perl5180delta - Man Page

Newly-added experimental features will now require this incantation:

    no warnings "experimental::feature_name";
    use feature "feature_name";  # would warn without the prev line

There is a new warnings category, called "experimental", containing warnings that the feature pragma emits when enabling experimental features.

Newly-added experimental features will also be given special warning IDs, which consist of "experimental::" followed by the name of the feature.  (The plan is to extend this mechanism eventually to all warnings, to allow them to be enabled or disabled individually, and not just by category.)

By saying

    no warnings "experimental::feature_name";

you are taking responsibility for any breakage that future changes to, or removal of, the feature may cause.

Since some features (like ~~ or my $_) now emit experimental warnings, and you may want to disable them in code that is also run on perls that do not recognize these warning categories, consider using the if pragma like this:

    no if $] >= 5.018, warnings => "experimental::feature_name";

Existing experimental features may begin emitting these warnings, too.  Please consult perlexperiment for information on which features are considered experimental.

Changes to the implementation of hashes in perl v5.18.0 will be one of the most visible changes to the behavior of existing code.

By default, two distinct hash variables with identical keys and values may now provide their contents in a different order where it was previously identical.

When encountering these changes, the key to cleaning up from them is to accept that hashes are unordered collections and to act accordingly.

Hash randomization

The seed used by Perl's hash function is now random.  This means that the order which keys/values will be returned from functions like keys(), values(), and each() will differ from run to run.

This change was introduced to make Perl's hashes more robust to algorithmic complexity attacks, and also because we discovered that it exposes hash ordering dependency bugs and makes them easier to track down.

Toolchain maintainers might want to invest in additional infrastructure to test for things like this.  Running tests several times in a row and then comparing results will make it easier to spot hash order dependencies in code.  Authors are strongly encouraged not to expose the key order of Perl's hashes to insecure audiences.

Further, every hash has its own iteration order, which should make it much more difficult to determine what the current hash seed is.

New hash functions

Perl v5.18 includes support for multiple hash functions, and changed the default (to ONE_AT_A_TIME_HARD), you can choose a different algorithm by defining a symbol at compile time.  For a current list, consult the INSTALL document.  Note that as of Perl v5.18 we can only recommend use of the default or SIPHASH. All the others are known to have security issues and are for research purposes only.

PERL_HASH_SEED environment variable now takes a hex value

PERL_HASH_SEED no longer accepts an integer as a parameter; instead the value is expected to be a binary value encoded in a hex string, such as "0xf5867c55039dc724".  This is to make the infrastructure support hash seeds of arbitrary lengths, which might exceed that of an integer.  (SipHash uses a 16 byte seed.)

PERL_PERTURB_KEYS environment variable added

The PERL_PERTURB_KEYS environment variable allows one to control the level of randomization applied to keys and friends.

When PERL_PERTURB_KEYS is 0, perl will not randomize the key order at all. The chance that keys changes due to an insert will be the same as in previous perls, basically only when the bucket size is changed.

When PERL_PERTURB_KEYS is 1, perl will randomize keys in a non-repeatable way. The chance that keys changes due to an insert will be very high.  This is the most secure and default mode.

When PERL_PERTURB_KEYS is 2, perl will randomize keys in a repeatable way. Repeated runs of the same program should produce the same output every time.

PERL_HASH_SEED implies a non-default PERL_PERTURB_KEYS setting. Setting PERL_HASH_SEED=0 (exactly one 0) implies PERL_PERTURB_KEYS=0 (hash key randomization disabled); setting PERL_HASH_SEED to any other value implies PERL_PERTURB_KEYS=2 (deterministic and repeatable hash key randomization). Specifying PERL_PERTURB_KEYS explicitly to a different level overrides this behavior.

Hash::Util::hash_seed() now returns a string

Hash::Util::hash_seed() now returns a string instead of an integer.  This is to make the infrastructure support hash seeds of arbitrary lengths which might exceed that of an integer.  (SipHash uses a 16 byte seed.)

Output of PERL_HASH_SEED_DEBUG has been changed

The environment variable PERL_HASH_SEED_DEBUG now makes perl show both the hash function perl was built with, and the seed, in hex, in use for that process. Code parsing this output, should it exist, must change to accommodate the new format.  Example of the new format:

    $ PERL_HASH_SEED_DEBUG=1 ./perl -e1
    HASH_FUNCTION = MURMUR3 HASH_SEED = 0x1476bb9f

Perl now supports Unicode 6.2.  A list of changes from Unicode 6.1 is at <http://www.unicode.org/versions/Unicode6.2.0>.

It is possible to define your own names for characters for use in \N{...}, charnames::vianame(), etc.  These names can now consist of characters from the whole Unicode range.  This allows for names to be in your native language, and not just English.  Certain restrictions apply to the characters that may be used (you can't define a name that has punctuation in it, for example).  See "CUSTOM ALIASES" in charnames.

The following new DTrace probes have been added:

• op-entry
• loading-file
• loaded-file

This new variable provides access to the filehandle that was last read. This is the handle used by $. and by tell and eof without arguments.

This is an experimental feature to allow matching against the union, intersection, etc., of sets of code points, similar to Unicode::Regex::Set.  It can also be used to extend /x processing to [bracketed] character classes, and as a replacement of user-defined properties, allowing more complex expressions than they do.  See "Extended Bracketed Character Classes" in perlrecharclass.

This new feature is still considered experimental.  To enable it:

    use 5.018;
    no warnings "experimental::lexical_subs";
    use feature "lexical_subs";

You can now declare subroutines with state sub foo, my sub foo, and our sub foo.  (state sub requires that the "state" feature be enabled, unless you write it as CORE::state sub foo.)

state sub creates a subroutine visible within the lexical scope in which it is declared.  The subroutine is shared between calls to the outer sub.

my sub declares a lexical subroutine that is created each time the enclosing block is entered.  state sub is generally slightly faster than my sub.

our sub declares a lexical alias to the package subroutine of the same name.

For more information, see "Lexical Subroutines" in perlsub.

The loop controls next, last and redo, and the special dump operator, now allow arbitrary expressions to be used to compute labels at run time.  Previously, any argument that was not a constant was treated as the empty string.

