{ "mode": "man", "parameter": "perl5120delta", "section": "1", "url": "https://www.chedong.com/phpMan.php/man/perl5120delta/1/json", "generated": "2026-08-22T03:40:40Z", "sections": { "NAME": { "content": "perl5120delta - what is new for perl v5.12.0\n", "subsections": [] }, "DESCRIPTION": { "content": "This document describes differences between the 5.10.0 release and the 5.12.0 release.\n\nMany of the bug fixes in 5.12.0 are already included in the 5.10.1 maintenance release.\n\nYou can see the list of those changes in the 5.10.1 release notes (perl5101delta).\n", "subsections": [] }, "Core Enhancements": { "content": "", "subsections": [ { "name": "New \"package NAME VERSION\" syntax", "content": "This new syntax allows a module author to set the $VERSION of a namespace when the namespace\nis declared with 'package'. It eliminates the need for \"our $VERSION = ...\" and similar\nconstructs. E.g.\n\npackage Foo::Bar 1.23;\n# $Foo::Bar::VERSION == 1.23\n\nThere are several advantages to this:\n\n• $VERSION is parsed in exactly the same way as \"use NAME VERSION\"\n\n• $VERSION is set at compile time\n\n• $VERSION is a version object that provides proper overloading of comparison operators so\ncomparing $VERSION to decimal (1.23) or dotted-decimal (v1.2.3) version numbers works\ncorrectly.\n\n• Eliminates \"$VERSION = ...\" and \"eval $VERSION\" clutter\n\n• As it requires VERSION to be a numeric literal or v-string literal, it can be statically\nparsed by toolchain modules without \"eval\" the way MM->parseversion does for \"$VERSION =\n...\"\n\nIt does not break old code with only \"package NAME\", but code that uses \"package NAME\nVERSION\" will need to be restricted to perl 5.12.0 or newer This is analogous to the change\nto \"open\" from two-args to three-args. Users requiring the latest Perl will benefit, and\nperhaps after several years, it will become a standard practice.\n\nHowever, \"package NAME VERSION\" requires a new, 'strict' version number format. See \"Version\nnumber formats\" for details.\n" }, { "name": "The \"...\" operator", "content": "A new operator, \"...\", nicknamed the Yada Yada operator, has been added. It is intended to\nmark placeholder code that is not yet implemented. See \"Yada Yada Operator\" in perlop.\n" }, { "name": "Implicit strictures", "content": "Using the \"use VERSION\" syntax with a version number greater or equal to 5.11.0 will\nlexically enable strictures just like \"use strict\" would do (in addition to enabling\nfeatures.) The following:\n\nuse 5.12.0;\n\nmeans:\n\nuse strict;\nuse feature ':5.12';\n" }, { "name": "Unicode improvements", "content": "Perl 5.12 comes with Unicode 5.2, the latest version available to us at the time of release.\nThis version of Unicode was released in October 2009. See\n for further details about what's changed in\nthis version of the standard. See perlunicode for instructions on installing and using other\nversions of Unicode.\n\nAdditionally, Perl's developers have significantly improved Perl's Unicode implementation.\nFor full details, see \"Unicode overhaul\" below.\n" }, { "name": "Y2038 compliance", "content": "Perl's core time-related functions are now Y2038 compliant. (It may not mean much to you, but\nyour kids will love it!)\n" }, { "name": "qr overloading", "content": "It is now possible to overload the \"qr//\" operator, that is, conversion to regexp, like it\nwas already possible to overload conversion to boolean, string or number of objects. It is\ninvoked when an object appears on the right hand side of the \"=~\" operator or when it is\ninterpolated into a regexp. See overload.\n" }, { "name": "Pluggable keywords", "content": "Extension modules can now cleanly hook into the Perl parser to define new kinds of keyword-\nheaded expression and compound statement. The syntax following the keyword is defined\nentirely by the extension. This allows a completely non-Perl sublanguage to be parsed inline,\nwith the correct ops cleanly generated.\n\nSee \"PLkeywordplugin\" in perlapi for the mechanism. The Perl core source distribution also\nincludes a new module XS::APItest::KeywordRPN, which implements reverse Polish notation\narithmetic via pluggable keywords. This module is mainly used for test purposes, and is not\nnormally installed, but also serves as an example of how to use the new mechanism.\n\nPerl's developers consider this feature to be experimental. We may remove it or change it in\na backwards-incompatible way in Perl 5.14.\n" }, { "name": "APIs for more internals", "content": "The lowest layers of the lexer and parts of the pad system now have C APIs available to XS\nextensions. These are necessary to support proper use of pluggable keywords, but have other\nuses too. The new APIs are experimental, and only cover a small proportion of what would be\nnecessary to take full advantage of the core's facilities in these areas. It is intended that\nthe Perl 5.13 development cycle will see the addition of a full range of clean, supported\ninterfaces.\n\nPerl's developers consider this feature to be experimental. We may remove it or change it in\na backwards-incompatible way in Perl 5.14.