.
3d3677c1 3579Non-Unicode matches such as C<\w> and C<[:word:]> should not generate the
5ed58cbd
RS
3580warning, as their definitions don't limit them to apply to only Unicode
3581code points. Now the message is only generated when matching against
3582C<\p{}> and C<\P{}>. There remains a bug, [perl #114148], for the very
3583few properties in Unicode that match just a single code point. The
3584warning is not generated if they are matched against an above-Unicode
3585code point.
3586
3587=item *
3588
3589Uninitialized warnings mentioning hash elements would only mention the
3590element name if it was not in the first bucket of the hash, due to an
3591off-by-one error.
3592
3593=item *
3594
3595A regular expression optimizer bug could cause multiline "^" to behave
3596incorrectly in the presence of line breaks, such that
3597C<"/\n\n" =~ m#\A(?:^/$)#im> would not match [perl #115242].
3598
3599=item *
3600
3601Failed C<fork> in list context no longer corrupts the stack.
3602C<@a = (1, 2, fork, 3)> used to gobble up the 2 and assign C<(1, undef, 3)>
3603if the C<fork> call failed.
3604
3605=item *
3606
3607Numerous memory leaks have been fixed, mostly involving tied variables that
3608die, regular expression character classes and code blocks, and syntax
3609errors.
3610
3611=item *
3612
3613Assigning a regular expression (C<${qr//}>) to a variable that happens to
3614hold a floating point number no longer causes assertion failures on
3615debugging builds.
3616
3617=item *
3618
3619Assigning a regular expression to a scalar containing a number no longer
f105b7be 3620causes subsequent numification to produce random numbers.
5ed58cbd
RS
3621
3622=item *
3623
3624Assigning a regular expression to a magic variable no longer wipes away the
e612b5a0 3625magic. This was a regression from v5.10.
5ed58cbd
RS
3626
3627=item *
3628
3629Assigning a regular expression to a blessed scalar no longer results in
e612b5a0 3630crashes. This was also a regression from v5.10.
5ed58cbd
RS
3631
3632=item *
3633
3634Regular expression can now be assigned to tied hash and array elements with
3635flattening into strings.
3636
3637=item *
3638
f105b7be 3639Numifying a regular expression no longer results in an uninitialized
5ed58cbd
RS
3640warning.
3641
3642=item *
3643
3644Negative array indices no longer cause EXISTS methods of tied variables to
e612b5a0 3645be ignored. This was a regression from v5.12.
5ed58cbd
RS
3646
3647=item *
3648
3649Negative array indices no longer result in crashes on arrays tied to
3650non-objects.
3651
3652=item *
3653
be12dd22
RS
3654C<$byte_overload .= $utf8> no longer results in doubly-encoded UTF-8 if the
3655left-hand scalar happened to have produced a UTF-8 string the last time
5ed58cbd
RS
3656overloading was invoked.
3657
3658=item *
3659
3660C<goto &sub> now uses the current value of @_, instead of using the array
3661the subroutine was originally called with. This means
3662C<local @_ = (...); goto &sub> now works [perl #43077].
3663
3664=item *
3665
3666If a debugger is invoked recursively, it no longer stomps on its own
3667lexical variables. Formerly under recursion all calls would share the same
3668set of lexical variables [perl #115742].
3669
3670=item *
3671
3672C<*_{ARRAY}> returned from a subroutine no longer spontaneously
3673becomes empty.
3674
3675=back
3676
3677=head1 Known Problems
3678
3679=over 4
3680
3681=item *
3682
7f50b25b 3683There are no known regressions. Please report any bugs you find!
5ed58cbd
RS
3684
3685=back
3686
32b79602
RS
3687=head1 Obituary
3688
3689Hojung Yoon (AMORETTE), 24, of Seoul, South Korea, went to his long rest
3690on May 8, 2013 with llama figurine and autographed TIMTOADY card. He
3691was a brilliant young Perl 5 & 6 hacker and a devoted member of
55c6e7a8 3692Seoul.pm. He programmed Perl, talked Perl, ate Perl, and loved Perl. We
32b79602
RS
3693believe that he is still programming in Perl with his broken IBM laptop
3694somewhere. He will be missed.
3695
5ed58cbd 3696=head1 Acknowledgements
a75569c0 3697
dd38235c
RS
3698Perl v5.18.0 represents approximately 12 months of development since
3699Perl v5.16.0 and contains approximately 400,000 lines of changes across
37002,100 files from 113 authors.