Several more built-in functions have been added as subroutines to the CORE:: namespace - namely, those non-overridable keywords that can be implemented without custom parsers: defined, delete, exists, glob, pos, prototype, scalar, split, study, and undef.

As some of these have prototypes, prototype('CORE::...') has been changed to not make a distinction between overridable and non-overridable keywords.  This is to make prototype('CORE::pos') consistent with prototype(&CORE::pos).

kill has always allowed a negative signal number, which kills the process group instead of a single process.  It has also allowed signal names.  But it did not behave consistently, because negative signal names were treated as 0.  Now negative signals names like -INT are supported and treated the same way as -2 [perl #112990].

Some of the changes in the hash overhaul were made to enhance security.  Please read that section.

The documentation for Storable now includes a section which warns readers of the danger of accepting Storable documents from untrusted sources. The short version is that deserializing certain types of data can lead to loading modules and other code execution. This is documented behavior and wanted behavior, but this opens an attack vector for malicious entities.

If users could provide a translation string to Locale::Maketext, this could be used to invoke arbitrary Perl subroutines available in the current process.

This has been fixed, but it is still possible to invoke any method provided by Locale::Maketext itself or a subclass that you are using. One of these methods in turn will invoke the Perl core's sprintf subroutine.

In summary, allowing users to provide translation strings without auditing them is a bad idea.

This vulnerability is documented in CVE-2012-6329.

Poorly written perl code that allows an attacker to specify the count to perl's x string repeat operator can already cause a memory exhaustion denial-of-service attack. A flaw in versions of perl before v5.15.5 can escalate that into a heap buffer overrun; coupled with versions of glibc before 2.16, it possibly allows the execution of arbitrary code.

The flaw addressed to this commit has been assigned identifier CVE-2012-5195 and was researched by Tim Brown.

Some of the changes in the hash overhaul are not fully compatible with previous versions of perl.  Please read that section.

Previously, it warned, and the Unicode REPLACEMENT CHARACTER was substituted.  Unicode now recommends that this situation be a syntax error.  Also, the previous behavior led to some confusing warnings and behaviors, and since the REPLACEMENT CHARACTER has no use other than as a stand-in for some unknown character, any code that has this problem is buggy.

Since v5.12.0, it has been deprecated to use certain characters in user-defined \N{...} character names.  These now cause a syntax error.  For example, it is now an error to begin a name with a digit, such as in

 my $undraftable = "\N{4F}";    # Syntax error!

or to have commas anywhere in the name.  See "CUSTOM ALIASES" in charnames.

Unicode 6.0 reused the name "BELL" for a different code point than it traditionally had meant.  Since Perl v5.14, use of this name still referred to U+0007, but would raise a deprecation warning.  Now, "BELL" refers to U+1F514, and the name for U+0007 is "ALERT".  All the functions in charnames have been correspondingly updated.

Unicode has now withdrawn their previous recommendation for regular expressions to automatically handle cases where a single character can match multiple characters case-insensitively, for example, the letter LATIN SMALL LETTER SHARP S and the sequence ss.  This is because it turns out to be impracticable to do this correctly in all circumstances.  Because Perl has tried to do this as best it can, it will continue to do so.  (We are considering an option to turn it off.) However, a new restriction is being added on such matches when they occur in [bracketed] character classes.  People were specifying things such as /[\0-\xff]/i, and being surprised that it matches the two character sequence ss (since LATIN SMALL LETTER SHARP S occurs in this range).  This behavior is also inconsistent with using a property instead of a range:  \p{Block=Latin1} also includes LATIN SMALL LETTER SHARP S, but /[\p{Block=Latin1}]/i does not match ss. The new rule is that for there to be a multi-character case-insensitive match within a bracketed character class, the character must be explicitly listed, and not as an end point of a range.  This more closely obeys the Principle of Least Astonishment.  See "Bracketed Character Classes" in perlrecharclass.  Note that a bug [perl #89774], now fixed as part of this change, prevented the previous behavior from working fully.

Due to an oversight, single character variable names in v5.16 were completely unrestricted.  This opened the door to several kinds of insanity.  As of v5.18, these now follow the rules of other identifiers, in addition to accepting characters that match the \p{POSIX_Punct} property.

There is no longer any difference in the parsing of identifiers specified by using braces versus without braces.  For instance, perl used to allow ${foo:bar} (with a single colon) but not $foo:bar. Now that both are handled by a single code path, they are both treated the same way: both are forbidden.  Note that this change is about the range of permissible literal identifiers, not other expressions.

No one could recall why \s didn't match \cK, the vertical tab. Now it does.  Given the extreme rarity of that character, very little breakage is expected.  That said, here's what it means:

\s in a regex now matches a vertical tab in all circumstances.

Literal vertical tabs in a regex literal are ignored when the /x modifier is used.

Leading vertical tabs, alone or mixed with other whitespace, are now ignored when interpreting a string as a number.  For example:

  $dec = " \cK \t 123";
  $hex = " \cK \t 0xF";

  say 0 + $dec;   # was 0 with warning, now 123
  say int $dec;   # was 0, now 123
  say oct $hex;   # was 0, now  15

The implementation of this feature has been almost completely rewritten. Although its main intent is to fix bugs, some behaviors, especially related to the scope of lexical variables, will have changed.  This is described more fully in the "Selected Bug Fixes" section.

It is no longer possible to abuse the way the parser parses s///e like this:

    %_=(_,"Just another ");
    $_="Perl hacker,\n";
    s//_}->{_/e;print

Instead of assigning to an implicit lexical $_, given now makes the global $_ an alias for its argument, just like foreach.  However, it still uses lexical $_ if there is lexical $_ in scope (again, just like foreach) [perl #114020].

Smart match, added in v5.10.0 and significantly revised in v5.10.1, has been a regular point of complaint.  Although there are a number of ways in which it is useful, it has also proven problematic and confusing for both users and implementors of Perl.  There have been a number of proposals on how to best address the problem.  It is clear that smartmatch is almost certainly either going to change or go away in the future.  Relying on its current behavior is not recommended.

Warnings will now be issued when the parser sees ~~, given, or when. To disable these warnings, you can add this line to the appropriate scope:

  no if $] >= 5.018, warnings => "experimental::smartmatch";

Consider, though, replacing the use of these features, as they may change behavior again before becoming stable.