\n" }, { "name": "Overridable function lookup", "content": "Where an extension module hooks the creation of rv2cv ops to modify the subroutine lookup\nprocess, this now works correctly for bareword subroutine calls. This means that prototypes\non subroutines referenced this way will be processed correctly. (Previously bareword\nsubroutine names were initially looked up, for parsing purposes, by an unhookable mechanism,\nso extensions could only properly influence subroutine names that appeared with an \"&\"\nsigil.)\n" }, { "name": "A proper interface for pluggable Method Resolution Orders", "content": "As of Perl 5.12.0 there is a new interface for plugging and using method resolution orders\nother than the default linear depth first search. The C3 method resolution order added in\n5.10.0 has been re-implemented as a plugin, without changing its Perl-space interface. See\nperlmroapi for more information.\n" }, { "name": "\"\\N\" experimental regex escape", "content": "Perl now supports \"\\N\", a new regex escape which you can think of as the inverse of \"\\n\". It\nwill match any character that is not a newline, independently from the presence or absence of\nthe single line match modifier \"/s\". It is not usable within a character class. \"\\N{3}\"\nmeans to match 3 non-newlines; \"\\N{5,}\" means to match at least 5. \"\\N{NAME}\" still means\nthe character or sequence named \"NAME\", but \"NAME\" no longer can be things like 3, or \"5,\".\n\nThis will break a custom charnames translator which allows numbers for character names, as\n\"\\N{3}\" will now mean to match 3 non-newline characters, and not the character whose name is\n3. (No name defined by the Unicode standard is a number, so only custom translators might be\naffected.)\n\nPerl's developers are somewhat concerned about possible user confusion with the existing\n\"\\N{...}\" construct which matches characters by their Unicode name. Consequently, this\nfeature is experimental. We may remove it or change it in a backwards-incompatible way in\nPerl 5.14.\n" }, { "name": "DTrace support", "content": "Perl now has some support for DTrace. See \"DTrace support\" in INSTALL.\n" }, { "name": "Support for \"configure_requires\" in CPAN module metadata", "content": "Both \"CPAN\" and \"CPANPLUS\" now support the \"configurerequires\" keyword in the META.yml\nmetadata file included in most recent CPAN distributions. This allows distribution authors\nto specify configuration prerequisites that must be installed before running Makefile.PL or\nBuild.PL.\n\nSee the documentation for \"ExtUtils::MakeMaker\" or \"Module::Build\" for more on how to specify\n\"configurerequires\" when creating a distribution for CPAN.\n" }, { "name": "\"each\", \"keys\", \"values\" are now more flexible", "content": "The \"each\", \"keys\", \"values\" function can now operate on arrays.\n" }, { "name": "\"when\" as a statement modifier", "content": "\"when\" is now allowed to be used as a statement modifier.\n" }, { "name": "$, flexibility", "content": "The variable $, may now be tied.\n" }, { "name": "// in when clauses", "content": "// now behaves like || in when clauses\n" }, { "name": "Enabling warnings from your shell environment", "content": "You can now set \"-W\" from the \"PERL5OPT\" environment variable\n" }, { "name": "\"delete local\"", "content": "\"delete local\" now allows you to locally delete a hash entry.\n" }, { "name": "New support for Abstract namespace sockets", "content": "Abstract namespace sockets are Linux-specific socket type that live in AFUNIX family,\nslightly abusing it to be able to use arbitrary character arrays as addresses: They start\nwith nul byte and are not terminated by nul byte, but with the length passed to the socket()\nsystem call.\n" }, { "name": "32-bit limit on substr arguments removed", "content": "The 32-bit limit on \"substr\" arguments has now been removed. The full range of the system's\nsigned and unsigned integers is now available for the \"pos\" and \"len\" arguments.\n" } ] }, "Potentially Incompatible Changes": { "content": "", "subsections": [ { "name": "Deprecations warn by default", "content": "Over the years, Perl's developers have deprecated a number of language features for a variety\nof reasons. Perl now defaults to issuing a warning if a deprecated language feature is used.\nMany of the deprecations Perl now warns you about have been deprecated for many years. You\ncan find a list of what was deprecated in a given release of Perl in the \"perl5xxdelta.pod\"\nfile for that release.\n\nTo disable this feature in a given lexical scope, you should use \"no warnings 'deprecated';\"\nFor information about which language features are deprecated and explanations of various\ndeprecation warnings, please see perldiag. See \"Deprecations\" below for the list of features\nand modules Perl's developers have deprecated as part of this release.\n" }, { "name": "Version number formats", "content": "Acceptable version number formats have been formalized into \"strict\" and \"lax\" rules.