3701
3702Perl continues to flourish into its third decade thanks to a vibrant
3703community of users and developers. The following people are known to
3704have contributed the improvements that became Perl v5.18.0:
3705
3706Aaron Crane, Aaron Trevena, Abhijit Menon-Sen, Adrian M. Enache, Alan
3707Haggai Alavi, Alexandr Ciornii, Andrew Tam, Andy Dougherty, Anton Nikishaev,
3708Aristotle Pagaltzis, Augustina Blair, Bob Ernst, Brad Gilbert, Breno G. de
3709Oliveira, Brian Carlson, Brian Fraser, Charlie Gonzalez, Chip Salzenberg, Chris
3710'BinGOs' Williams, Christian Hansen, Colin Kuskie, Craig A. Berry, Dagfinn
3711Ilmari Mannsåker, Daniel Dragan, Daniel Perrett, Darin McBride, Dave Rolsky,
3712David Golden, David Leadbeater, David Mitchell, David Nicol, Dominic
3713Hargreaves, E. Choroba, Eric Brine, Evan Miller, Father Chrysostomos, Florian
3714Ragwitz, François Perrad, George Greer, Goro Fuji, H.Merijn Brand, Herbert
3715Breunung, Hugo van der Sanden, Igor Zaytsev, James E Keenan, Jan Dubois,
3716Jasmine Ahuja, Jerry D. Hedden, Jess Robinson, Jesse Luehrs, Joaquin Ferrero,
3717Joel Berger, John Goodyear, John Peacock, Karen Etheridge, Karl Williamson,
3718Karthik Rajagopalan, Kent Fredric, Leon Timmermans, Lucas Holt, Lukas Mai,
3719Marcus Holland-Moritz, Markus Jansen, Martin Hasch, Matthew Horsfall, Max
3720Maischein, Michael G Schwern, Michael Schroeder, Moritz Lenz, Nicholas Clark,
3721Niko Tyni, Oleg Nesterov, Patrik Hägglund, Paul Green, Paul Johnson, Paul
3722Marquess, Peter Martini, Rafael Garcia-Suarez, Reini Urban, Renee Baecker,
3723Rhesa Rozendaal, Ricardo Signes, Robin Barker, Ronald J. Kimball, Ruslan
3724Zakirov, Salvador Fandiño, Sawyer X, Scott Lanning, Sergey Alekseev, Shawn M
3725Moore, Shirakata Kentaro, Shlomi Fish, Sisyphus, Smylers, Steffen Müller,
3726Steve Hay, Steve Peters, Steven Schubiger, Sullivan Beck, Sven Strickroth,
3727Sébastien Aperghis-Tramoni, Thomas Sibley, Tobias Leich, Tom Wyant, Tony Cook,
3728Vadim Konovalov, Vincent Pit, Volker Schatz, Walt Mankowski, Yves Orton,
3729Zefram.
3730
3731The list above is almost certainly incomplete as it is automatically generated
3732from version control history. In particular, it does not include the names of
3733the (very much appreciated) contributors who reported issues to the Perl bug
3734tracker.
3735
3736Many of the changes included in this version originated in the CPAN modules
3737included in Perl's core. We're grateful to the entire CPAN community for
3738helping Perl to flourish.
3739
3740For a more complete list of all of Perl's historical contributors, please see
3741the F<AUTHORS> file in the Perl source distribution.
f5b73711 3742
44691e6f
AB
3743=head1 Reporting Bugs
3744
e08634c5
SH
3745If you find what you think is a bug, you might check the articles recently
3746posted to the comp.lang.perl.misc newsgroup and the perl bug database at
3747http://rt.perl.org/perlbug/ . There may also be information at
3748http://www.perl.org/ , the Perl Home Page.
44691e6f 3749
e08634c5
SH
3750If you believe you have an unreported bug, please run the L<perlbug> program
3751included with your release. Be sure to trim your bug down to a tiny but
3752sufficient test case. Your bug report, along with the output of C<perl -V>,
3753will be sent off to perlbug@perl.org to be analysed by the Perl porting team.
44691e6f
AB
3754
3755If the bug you are reporting has security implications, which make it
e08634c5
SH
3756inappropriate to send to a publicly archived mailing list, then please send it
3757to perl5-security-report@perl.org. This points to a closed subscription
3758unarchived mailing list, which includes all the core committers, who will be
3759able to help assess the impact of issues, figure out a resolution, and help
f9001595 3760co-ordinate the release of patches to mitigate or fix the problem across all
e08634c5
SH
3761platforms on which Perl is supported. Please only use this address for
3762security issues in the Perl core, not for modules independently distributed on
3763CPAN.
44691e6f
AB
3764
3765=head1 SEE ALSO
3766
e08634c5
SH
3767The F<Changes> file for an explanation of how to view exhaustive details on
3768what changed.
44691e6f
AB
3769
3770The F<INSTALL> file for how to build Perl.
3771
3772The F<README> file for general stuff.
3773
3774The F<Artistic> and F<Copying> files for copyright information.
3775
3776=cut