Since it was introduced in Perl v5.10, it has caused much confusion with no obvious solution:

• Various modules (e.g., List::Util) expect callback routines to use the global $_.  use List::Util 'first'; my $_; first { $_ == 1 } @list does not work as one would expect.
• A my $_ declaration earlier in the same file can cause confusing closure warnings.
• The "_" subroutine prototype character allows called subroutines to access your lexical $_, so it is not really private after all.
• Nevertheless, subroutines with a "(@)" prototype and methods cannot access the caller's lexical $_, unless they are written in XS.
• But even XS routines cannot access a lexical $_ declared, not in the calling subroutine, but in an outer scope, iff that subroutine happened not to mention $_ or use any operators that default to $_.

It is our hope that lexical $_ can be rehabilitated, but this may cause changes in its behavior.  Please use it with caution until it becomes stable.

Previously, when reading from a stream with I/O layers such as encoding, the readline() function, otherwise known as the <> operator, would read N bytes from the top-most layer. [perl #79960]

Now, N characters are read instead.

There is no change in behaviour when reading from streams with no extra layers, since bytes map exactly to characters.

glob overrides used to be passed a magical undocumented second argument that identified the caller.  Nothing on CPAN was using this, and it got in the way of a bug fix, so it was removed.  If you really need to identify the caller, see Devel::Callsite on CPAN.

The body of a here document inside a quote-like operator now always begins on the line after the "<<foo" marker.  Previously, it was documented to begin on the line following the containing quote-like operator, but that was only sometimes the case [perl #114040].

You may no longer write something like:

 m/a/and 1

Instead you must write

 m/a/ and 1

with whitespace separating the operator from the closing delimiter of the regular expression.  Not having whitespace has resulted in a deprecation warning since Perl v5.14.0.

qw lists used to fool the parser into thinking they were always surrounded by parentheses.  This permitted some surprising constructions such as foreach $x qw(a b c) {...}, which should really be written foreach $x (qw(a b c)) {...}.  These would sometimes get the lexer into the wrong state, so they didn't fully work, and the similar foreach qw(a b c) {...} that one might expect to be permitted never worked at all.

This side effect of qw has now been abolished.  It has been deprecated since Perl v5.13.11.  It is now necessary to use real parentheses everywhere that the grammar calls for them.

Turning on any lexical warnings used first to disable all default warnings if lexical warnings were not already enabled:

    $*; # deprecation warning
    use warnings "void";
    $#; # void warning; no deprecation warning

Now, the debugging, deprecated, glob, inplace and malloc warnings categories are left on when turning on lexical warnings (unless they are turned off by no warnings, of course).

This may cause deprecation warnings to occur in code that used to be free of warnings.

Those are the only categories consisting only of default warnings.  Default warnings in other categories are still disabled by use warnings "category", as we do not yet have the infrastructure for controlling individual warnings.

Due to an accident of history, state sub and our sub were equivalent to a plain sub, so one could even create an anonymous sub with our sub { ... }.  These are now disallowed outside of the "lexical_subs" feature.  Under the "lexical_subs" feature they have new meanings described in "Lexical Subroutines" in perlsub.

A value stored in an environment variable has always been stringified when inherited by child processes.

In this release, when assigning to %ENV, values are immediately stringified, and converted to be only a byte string.

First, it is forced to be only a string.  Then if the string is utf8 and the equivalent of utf8::downgrade() works, that result is used; otherwise, the equivalent of utf8::encode() is used, and a warning is issued about wide characters ("Diagnostics").

When require encounters an unreadable file, it now dies.  It used to ignore the file and continue searching the directories in @INC [perl #113422].

The various gv_fetchmeth_* XS functions used to treat a package whose named ended with ::SUPER specially.  A method lookup on the Foo::SUPER package would be treated as a SUPER method lookup on the Foo package.  This is no longer the case.  To do a SUPER lookup, pass the Foo stash and the GV_SUPER flag.

After some changes earlier in v5.17, split's behavior has been simplified: if the PATTERN argument evaluates to a string containing one space, it is treated the way that a literal string containing one space once was.

The following modules will be removed from the core distribution in a future release, and will at that time need to be installed from CPAN. Distributions on CPAN which require these modules will need to list them as prerequisites.

The core versions of these modules will now issue "deprecated"-category warnings to alert you to this fact. To silence these deprecation warnings, install the modules in question from CPAN.

Note that these are (with rare exceptions) fine modules that you are encouraged to continue to use. Their disinclusion from core primarily hinges on their necessity to bootstrapping a fully functional, CPAN-capable Perl installation, not usually on concerns over their design.

encoding

The use of this pragma is now strongly discouraged. It conflates the encoding of source text with the encoding of I/O data, reinterprets escape sequences in source text (a questionable choice), and introduces the UTF-8 bug to all runtime handling of character strings. It is broken as designed and beyond repair.

For using non-ASCII literal characters in source text, please refer to utf8. For dealing with textual I/O data, please refer to Encode and open.

Archive::Extract

B::Lint

B::Lint::Debug

CPANPLUS and all included CPANPLUS::* modules

Devel::InnerPackage

Log::Message

Log::Message::Config

Log::Message::Handlers

Log::Message::Item

Log::Message::Simple

Module::Pluggable

Module::Pluggable::Object

Object::Accessor

Pod::LaTeX

Term::UI

Term::UI::History

The following utilities will be removed from the core distribution in a future release as their associated modules have been deprecated. They will remain available with the applicable CPAN distribution.

cpanp

cpanp-run-perl

cpan2dist

These items are part of the CPANPLUS distribution.

pod2latex

This item is part of the Pod::LaTeX distribution.

This interpreter-global variable used to track the total number of Perl objects in the interpreter. It is no longer maintained and will be removed altogether in Perl v5.20.

When a regular expression pattern is compiled with /x, Perl treats 6 characters as white space to ignore, such as SPACE and TAB.  However, Unicode recommends 11 characters be treated thusly.  We will conform with this in a future Perl version.  In the meantime, use of any of the missing characters will raise a deprecation warning, unless turned off. The five characters are:

    U+0085 NEXT LINE
    U+200E LEFT-TO-RIGHT MARK
    U+200F RIGHT-TO-LEFT MARK
    U+2028 LINE SEPARATOR
    U+2029 PARAGRAPH SEPARATOR

A user-defined character name with trailing or multiple spaces in a row is likely a typo.  This now generates a warning when defined, on the assumption that uses of it will be unlikely to include the excess whitespace.