\n\"package NAME VERSION\" takes a strict version number. \"UNIVERSAL::VERSION\" and the version\nobject constructors take lax version numbers. Providing an invalid version will result in a\nfatal error. The version argument in \"use NAME VERSION\" is first parsed as a numeric literal\nor v-string and then passed to \"UNIVERSAL::VERSION\" (and must then pass the \"lax\" format\ntest).\n\nThese formats are documented fully in the version module. To a first approximation, a\n\"strict\" version number is a positive decimal number (integer or decimal-fraction) without\nexponentiation or else a dotted-decimal v-string with a leading 'v' character and at least\nthree components. A \"lax\" version number allows v-strings with fewer than three components or\nwithout a leading 'v'. Under \"lax\" rules, both decimal and dotted-decimal versions may have a\ntrailing \"alpha\" component separated by an underscore character after a fractional or dotted-\ndecimal component.\n\nThe version module adds \"version::isstrict\" and \"version::islax\" functions to check a\nscalar against these rules.\n" }, { "name": "@INC reorganization", "content": "In @INC, \"ARCHLIB\" and \"PRIVLIB\" now occur after the current version's \"siteperl\" and\n\"vendorperl\". Modules installed into \"siteperl\" and \"vendorperl\" will now be loaded in\npreference to those installed in \"ARCHLIB\" and \"PRIVLIB\".\n" }, { "name": "REGEXPs are now first class", "content": "Internally, Perl now treats compiled regular expressions (such as those created with \"qr//\")\nas first class entities. Perl modules which serialize, deserialize or otherwise have deep\ninteraction with Perl's internal data structures need to be updated for this change. Most\naffected CPAN modules have already been updated as of this writing.\n" }, { "name": "Switch statement changes", "content": "The \"given\"/\"when\" switch statement handles complex statements better than Perl 5.10.0 did\n(These enhancements are also available in 5.10.1 and subsequent 5.10 releases.) There are two\nnew cases where \"when\" now interprets its argument as a boolean, instead of an expression to\nbe used in a smart match:\n\nflip-flop operators\nThe \"..\" and \"...\" flip-flop operators are now evaluated in boolean context, following\ntheir usual semantics; see \"Range Operators\" in perlop.\n\nNote that, as in perl 5.10.0, \"when (1..10)\" will not work to test whether a given value\nis an integer between 1 and 10; you should use \"when ([1..10])\" instead (note the array\nreference).\n\nHowever, contrary to 5.10.0, evaluating the flip-flop operators in boolean context\nensures it can now be useful in a when(), notably for implementing bistable conditions,\nlike in:\n\nwhen (/^=begin/ .. /^=end/) {\n# do something\n}\n\ndefined-or operator\nA compound expression involving the defined-or operator, as in \"when (expr1 // expr2)\",\nwill be treated as boolean if the first expression is boolean. (This just extends the\nexisting rule that applies to the regular or operator, as in \"when (expr1 || expr2)\".)\n" }, { "name": "Smart match changes", "content": "Since Perl 5.10.0, Perl's developers have made a number of changes to the smart match\noperator. These, of course, also alter the behaviour of the switch statements where smart\nmatching is implicitly used. These changes were also made for the 5.10.1 release, and will\nremain in subsequent 5.10 releases.\n\nChanges to type-based dispatch\n\nThe smart match operator \"~~\" is no longer commutative. The behaviour of a smart match now\ndepends primarily on the type of its right hand argument. Moreover, its semantics have been\nadjusted for greater consistency or usefulness in several cases. While the general backwards\ncompatibility is maintained, several changes must be noted:\n\n• Code references with an empty prototype are no longer treated specially. They are passed\nan argument like the other code references (even if they choose to ignore it).\n\n• \"%hash ~~ sub {}\" and \"@array ~~ sub {}\" now test that the subroutine returns a true\nvalue for each key of the hash (or element of the array), instead of passing the whole\nhash or array as a reference to the subroutine.\n\n• Due to the commutativity breakage, code references are no longer treated specially when\nappearing on the left of the \"~~\" operator, but like any vulgar scalar.\n\n• \"undef ~~ %hash\" is always false (since \"undef\" can't be a key in a hash). No implicit\nconversion to \"\" is done (as was the case in perl 5.10.0).\n\n• \"$scalar ~~ @array\" now always distributes the smart match across the elements of the\narray. It's true if one element in @array verifies \"$scalar ~~ $element\". This is a\ngeneralization of the old behaviour that tested whether the array contained the scalar.\n\nThe full dispatch table for the smart match operator is given in \"Smart matching in detail\"\nin perlsyn.