All the functions used to classify characters will be removed from a future version of Perl, and should not be used.  With participating C compilers (e.g., gcc), compiling any file that uses any of these will generate a warning.  These were not intended for public use; there are equivalent, faster, macros for most of them.

See "Character classes" in perlapi.  The complete list is:

is_uni_alnum, is_uni_alnumc, is_uni_alnumc_lc, is_uni_alnum_lc, is_uni_alpha, is_uni_alpha_lc, is_uni_ascii, is_uni_ascii_lc, is_uni_blank, is_uni_blank_lc, is_uni_cntrl, is_uni_cntrl_lc, is_uni_digit, is_uni_digit_lc, is_uni_graph, is_uni_graph_lc, is_uni_idfirst, is_uni_idfirst_lc, is_uni_lower, is_uni_lower_lc, is_uni_print, is_uni_print_lc, is_uni_punct, is_uni_punct_lc, is_uni_space, is_uni_space_lc, is_uni_upper, is_uni_upper_lc, is_uni_xdigit, is_uni_xdigit_lc, is_utf8_alnum, is_utf8_alnumc, is_utf8_alpha, is_utf8_ascii, is_utf8_blank, is_utf8_char, is_utf8_cntrl, is_utf8_digit, is_utf8_graph, is_utf8_idcont, is_utf8_idfirst, is_utf8_lower, is_utf8_mark, is_utf8_perl_space, is_utf8_perl_word, is_utf8_posix_digit, is_utf8_print, is_utf8_punct, is_utf8_space, is_utf8_upper, is_utf8_xdigit, is_utf8_xidcont, is_utf8_xidfirst.

In addition these three functions that have never worked properly are deprecated: to_uni_lower_lc, to_uni_title_lc, and to_uni_upper_lc.

There are three pairs of characters that Perl recognizes as metacharacters in regular expression patterns: {}, [], and (). These can be used as well to delimit patterns, as in:

  m{foo}
  s(foo)(bar)

Since they are metacharacters, they have special meaning to regular expression patterns, and it turns out that you can't turn off that special meaning by the normal means of preceding them with a backslash, if you use them, paired, within a pattern delimited by them.  For example, in

  m{foo\{1,3\}}

the backslashes do not change the behavior, and this matches "f o" followed by one to three more occurrences of "o".

Usages like this, where they are interpreted as metacharacters, are exceedingly rare; we think there are none, for example, in all of CPAN. Hence, this deprecation should affect very little code.  It does give notice, however, that any such code needs to change, which will in turn allow us to change the behavior in future Perl versions so that the backslashes do have an effect, and without fear that we are silently breaking any existing code.

A deprecation warning is now raised if the ( and ? are separated by white space or comments in (?...) regular expression constructs. Similarly, if the ( and * are separated in (*VERB...) constructs.

In theory, you can currently build perl without PerlIO.  Instead, you'd use a wrapper around stdio or sfio.  In practice, this isn't very useful.  It's not well tested, and without any support for IO layers or (thus) Unicode, it's not much of a perl.  Building without PerlIO will most likely be removed in the next version of perl.

PerlIO supports a stdio layer if stdio use is desired.  Similarly a sfio layer could be produced in the future, if needed.

• Config::Perl::V version 0.16 has been added as a dual-lifed module. It provides structured data retrieval of perl -V output including information only known to the perl binary and not available via Config.

For a complete list of updates, run:

  $ corelist --diff 5.16.0 5.18.0

You can substitute your favorite version in place of 5.16.0, too.

• Archive::Extract has been upgraded to 0.68.

  Work around an edge case on Linux with Busybox's unzip.

• Archive::Tar has been upgraded to 1.90.

  ptar now supports the -T option as well as dashless options [rt.cpan.org #75473], [rt.cpan.org #75475].

  Auto-encode filenames marked as UTF-8 [rt.cpan.org #75474].

  Don't use tell on IO::Zlib handles [rt.cpan.org #64339].

  Don't try to chown on symlinks.

• autodie has been upgraded to 2.13.

  autodie now plays nicely with the 'open' pragma.

• B has been upgraded to 1.42.

  The stashoff method of COPs has been added.   This provides access to an internal field added in perl 5.16 under threaded builds [perl #113034].

  B::COP::stashpv now supports UTF-8 package names and embedded NULs.

  All CVf_* and GVf_* and more SV-related flag values are now provided as constants in the B:: namespace and available for export.  The default export list has not changed.

  This makes the module work with the new pad API.

• B::Concise has been upgraded to 0.95.

  The -nobanner option has been fixed, and formats can now be dumped. When passed a sub name to dump, it will check also to see whether it is the name of a format.  If a sub and a format share the same name, it will dump both.

  This adds support for the new OpMAYBE_TRUEBOOL and OPpTRUEBOOL flags.

• B::Debug has been upgraded to 1.18.

  This adds support (experimentally) for B::PADLIST, which was added in Perl 5.17.4.

• B::Deparse has been upgraded to 1.20.

  Avoid warning when run under perl -w.

  It now deparses loop controls with the correct precedence, and multiple statements in a format line are also now deparsed correctly.

  This release suppresses trailing semicolons in formats.

  This release adds stub deparsing for lexical subroutines.

  It no longer dies when deparsing sort without arguments.  It now correctly omits the comma for system $prog @args and exec $prog @args.

• bignum, bigint and bigrat have been upgraded to 0.33.

  The overrides for hex and oct have been rewritten, eliminating several problems, and making one incompatible change:

  • Formerly, whichever of use bigint or use bigrat was compiled later would take precedence over the other, causing hex and oct not to respect the other pragma when in scope.
  • Using any of these three pragmata would cause hex and oct anywhere else in the program to evaluate their arguments in list context and prevent them from inferring $_ when called without arguments.
  • Using any of these three pragmata would make oct("1234") return 1234 (for any number not beginning with 0) anywhere in the program.  Now "1234" is translated from octal to decimal, whether within the pragma's scope or not.
  • The global overrides that facilitate lexical use of hex and oct now respect any existing overrides that were in place before the new overrides were installed, falling back to them outside of the scope of use bignum.
  • use bignum "hex", use bignum "oct" and similar invocations for bigint and bigrat now export a hex or oct function, instead of providing a global override.

• Carp has been upgraded to 1.29.

  Carp is no longer confused when caller returns undef for a package that has been deleted.

  The longmess() and shortmess() functions are now documented.

• CGI has been upgraded to 3.63.

  Unrecognized HTML escape sequences are now handled better, problematic trailing newlines are no longer inserted after <form> tags by startform() or start_form(), and bogus "Insecure Dependency" warnings appearing with some versions of perl are now worked around.

• Class::Struct has been upgraded to 0.64.

  The constructor now respects overridden accessor methods [perl #29230].

• Compress::Raw::Bzip2 has been upgraded to 2.060.

  The misuse of Perl's "magic" API has been fixed.

• Compress::Raw::Zlib has been upgraded to 2.060.

  Upgrade bundled zlib to version 1.2.7.

  Fix build failures on Irix, Solaris, and Win32, and also when building as C++ [rt.cpan.org #69985], [rt.cpan.org #77030], [rt.cpan.org #75222].

  The misuse of Perl's "magic" API has been fixed.

  compress(), uncompress(), memGzip() and memGunzip() have been speeded up by making parameter validation more efficient.

• CPAN::Meta::Requirements has been upgraded to 2.122.

  Treat undef requirements to from_string_hash as 0 (with a warning).

  Added requirements_for_module method.

• CPANPLUS has been upgraded to 0.9135.

  Allow adding blib/script to PATH.

  Save the history between invocations of the shell.

  Handle multiple makemakerargs and makeflags arguments better.

  This resolves issues with the SQLite source engine.

• Data::Dumper has been upgraded to 2.145.

  It has been optimized to only build a seen-scalar hash as necessary, thereby speeding up serialization drastically.

  Additional tests were added in order to improve statement, branch, condition and subroutine coverage.  On the basis of the coverage analysis, some of the internals of Dumper.pm were refactored.  Almost all methods are now documented.

• DB_File has been upgraded to 1.827.

  The main Perl module no longer uses the "@_" construct.

• Devel::Peek has been upgraded to 1.11.

  This fixes compilation with C++ compilers and makes the module work with the new pad API.

• Digest::MD5 has been upgraded to 2.52.

  Fix Digest::Perl::MD5 OO fallback [rt.cpan.org #66634].

• Digest::SHA has been upgraded to 5.84.

  This fixes a double-free bug, which might have caused vulnerabilities in some cases.

• DynaLoader has been upgraded to 1.18.

  This is due to a minor code change in the XS for the VMS implementation.

  This fixes warnings about using CODE sections without an OUTPUT section.

• Encode has been upgraded to 2.49.

  The Mac alias x-mac-ce has been added, and various bugs have been fixed in Encode::Unicode, Encode::UTF7 and Encode::GSM0338.

• Env has been upgraded to 1.04.

  Its SPLICE implementation no longer misbehaves in list context.

• ExtUtils::CBuilder has been upgraded to 0.280210.

  Manifest files are now correctly embedded for those versions of VC++ which make use of them. [perl #111782, #111798].

  A list of symbols to export can now be passed to link() when on Windows, as on other OSes [perl #115100].

• ExtUtils::ParseXS has been upgraded to 3.18.

  The generated C code now avoids unnecessarily incrementing PL_amagic_generation on Perl versions where it's done automatically (or on current Perl where the variable no longer exists).

  This avoids a bogus warning for initialised XSUB non-parameters [perl #112776].

• File::Copy has been upgraded to 2.26.

  copy() no longer zeros files when copying into the same directory, and also now fails (as it has long been documented to do) when attempting to copy a file over itself.

• File::DosGlob has been upgraded to 1.10.

  The internal cache of file names that it keeps for each caller is now freed when that caller is freed.  This means use File::DosGlob 'glob'; eval 'scalar <*>' no longer leaks memory.

• File::Fetch has been upgraded to 0.38.

  Added the 'file_default' option for URLs that do not have a file component.

  Use File::HomeDir when available, and provide PERL5_CPANPLUS_HOME to override the autodetection.

  Always re-fetch CHECKSUMS if fetchdir is set.

• File::Find has been upgraded to 1.23.

  This fixes inconsistent unixy path handling on VMS.

  Individual files may now appear in list of directories to be searched [perl #59750].

• File::Glob has been upgraded to 1.20.

  File::Glob has had exactly the same fix as File::DosGlob.  Since it is what Perl's own glob operator itself uses (except on VMS), this means eval 'scalar <*>' no longer leaks.

  A space-separated list of patterns return long lists of results no longer results in memory corruption or crashes.  This bug was introduced in Perl 5.16.0.  [perl #114984]

• File::Spec::Unix has been upgraded to 3.40.

  abs2rel could produce incorrect results when given two relative paths or the root directory twice [perl #111510].

• File::stat has been upgraded to 1.07.

  File::stat ignores the filetest pragma, and warns when used in combination therewith.  But it was not warning for -r.  This has been fixed [perl #111640].

  -p now works, and does not return false for pipes [perl #111638].

  Previously File::stat's overloaded -x and -X operators did not give the correct results for directories or executable files when running as root. They had been treating executable permissions for root just like for any other user, performing group membership tests etc for files not owned by root. They now follow the correct Unix behaviour - for a directory they are always true, and for a file if any of the three execute permission bits are set then they report that root can execute the file. Perl's builtin -x and -X operators have always been correct.

• File::Temp has been upgraded to 0.23

  Fixes various bugs involving directory removal.  Defers unlinking tempfiles if the initial unlink fails, which fixes problems on NFS.

• GDBM_File has been upgraded to 1.15.

  The undocumented optional fifth parameter to TIEHASH has been removed. This was intended to provide control of the callback used by gdbm* functions in case of fatal errors (such as filesystem problems), but did not work (and could never have worked). No code on CPAN even attempted to use it. The callback is now always the previous default, croak. Problems on some platforms with how the C croak function is called have also been resolved.

• Hash::Util has been upgraded to 0.15.

  hash_unlocked and hashref_unlocked now returns true if the hash is unlocked, instead of always returning false [perl #112126].

  hash_unlocked, hashref_unlocked, lock_hash_recurse and unlock_hash_recurse are now exportable [perl #112126].