\n\nSmart match and overloading\n\nAccording to the rule of dispatch based on the rightmost argument type, when an object\noverloading \"~~\" appears on the right side of the operator, the overload routine will always\nbe called (with a 3rd argument set to a true value, see overload.) However, when the object\nwill appear on the left, the overload routine will be called only when the rightmost argument\nis a simple scalar. This way, distributivity of smart match across arrays is not broken, as\nwell as the other behaviours with complex types (coderefs, hashes, regexes). Thus, writers of\noverloading routines for smart match mostly need to worry only with comparing against a\nscalar, and possibly with stringification overloading; the other common cases will be\nautomatically handled consistently.\n\n\"~~\" will now refuse to work on objects that do not overload it (in order to avoid relying on\nthe object's underlying structure). (However, if the object overloads the stringification or\nthe numification operators, and if overload fallback is active, it will be used instead, as\nusual.)\n" }, { "name": "Other potentially incompatible changes", "content": "• The definitions of a number of Unicode properties have changed to match those of the\ncurrent Unicode standard. These are listed above under \"Unicode overhaul\". This change\nmay break code that expects the old definitions.\n\n• The boolkeys op has moved to the group of hash ops. This breaks binary compatibility.\n\n• Filehandles are now always blessed into \"IO::File\".\n\nThe previous behaviour was to bless Filehandles into FileHandle (an empty proxy class) if\nit was loaded into memory and otherwise to bless them into \"IO::Handle\".\n\n• The semantics of \"use feature :5.10*\" have changed slightly. See \"Modules and Pragmata\"\nfor more information.\n\n• Perl's developers now use git, rather than Perforce. This should be a purely internal\nchange only relevant to people actively working on the core. However, you may see minor\ndifference in perl as a consequence of the change. For example in some of details of the\noutput of \"perl -V\". See perlrepository for more information.\n\n• As part of the \"Test::Harness\" 2.x to 3.x upgrade, the experimental\n\"Test::Harness::Straps\" module has been removed. See \"Modules and Pragmata\" for more\ndetails.\n\n• As part of the \"ExtUtils::MakeMaker\" upgrade, the \"ExtUtils::MakeMaker::bytes\" and\n\"ExtUtils::MakeMaker::vmsish\" modules have been removed from this distribution.\n\n• \"Module::CoreList\" no longer contains the %:patchlevel hash.\n\n• \"length undef\" now returns undef.\n\n• Unsupported private C API functions are now declared \"static\" to prevent leakage to\nPerl's public API.\n\n• To support the bootstrapping process, miniperl no longer builds with UTF-8 support in the\nregexp engine.\n\nThis allows a build to complete with PERLUNICODE set and a UTF-8 locale. Without this\nthere's a bootstrapping problem, as miniperl can't load the UTF-8 components of the\nregexp engine, because they're not yet built.\n\n• miniperl's @INC is now restricted to just \"-I...\", the split of $ENV{PERL5LIB}, and \"\".\"\"\n\n• A space or a newline is now required after a \"#line XXX\" directive.\n\n• Tied filehandles now have an additional method EOF which provides the EOF type.\n\n• To better match all other flow control statements, \"foreach\" may no longer be used as an\nattribute.\n\n• Perl's command-line switch \"-P\", which was deprecated in version 5.10.0, has now been\nremoved. The CPAN module \"Filter::cpp\" can be used as an alternative.\n" } ] }, "Deprecations": { "content": "From time to time, Perl's developers find it necessary to deprecate features or modules we've\npreviously shipped as part of the core distribution. We are well aware of the pain and\nfrustration that a backwards-incompatible change to Perl can cause for developers building or\nmaintaining software in Perl. You can be sure that when we deprecate a functionality or\nsyntax, it isn't a choice we make lightly. Sometimes, we choose to deprecate functionality or\nsyntax because it was found to be poorly designed or implemented. Sometimes, this is because\nthey're holding back other features or causing performance problems. Sometimes, the reasons\nare more complex. Wherever possible, we try to keep deprecated functionality available to\ndevelopers in its previous form for at least one major release. So long as a deprecated\nfeature isn't actively disrupting our ability to maintain and extend Perl, we'll try to leave\nit in place as long as possible.\n\nThe following items are now deprecated:\n\nsuidperl\n\"suidperl\" is no longer part of Perl. It used to provide a mechanism to emulate setuid\npermission bits on systems that don't support it properly.\n\nUse of \":=\" to mean an empty attribute list\nAn accident of Perl's parser meant that these constructions were all equivalent:\n\nmy $pi := 4;\nmy $pi : = 4;\nmy $pi : = 4;\n\nwith the \":\" being treated as the start of an attribute list, which ends before the \"=\".\nAs whitespace is not significant here, all are parsed as an empty attribute list, hence\nall the above are equivalent to, and better written as\n\nmy $pi = 4;\n\nbecause no attribute processing is done for an empty list.\n\nAs is, this meant that \":=\" cannot be used as a new token, without silently changing the\nmeaning of existing code. Hence that particular form is now deprecated, and will become a\nsyntax error. If it is absolutely necessary to have empty attribute lists (for example,\nbecause of a code generator) then avoid the warning by adding a space before the \"=\".\n\n\"UNIVERSAL->import()\"\nThe method \"UNIVERSAL->import()\" is now deprecated. Attempting to pass import arguments\nto a \"use UNIVERSAL\" statement will result in a deprecation warning.\n\nUse of \"goto\" to jump into a construct\nUsing \"goto\" to jump from an outer scope into an inner scope is now deprecated. This rare\nuse case was causing problems in the implementation of scopes.\n\nCustom character names in \\N{name} that don't look like names\nIn \"\\N{name}\", name can be just about anything. The standard Unicode names have a very\nlimited domain, but a custom name translator could create names that are, for example,\nmade up entirely of punctuation symbols. It is now deprecated to make names that don't\nbegin with an alphabetic character, and aren't alphanumeric or contain other than a very\nfew other characters, namely spaces, dashes, parentheses and colons. Because of the added\nmeaning of \"\\N\" (See \"\"\\N\" experimental regex escape\"), names that look like curly brace\n-enclosed quantifiers won't work. For example, \"\\N{3,4}\" now means to match 3 to 4 non-\nnewlines; before a custom name \"3,4\" could have been created.\n\nDeprecated Modules\nThe following modules will be removed from the core distribution in a future release, and\nshould be installed from CPAN instead. Distributions on CPAN which require these should\nadd them to their prerequisites. The core versions of these modules warnings will issue a\ndeprecation warning.\n\nIf you ship a packaged version of Perl, either alone or as part of a larger system, then\nyou should carefully consider the repercussions of core module deprecations. You may want\nto consider shipping your default build of Perl with packages for some or all deprecated\nmodules which install into \"vendor\" or \"site\" perl library directories. This will inhibit\nthe deprecation warnings.\n\nAlternatively, you may want to consider patching lib/deprecate.pm to provide deprecation\nwarnings specific to your packaging system or distribution of Perl, consistent with how\nyour packaging system or distribution manages a staged transition from a release where\nthe installation of a single package provides the given functionality, to a later release\nwhere the system administrator needs to know to install multiple packages to get that\nsame functionality.\n\nYou can silence these deprecation warnings by installing the modules in question from\nCPAN. To install the latest version of all of them, just install\n\"Task::Deprecations::512\".\n\nClass::ISA\nPod::Plainer\nShell\nSwitch\nSwitch is buggy and should be avoided. You may find Perl's new \"given\"/\"when\" feature\na suitable replacement. See \"Switch statements\" in perlsyn for more information.\n\nAssignment to $[\nUse of the attribute :locked on subroutines\nUse of \"locked\" with the attributes pragma\nUse of \"unique\" with the attributes pragma\nPerlpmflag\n\"Perlpmflag\" is no longer part of Perl's public API. Calling it now generates a\ndeprecation warning, and it will be removed in a future release. Although listed as part\nof the API, it was never documented, and only ever used in toke.c, and prior to 5.10,\nregcomp.c. In core, it has been replaced by a static function.\n\nNumerous Perl 4-era libraries\ntermcap.pl, tainted.pl, stat.pl, shellwords.pl, pwd.pl, open3.pl, open2.pl, newgetopt.pl,\nlook.pl, find.pl, finddepth.pl, importenv.pl, hostname.pl, getopts.pl, getopt.pl,\ngetcwd.pl, flush.pl, fastcwd.pl, exceptions.pl, ctime.pl, complete.pl, cacheout.pl,\nbigrat.pl, bigint.pl, bigfloat.pl, assert.pl, abbrev.pl, dotsh.pl, and timelocal.pl are\nall now deprecated. Earlier, Perl's developers intended to remove these libraries from\nPerl's core for the 5.14.0 release.\n\nDuring final testing before the release of 5.12.0, several developers discovered current\nproduction code using these ancient libraries, some inside the Perl core itself.\nAccordingly, the pumpking granted them a stay of execution. They will begin to warn about\ntheir deprecation in the 5.14.0 release and will be removed in the 5.16.0 release.\n", "subsections": [] }, "Unicode overhaul": { "content": "Perl's developers have made a concerted effort to update Perl to be in sync with the latest\nUnicode standard. Changes for this include:\n\nPerl can now handle every Unicode character property. New documentation, perluniprops, lists\nall available non-Unihan character properties. By default, perl does not expose Unihan,\ndeprecated or Unicode-internal properties. See below for more details on these; there is\nalso a section in the pod listing them, and explaining why they are not exposed.\n\nPerl now fully supports the Unicode compound-style of using \"=\" and \":\" in writing regular\nexpressions: \"\\p{property=value}\" and \"\\p{property:value}\" (both of which mean the same\nthing).