  Two new functions, hash_locked and hashref_locked, have been added. Oddly enough, these two functions were already exported, even though they did not exist [perl #112126].

• HTTP::Tiny has been upgraded to 0.025.

  Add SSL verification features [github #6], [github #9].

  Include the final URL in the response hashref.

  Add local_address option.

  This improves SSL support.

• IO has been upgraded to 1.28.

  sync() can now be called on read-only file handles [perl #64772].

  IO::Socket tries harder to cache or otherwise fetch socket information.

• IPC::Cmd has been upgraded to 0.80.

  Use POSIX::_exit instead of exit in run_forked [rt.cpan.org #76901].

• IPC::Open3 has been upgraded to 1.13.

  The open3() function no longer uses POSIX::close() to close file descriptors since that breaks the ref-counting of file descriptors done by PerlIO in cases where the file descriptors are shared by PerlIO streams, leading to attempts to close the file descriptors a second time when any such PerlIO streams are closed later on.

• Locale::Codes has been upgraded to 3.25.

  It includes some new codes.

• Memoize has been upgraded to 1.03.

  Fix the MERGE cache option.

• Module::Build has been upgraded to 0.4003.

  Fixed bug where modules without $VERSION might have a version of '0' listed in 'provides' metadata, which will be rejected by PAUSE.

  Fixed bug in PodParser to allow numerals in module names.

  Fixed bug where giving arguments twice led to them becoming arrays, resulting in install paths like ARRAY(0xdeadbeef)/lib/Foo.pm.

  A minor bug fix allows markup to be used around the leading "Name" in a POD "abstract" line, and some documentation improvements have been made.

• Module::CoreList has been upgraded to 2.90

  Version information is now stored as a delta, which greatly reduces the size of the CoreList.pm file.

  This restores compatibility with older versions of perl and cleans up the corelist data for various modules.

• Module::Load::Conditional has been upgraded to 0.54.

  Fix use of requires on perls installed to a path with spaces.

  Various enhancements include the new use of Module::Metadata.

• Module::Metadata has been upgraded to 1.000011.

  The creation of a Module::Metadata object for a typical module file has been sped up by about 40%, and some spurious warnings about $VERSIONs have been suppressed.

• Module::Pluggable has been upgraded to 4.7.

  Amongst other changes, triggers are now allowed on events, which gives a powerful way to modify behaviour.

• Net::Ping has been upgraded to 2.41.

  This fixes some test failures on Windows.

• Opcode has been upgraded to 1.25.

  Reflect the removal of the boolkeys opcode and the addition of the clonecv, introcv and padcv opcodes.

• overload has been upgraded to 1.22.

  no overload now warns for invalid arguments, just like use overload.

• PerlIO::encoding has been upgraded to 0.16.

  This is the module implementing the ":encoding(...)" I/O layer.  It no longer corrupts memory or crashes when the encoding back-end reallocates the buffer or gives it a typeglob or shared hash key scalar.

• PerlIO::scalar has been upgraded to 0.16.

  The buffer scalar supplied may now only contain code points 0xFF or lower. [perl #109828]

• Perl::OSType has been upgraded to 1.003.

  This fixes a bug detecting the VOS operating system.

• Pod::Html has been upgraded to 1.18.

  The option --libpods has been reinstated. It is deprecated, and its use does nothing other than issue a warning that it is no longer supported.

  Since the HTML files generated by pod2html claim to have a UTF-8 charset, actually write the files out using UTF-8 [perl #111446].

• Pod::Simple has been upgraded to 3.28.

  Numerous improvements have been made, mostly to Pod::Simple::XHTML, which also has a compatibility change: the codes_in_verbatim option is now disabled by default.  See cpan/Pod-Simple/ChangeLog for the full details.

• re has been upgraded to 0.23

  Single character [class]es like /[s]/ or /[s]/i are now optimized as if they did not have the brackets, i.e. /s/ or /s/i.

  See note about op_comp in the "Internal Changes" section below.

• Safe has been upgraded to 2.35.

  Fix interactions with Devel::Cover.

  Don't eval code under no strict.

• Scalar::Util has been upgraded to version 1.27.

  Fix an overloading issue with sum.

  first and reduce now check the callback first (so &first(1) is disallowed).

  Fix tainted on magical values [rt.cpan.org #55763].

  Fix sum on previously magical values [rt.cpan.org #61118].

  Fix reading past the end of a fixed buffer [rt.cpan.org #72700].

• Search::Dict has been upgraded to 1.07.

  No longer require stat on filehandles.

  Use fc for casefolding.

• Socket has been upgraded to 2.009.

  Constants and functions required for IP multicast source group membership have been added.

  unpack_sockaddr_in() and unpack_sockaddr_in6() now return just the IP address in scalar context, and inet_ntop() now guards against incorrect length scalars being passed in.

  This fixes an uninitialized memory read.

• Storable has been upgraded to 2.41.

  Modifying $_[0] within STORABLE_freeze no longer results in crashes [perl #112358].

  An object whose class implements STORABLE_attach is now thawed only once when there are multiple references to it in the structure being thawed [perl #111918].

  Restricted hashes were not always thawed correctly [perl #73972].

  Storable would croak when freezing a blessed REF object with a STORABLE_freeze() method [perl #113880].

  It can now freeze and thaw vstrings correctly.  This causes a slight incompatible change in the storage format, so the format version has increased to 2.9.

  This contains various bugfixes, including compatibility fixes for older versions of Perl and vstring handling.

• Sys::Syslog has been upgraded to 0.32.

  This contains several bug fixes relating to getservbyname(), setlogsock()and log levels in syslog(), together with fixes for Windows, Haiku-OS and GNU/kFreeBSD.  See cpan/Sys-Syslog/Changes for the full details.

• Term::ANSIColor has been upgraded to 4.02.

  Add support for italics.

  Improve error handling.

• Term::ReadLine has been upgraded to 1.10.  This fixes the use of the cpan and cpanp shells on Windows in the event that the current drive happens to contain a \dev\tty file.

• Test::Harness has been upgraded to 3.26.

  Fix glob semantics on Win32 [rt.cpan.org #49732].

  Don't use Win32::GetShortPathName when calling perl [rt.cpan.org #47890].

  Ignore -T when reading shebang [rt.cpan.org #64404].