\n\nPerl now fully supports the Unicode loose matching rules for text between the braces in\n\"\\p{...}\" constructs. In addition, Perl allows underscores between digits of numbers.\n\nPerl now accepts all the Unicode-defined synonyms for properties and property values.\n\n\"qr/\\X/\", which matches a Unicode logical character, has been expanded to work better with\nvarious Asian languages. It now is defined as an extended grapheme cluster. (See\n). Anything matched previously and that made sense\nwill continue to be accepted. Additionally:\n\n• \"\\X\" will not break apart a \"CR LF\" sequence.\n\n• \"\\X\" will now match a sequence which includes the \"ZWJ\" and \"ZWNJ\" characters.\n\n• \"\\X\" will now always match at least one character, including an initial mark. Marks\ngenerally come after a base character, but it is possible in Unicode to have them in\nisolation, and \"\\X\" will now handle that case, for example at the beginning of a line, or\nafter a \"ZWSP\". And this is the part where \"\\X\" doesn't match the things that it used to\nthat don't make sense. Formerly, for example, you could have the nonsensical case of an\naccented LF.\n\n• \"\\X\" will now match a (Korean) Hangul syllable sequence, and the Thai and Lao exception\ncases.\n\nOtherwise, this change should be transparent for the non-affected languages.\n\n\"\\p{...}\" matches using the CanonicalCombiningClass property were completely broken in\nprevious releases of Perl. They should now work correctly.\n\nBefore Perl 5.12, the Unicode \"DecompositionType=Compat\" property and a Perl extension had\nthe same name, which led to neither matching all the correct values (with more than 100\nmistakes in one, and several thousand in the other). The Perl extension has now been renamed\nto be \"DecompositionType=Noncanonical\" (short: \"dt=noncanon\"). It has the same meaning as\nwas previously intended, namely the union of all the non-canonical Decomposition types, with\nUnicode \"Compat\" being just one of those.\n\n\"\\p{DecompositionType=Canonical}\" now includes the Hangul syllables.\n\n\"\\p{Uppercase}\" and \"\\p{Lowercase}\" now work as the Unicode standard says they should. This\nmeans they each match a few more characters than they used to.\n\n\"\\p{Cntrl}\" now matches the same characters as \"\\p{Control}\". This means it no longer will\nmatch Private Use (gc=co), Surrogates (gc=cs), nor Format (gc=cf) code points. The Format\ncode points represent the biggest possible problem. All but 36 of them are either officially\ndeprecated or strongly discouraged from being used. Of those 36, likely the most widely used\nare the soft hyphen (U+00AD), and BOM, ZWSP, ZWNJ, WJ, and similar characters, plus\nbidirectional controls.\n\n\"\\p{Alpha}\" now matches the same characters as \"\\p{Alphabetic}\". Before 5.12, Perl's\ndefinition included a number of things that aren't really alpha (all marks) while omitting\nmany that were. The definitions of \"\\p{Alnum}\" and \"\\p{Word}\" depend on Alpha's definition\nand have changed accordingly.\n\n\"\\p{Word}\" no longer incorrectly matches non-word characters such as fractions.\n\n\"\\p{Print}\" no longer matches the line control characters: Tab, LF, CR, FF, VT, and NEL. This\nbrings it in line with standards and the documentation.\n\n\"\\p{XDigit}\" now matches the same characters as \"\\p{HexDigit}\". This means that in addition\nto the characters it currently matches, \"[A-Fa-f0-9]\", it will also match the 22 fullwidth\nequivalents, for example U+FF10: FULLWIDTH DIGIT ZERO.\n\nThe Numeric type property has been extended to include the Unihan characters.\n\nThere is a new Perl extension, the 'PresentIn', or simply 'In', property. This is an\nextension of the Unicode Age property, but \"\\p{In=5.0}\" matches any code point whose usage\nhas been determined as of Unicode version 5.0. The \"\\p{Age=5.0}\" only matches code points\nadded in precisely version 5.0.\n\nA number of properties now have the correct values for unassigned code points. The affected\nproperties are BidiClass, EastAsianWidth, JoiningType, DecompositionType,\nHangulSyllableType, NumericType, and LineBreak.\n\nThe DefaultIgnorableCodePoint, IDContinue, and IDStart properties are now up to date\nwith current Unicode definitions.\n\nEarlier versions of Perl erroneously exposed certain properties that are supposed to be\nUnicode internal-only. Use of these in regular expressions will now generate, if enabled, a\ndeprecation warning message. The properties are: OtherAlphabetic,\nOtherDefaultIgnorableCodePoint, OtherGraphemeExtend, OtherIDContinue, OtherIDStart,\nOtherLowercase, OtherMath, and OtherUppercase.\n\nIt is now possible to change which Unicode properties Perl understands on a per-installation\nbasis. As mentioned above, certain properties are turned off by default. These include all\nthe Unihan properties (which should be accessible via the CPAN module Unicode::Unihan) and\nany deprecated or Unicode internal-only property that Perl has never exposed.\n\nThe generated files in the \"lib/unicore/To\" directory are now more clearly marked as being\nstable, directly usable by applications. New hash entries in them give the format of the\nnormal entries, which allows for easier machine parsing. Perl can generate files in this\ndirectory for any property, though most are suppressed. You can find instructions for\nchanging which are written in perluniprops.\n", "subsections": [] }, "Modules and Pragmata": { "content": "", "subsections": [ { "name": "New Modules and Pragmata", "content": "\"autodie\"\n\"autodie\" is a new lexically-scoped alternative for the \"Fatal\" module. The bundled\nversion is 2.0601. Note that in this release, using a string eval when \"autodie\" is in\neffect can cause the autodie behaviour to leak into the surrounding scope. See \"BUGS\" in\nautodie for more details.\n\nVersion 2.0601 has been added to the Perl core.\n\n\"Compress::Raw::Bzip2\"\nVersion 2.024 has been added to the Perl core.\n\n\"overloading\"\n\"overloading\" allows you to lexically disable or enable overloading for some or all\noperations.\n\nVersion 0.001 has been added to the Perl core.\n\n\"parent\"\n\"parent\" establishes an ISA relationship with base classes at compile time. It provides\nthe key feature of \"base\" without further unwanted behaviors.\n\nVersion 0.223 has been added to the Perl core.\n\n\"Parse::CPAN::Meta\"\nVersion 1.40 has been added to the Perl core.\n\n\"VMS::DCLsym\"\nVersion 1.03 has been added to the Perl core.\n\n\"VMS::Stdio\"\nVersion 2.4 has been added to the Perl core.\n\n\"XS::APItest::KeywordRPN\"\nVersion 0.003 has been added to the Perl core.\n" }, { "name": "Updated Pragmata", "content": "\"base\"\nUpgraded from version 2.13 to 2.15.\n\n\"bignum\"\nUpgraded from version 0.22 to 0.23.\n\n\"charnames\"\n\"charnames\" now contains the Unicode NameAliases.txt database file. This has the effect\nof adding some extra \"\\N\" character names that formerly wouldn't have been recognised;\nfor example, \"\\N{LATIN CAPITAL LETTER GHA}\".\n\nUpgraded from version 1.06 to 1.07.\n\n\"constant\"\nUpgraded from version 1.13 to 1.20.\n\n\"diagnostics\"\n\"diagnostics\" now supports %.0f formatting internally.\n\n\"diagnostics\" no longer suppresses \"Use of uninitialized value in range (or flip)\"\nwarnings. [perl #71204]\n\nUpgraded from version 1.17 to 1.19.\n\n\"feature\"\nIn \"feature\", the meaning of the \":5.10\" and \":5.10.X\" feature bundles has changed\nslightly. The last component, if any (i.e. \"X\") is simply ignored. This is predicated on\nthe assumption that new features will not, in general, be added to maintenance releases.\nSo \":5.10\" and \":5.10.X\" have identical effect. This is a change to the behaviour\ndocumented for 5.10.0.\n\n\"feature\" now includes the \"unicodestrings\" feature:\n\nuse feature \"unicodestrings\";\n\nThis pragma turns on Unicode semantics for the case-changing operations (\"uc\", \"lc\",\n\"ucfirst\", \"lcfirst\") on strings that don't have the internal UTF-8 flag set, but that\ncontain single-byte characters between 128 and 255.\n\nUpgraded from version 1.11 to 1.16.\n\n\"less\"\n\"less\" now includes the \"stashname\" method to allow subclasses of \"less\" to pick where\nin %^H to store their stash.\n\nUpgraded from version 0.02 to 0.03.\n\n\"lib\"\nUpgraded from version 0.5565 to 0.62.\n\n\"mro\"\n\"mro\" is now implemented as an XS extension. The documented interface has not changed.\nCode relying on the implementation detail that some \"mro::\" methods happened to be\navailable at all times gets to \"keep both pieces\".\n\nUpgraded from version 1.00 to 1.02.\n\n\"overload\"\n\"overload\" now allow overloading of 'qr'.\n\nUpgraded from version 1.06 to 1.10.\n\n\"threads\"\nUpgraded from version 1.67 to 1.75.\n\n\"threads::shared\"\nUpgraded from version 1.14 to 1.32.\n\n\"version\"\n\"version\" now has support for \"Version number formats\" as described earlier in this\ndocument and in its own documentation.\n\nUpgraded from version 0.74 to 0.82.\n\n\"warnings\"\n\"warnings\" has a new warnings::fatalenabled() function. It also includes a new\n\"illegalproto\" warning category. See also \"New or Changed Diagnostics\" for this change.\n\nUpgraded from version 1.06 to 1.09.\n" }, { "name": "Updated Modules", "content": "\"Archive::Extract\"\nUpgraded from version 0.24 to 0.38.\n\n\"Archive::Tar\"\nUpgraded from version 1.38 to 1.54.\n\n\"Attribute::Handlers\"\nUpgraded from version 0.79 to 0.87.\n\n\"AutoLoader\"\nUpgraded from version 5.63 to 5.70.\n\n\"B::Concise\"\nUpgraded from version 0.74 to 0.78.\n\n\"B::Debug\"\nUpgraded from version 1.05 to 1.12.\n\n\"B::Deparse\"\nUpgraded from version 0.83 to 0.96.\n\n\"B::Lint\"\nUpgraded from version 1.09 to 1.1101.\n\n\"CGI\"\nUpgraded from version 3.29 to 3.48.\n\n\"Class::ISA\"\nUpgraded from version 0.33 to 0.36.\n\nNOTE: \"Class::ISA\" is deprecated and may be removed from a future version of Perl.\n\n\"Compress::Raw::Zlib\"\nUpgraded from version 2.008 to 2.024.\n\n\"CPAN\"\nUpgraded from version 1.9205 to 1.9456.\n\n\"CPANPLUS\"\nUpgraded from version 0.84 to 0.90.\n\n\"CPANPLUS::Dist::Build\"\nUpgraded from version 0.0602 to 0.46.\n\n\"Data::Dumper\"\nUpgraded from version 2.12114 to 2.125.\n\n\"DBFile\"\nUpgraded from version 1.8161 to 1.820.