  Handle the case where we don't know the wait status of the test more gracefully.

  Make the test summary 'ok' line overridable so that it can be changed to a plugin to make the output of prove idempotent.

  Don't run world-writable files.

• Text::Tabs and Text::Wrap have been upgraded to 2012.0818.  Support for Unicode combining characters has been added to them both.

• threads::shared has been upgraded to 1.31.

  This adds the option to warn about or ignore attempts to clone structures that can't be cloned, as opposed to just unconditionally dying in that case.

  This adds support for dual-valued values as created by Scalar::Util::dualvar.

• Tie::StdHandle has been upgraded to 4.3.

  READ now respects the offset argument to read [perl #112826].

• Time::Local has been upgraded to 1.2300.

  Seconds values greater than 59 but less than 60 no longer cause timegm() and timelocal() to croak.

• Unicode::UCD has been upgraded to 0.53.

  This adds a function all_casefolds() that returns all the casefolds.

• Win32 has been upgraded to 0.47.

  New APIs have been added for getting and setting the current code page.

• Version::Requirements has been removed from the core distribution.  It is available under a different name: CPAN::Meta::Requirements.

perlcheat

• perlcheat has been reorganized, and a few new sections were added.

perldata

• Now explicitly documents the behaviour of hash initializer lists that contain duplicate keys.

perldiag

• The explanation of symbolic references being prevented by "strict refs" now doesn't assume that the reader knows what symbolic references are.

perlfaq

• perlfaq has been synchronized with version 5.0150040 from CPAN.

perlfunc

• The return value of pipe is now documented.
• Clarified documentation of our.

perlop

• Loop control verbs (dump, goto, next, last and redo) have always had the same precedence as assignment operators, but this was not documented until now.

Diagnostics

The following additions or changes have been made to diagnostic output, including warnings and fatal error messages.  For the complete list of diagnostic messages, see perldiag.

New Errors

• Unterminated delimiter for here document

  This message now occurs when a here document label has an initial quotation mark but the final quotation mark is missing.

  This replaces a bogus and misleading error message about not finding the label itself [perl #114104].

• panic: child pseudo-process was never scheduled

  This error is thrown when a child pseudo-process in the ithreads implementation on Windows was not scheduled within the time period allowed and therefore was not able to initialize properly [perl #88840].

• Group name must start with a non-digit word character in regex; marked by <-- HERE in m/%s/

  This error has been added for (?&0), which is invalid.  It used to produce an incomprehensible error message [perl #101666].

• Can't use an undefined value as a subroutine reference

  Calling an undefined value as a subroutine now produces this error message. It used to, but was accidentally disabled, first in Perl 5.004 for non-magical variables, and then in Perl v5.14 for magical (e.g., tied) variables.  It has now been restored.  In the mean time, undef was treated as an empty string [perl #113576].

• Experimental "%s" subs not enabled

  To use lexical subs, you must first enable them:

      no warnings 'experimental::lexical_subs';
      use feature 'lexical_subs';
      my sub foo { ... }

New Warnings

• 'Strings with code points over 0xFF may not be mapped into in-memory file handles'
• '%s' resolved to '\o{%s}%d'
• 'Trailing white-space in a charnames alias definition is deprecated'
• 'A sequence of multiple spaces in a charnames alias definition is deprecated'
• 'Passing malformed UTF-8 to "%s" is deprecated'

• Subroutine "&%s" is not available

  (W closure) During compilation, an inner named subroutine or eval is attempting to capture an outer lexical subroutine that is not currently available.  This can happen for one of two reasons.  First, the lexical subroutine may be declared in an outer anonymous subroutine that has not yet been created.  (Remember that named subs are created at compile time, while anonymous subs are created at run-time.)  For example,

      sub { my sub a {...} sub f { \&a } }

  At the time that f is created, it can't capture the current the "a" sub, since the anonymous subroutine hasn't been created yet.  Conversely, the following won't give a warning since the anonymous subroutine has by now been created and is live:

      sub { my sub a {...} eval 'sub f { \&a }' }->();

  The second situation is caused by an eval accessing a variable that has gone out of scope, for example,

      sub f {
          my sub a {...}
          sub { eval '\&a' }
      }
      f()->();

  Here, when the '\&a' in the eval is being compiled, f() is not currently being executed, so its &a is not available for capture.

• "%s" subroutine &%s masks earlier declaration in same %s

  (W misc) A "my" or "state" subroutine has been redeclared in the current scope or statement, effectively eliminating all access to the previous instance.  This is almost always a typographical error. Note that the earlier subroutine will still exist until the end of the scope or until all closure references to it are destroyed.

• The %s feature is experimental

  (S experimental) This warning is emitted if you enable an experimental feature via use feature.  Simply suppress the warning if you want to use the feature, but know that in doing so you are taking the risk of using an experimental feature which may change or be removed in a future Perl version:

      no warnings "experimental::lexical_subs";
      use feature "lexical_subs";

• sleep(%u) too large

  (W overflow) You called sleep with a number that was larger than it can reliably handle and sleep probably slept for less time than requested.

• Wide character in setenv

  Attempts to put wide characters into environment variables via %ENV now provoke this warning.

• "Invalid negative number (%s) in chr"

  chr() now warns when passed a negative value [perl #83048].

• "Integer overflow in srand"

  srand() now warns when passed a value that doesn't fit in a UV (since the value will be truncated rather than overflowing) [perl #40605].

• "-i used with no filenames on the command line, reading from STDIN"

  Running perl with the -i flag now warns if no input files are provided on the command line [perl #113410].

• $* is no longer supported

  The warning that use of $* and $# is no longer supported is now generated for every location that references them.  Previously it would fail to be generated if another variable using the same typeglob was seen first (e.g. @* before $*), and would not be generated for the second and subsequent uses.  (It's hard to fix the failure to generate warnings at all without also generating them every time, and warning every time is consistent with the warnings that $[ used to generate.)

• The warnings for \b{ and \B{ were added.  They are a deprecation warning which should be turned off by that category.  One should not have to turn off regular regexp warnings as well to get rid of these.

• Constant(%s): Call to &{$^H{%s}} did not return a defined value

  Constant overloading that returns undef results in this error message. For numeric constants, it used to say "Constant(undef)".  "undef" has been replaced with the number itself.

• The error produced when a module cannot be loaded now includes a hint that the module may need to be installed: "Can't locate hopping.pm in @INC (you may need to install the hopping module) (@INC contains: ...)"

• vector argument not supported with alpha versions

  This warning was not suppressible, even with no warnings.  Now it is suppressible, and has been moved from the "internal" category to the "printf" category.

• Can't do {n,m} with n > m in regex; marked by <-- HERE in m/%s/

  This fatal error has been turned into a warning that reads:

  Quantifier {n,m} with n > m can't match in regex

  (W regexp) Minima should be less than or equal to maxima.  If you really want your regexp to match something 0 times, just put {0}.

• The "Runaway prototype" warning that occurs in bizarre cases has been removed as being unhelpful and inconsistent.
• The "Not a format reference" error has been removed, as the only case in which it could be triggered was a bug.
• The "Unable to create sub named %s" error has been removed for the same reason.
• The 'Can't use "my %s" in sort comparison' error has been downgraded to a warning, '"my %s" used in sort comparison' (with 'state' instead of 'my' for state variables).  In addition, the heuristics for guessing whether lexical $a or $b has been misused have been improved to generate fewer false positives.  Lexical $a and $b are no longer disallowed if they are outside the sort block.  Also, a named unary or list operator inside the sort block no longer causes the $a or $b to be ignored [perl #86136].

BeOS

BeOS was an operating system for personal computers developed by Be Inc, initially for their BeBox hardware. The OS Haiku was written as an open source replacement for/continuation of BeOS, and its perl port is current and actively maintained.

UTS Global

Support code relating to UTS global has been removed.  UTS was a mainframe version of System V created by Amdahl, subsequently sold to UTS Global.  The port has not been touched since before Perl v5.8.0, and UTS Global is now defunct.

VM/ESA

Support for VM/ESA has been removed. The port was tested on 2.3.0, which IBM ended service on in March 2002. 2.4.0 ended service in June 2003, and was superseded by Z/VM. The current version of Z/VM is V6.2.0, and scheduled for end of service on 2015/04/30.

MPE/IX

Support for MPE/IX has been removed.

EPOC

Support code relating to EPOC has been removed.  EPOC was a family of operating systems developed by Psion for mobile devices.  It was the predecessor of Symbian.  The port was last updated in April 2002.

Rhapsody

Support for Rhapsody has been removed.

AIX

Configure now always adds -qlanglvl=extc99 to the CC flags on AIX when using xlC.  This will make it easier to compile a number of XS-based modules that assume C99 [perl #113778].

clang++

There is now a workaround for a compiler bug that prevented compiling with clang++ since Perl v5.15.7 [perl #112786].

C++

When compiling the Perl core as C++ (which is only semi-supported), the mathom functions are now compiled as extern "C", to ensure proper binary compatibility.  (However, binary compatibility isn't generally guaranteed anyway in the situations where this would matter.)

Darwin

Stop hardcoding an alignment on 8 byte boundaries to fix builds using -Dusemorebits.

Haiku

Perl should now work out of the box on Haiku R1 Alpha 4.

MidnightBSD

libc_r was removed from recent versions of MidnightBSD and older versions work better with pthread. Threading is now enabled using pthread which corrects build errors with threading enabled on 0.4-CURRENT.

Solaris

In Configure, avoid running sed commands with flags not supported on Solaris.

VMS

• Where possible, the case of filenames and command-line arguments is now preserved by enabling the CRTL features DECC$EFS_CASE_PRESERVE and DECC$ARGV_PARSE_STYLE at start-up time.  The latter only takes effect when extended parse is enabled in the process from which Perl is run.
• The character set for Extended Filename Syntax (EFS) is now enabled by default on VMS.  Among other things, this provides better handling of dots in directory names, multiple dots in filenames, and spaces in filenames.  To obtain the old behavior, set the logical name DECC$EFS_CHARSET to DISABLE.
• Fixed linking on builds configured with -Dusemymalloc=y.
• Experimental support for building Perl with the HP C++ compiler is available by configuring with -Dusecxx.
• All C header files from the top-level directory of the distribution are now installed on VMS, providing consistency with a long-standing practice on other platforms. Previously only a subset were installed, which broke non-core extension builds for extensions that depended on the missing include files.
• Quotes are now removed from the command verb (but not the parameters) for commands spawned via system, backticks, or a piped open.  Previously, quotes on the verb were passed through to DCL, which would fail to recognize the command.  Also, if the verb is actually a path to an image or command procedure on an ODS-5 volume, quoting it now allows the path to contain spaces.
• The a2p build has been fixed for the HP C++ compiler on OpenVMS.

Win32

• Perl can now be built using Microsoft's Visual C++ 2012 compiler by specifying CCTYPE=MSVC110 (or MSVC110FREE if you are using the free Express edition for Windows Desktop) in win32/Makefile.
• The option to build without USE_SOCKETS_AS_HANDLES has been removed.
• Fixed a problem where perl could crash while cleaning up threads (including the main thread) in threaded debugging builds on Win32 and possibly other platforms [perl #114496].
• A rare race condition that would lead to sleep taking more time than requested, and possibly even hanging, has been fixed [perl #33096].

• link on Win32 now attempts to set $! to more appropriate values based on the Win32 API error code. [perl #112272]

  Perl no longer mangles the environment block, e.g. when launching a new sub-process, when the environment contains non-ASCII characters. Known problems still remain, however, when the environment contains characters outside of the current ANSI codepage (e.g. see the item about Unicode in %ENV in <http://perl5.git.perl.org/perl.git/blob/HEAD:/Porting/todo.pod>). [perl #113536]

• Building perl with some Windows compilers used to fail due to a problem with miniperl's glob operator (which uses the perlglob program) deleting the PATH environment variable [perl #113798].

• A new makefile option, USE_64_BIT_INT, has been added to the Windows makefiles.  Set this to "define" when building a 32-bit perl if you want it to use 64-bit integers.

  Machine code size reductions, already made to the DLLs of XS modules in Perl v5.17.2, have now been extended to the perl DLL itself.

  Building with VC++ 6.0 was inadvertently broken in Perl v5.17.2 but has now been fixed again.

WinCE

Building on WinCE is now possible once again, although more work is required to fully restore a clean build.