\n\n\"Devel::PPPort\"\nUpgraded from version 3.13 to 3.19.\n\n\"Digest\"\nUpgraded from version 1.15 to 1.16.\n\n\"Digest::MD5\"\nUpgraded from version 2.3601 to 2.39.\n\n\"Digest::SHA\"\nUpgraded from version 5.45 to 5.47.\n\n\"Encode\"\nUpgraded from version 2.23 to 2.39.\n\n\"Exporter\"\nUpgraded from version 5.62 to 5.6401.\n\n\"ExtUtils::CBuilder\"\nUpgraded from version 0.21 to 0.27.\n\n\"ExtUtils::Command\"\nUpgraded from version 1.13 to 1.16.\n\n\"ExtUtils::Constant\"\nUpgraded from version 0.2 to 0.22.\n\n\"ExtUtils::Install\"\nUpgraded from version 1.44 to 1.55.\n\n\"ExtUtils::MakeMaker\"\nUpgraded from version 6.42 to 6.56.\n\n\"ExtUtils::Manifest\"\nUpgraded from version 1.5101 to 1.57.\n\n\"ExtUtils::ParseXS\"\nUpgraded from version 2.1802 to 2.21.\n\n\"File::Fetch\"\nUpgraded from version 0.14 to 0.24.\n\n\"File::Path\"\nUpgraded from version 2.04 to 2.0801.\n\n\"File::Temp\"\nUpgraded from version 0.18 to 0.22.\n\n\"Filter::Simple\"\nUpgraded from version 0.82 to 0.84.\n\n\"Filter::Util::Call\"\nUpgraded from version 1.07 to 1.08.\n\n\"Getopt::Long\"\nUpgraded from version 2.37 to 2.38.\n\n\"IO\"\nUpgraded from version 1.2301 to 1.2502.\n\n\"IO::Zlib\"\nUpgraded from version 1.07 to 1.10.\n\n\"IPC::Cmd\"\nUpgraded from version 0.401 to 0.54.\n\n\"IPC::SysV\"\nUpgraded from version 1.05 to 2.01.\n\n\"Locale::Maketext\"\nUpgraded from version 1.12 to 1.14.\n\n\"Locale::Maketext::Simple\"\nUpgraded from version 0.18 to 0.21.\n\n\"Log::Message\"\nUpgraded from version 0.01 to 0.02.\n\n\"Log::Message::Simple\"\nUpgraded from version 0.04 to 0.06.\n\n\"Math::BigInt\"\nUpgraded from version 1.88 to 1.8901.\n\n\"Math::BigInt::FastCalc\"\nUpgraded from version 0.16 to 0.19.\n\n\"Math::BigRat\"\nUpgraded from version 0.21 to 0.24.\n\n\"Math::Complex\"\nUpgraded from version 1.37 to 1.56.\n\n\"Memoize\"\nUpgraded from version 1.0102 to 1.0103.\n\n\"MIME::Base64\"\nUpgraded from version 3.0701 to 3.08.\n\n\"Module::Build\"\nUpgraded from version 0.280801 to 0.3603.\n\n\"Module::CoreList\"\nUpgraded from version 2.12 to 2.29.\n\n\"Module::Load\"\nUpgraded from version 0.12 to 0.16.\n\n\"Module::Load::Conditional\"\nUpgraded from version 0.22 to 0.34.\n\n\"Module::Loaded\"\nUpgraded from version 0.01 to 0.06.\n\n\"Module::Pluggable\"\nUpgraded from version 3.6 to 3.9.\n\n\"Net::Ping\"\nUpgraded from version 2.33 to 2.36.\n\n\"NEXT\"\nUpgraded from version 0.6001 to 0.64.\n\n\"Object::Accessor\"\nUpgraded from version 0.32 to 0.36.\n\n\"Package::Constants\"\nUpgraded from version 0.01 to 0.02.\n\n\"PerlIO\"\nUpgraded from version 1.04 to 1.06.\n\n\"Pod::Parser\"\nUpgraded from version 1.35 to 1.37.\n\n\"Pod::Perldoc\"\nUpgraded from version 3.1402 to 3.1502.\n\n\"Pod::Plainer\"\nUpgraded from version 0.01 to 1.02.\n\nNOTE: \"Pod::Plainer\" is deprecated and may be removed from a future version of Perl.\n\n\"Pod::Simple\"\nUpgraded from version 3.05 to 3.13.\n\n\"Safe\"\nUpgraded from version 2.12 to 2.22.\n\n\"SelfLoader\"\nUpgraded from version 1.11 to 1.17.\n\n\"Storable\"\nUpgraded from version 2.18 to 2.22.\n\n\"Switch\"\nUpgraded from version 2.13 to 2.16.\n\nNOTE: \"Switch\" is deprecated and may be removed from a future version of Perl.\n\n\"Sys::Syslog\"\nUpgraded from version 0.22 to 0.27.\n\n\"Term::ANSIColor\"\nUpgraded from version 1.12 to 2.02.\n\n\"Term::UI\"\nUpgraded from version 0.18 to 0.20.\n\n\"Test\"\nUpgraded from version 1.25 to 1.2502.\n\n\"Test::Harness\"\nUpgraded from version 2.64 to 3.17.\n\n\"Test::Simple\"\nUpgraded from version 0.72 to 0.94.\n\n\"Text::Balanced\"\nUpgraded from version 2.0.0 to 2.02.\n\n\"Text::ParseWords\"\nUpgraded from version 3.26 to 3.27.\n\n\"Text::Soundex\"\nUpgraded from version 3.03 to 3.0301.\n\n\"Thread::Queue\"\nUpgraded from version 2.00 to 2.11.\n\n\"Thread::Semaphore\"\nUpgraded from version 2.01 to 2.09.\n\n\"Tie::RefHash\"\nUpgraded from version 1.37 to 1.38.\n\n\"Time::HiRes\"\nUpgraded from version 1.9711 to 1.9719.\n\n\"Time::Local\"\nUpgraded from version 1.18 to 1.190101.\n\n\"Time::Piece\"\nUpgraded from version 1.12 to 1.15.\n\n\"Unicode::Collate\"\nUpgraded from version 0.52 to 0.5201.\n\n\"Unicode::Normalize\"\nUpgraded from version 1.02 to 1.03.\n\n\"Win32\"\nUpgraded from version 0.34 to 0.39.\n\n\"Win32API::File\"\nUpgraded from version 0.100101 to 0.1101.\n\n\"XSLoader\"\nUpgraded from version 0.08 to 0.10.\n" }, { "name": "Removed Modules and Pragmata", "content": "\"attrs\"\nRemoved from the Perl core. Prior version was 1.02.\n\n\"CPAN::API::HOWTO\"\nRemoved from the Perl core. Prior version was 'undef'.\n\n\"CPAN::DeferedCode\"\nRemoved from the Perl core. Prior version was 5.50.\n\n\"CPANPLUS::inc\"\nRemoved from the Perl core. Prior version was 'undef'.\n\n\"DCLsym\"\nRemoved from the Perl core. Prior version was 1.03.\n\n\"ExtUtils::MakeMaker::bytes\"\nRemoved from the Perl core. Prior version was 6.42.\n\n\"ExtUtils::MakeMaker::vmsish\"\nRemoved from the Perl core. Prior version was 6.42.\n\n\"Stdio\"\nRemoved from the Perl core. Prior version was 2.3.\n\n\"Test::Harness::Assert\"\nRemoved from the Perl core. Prior version was 0.02.\n\n\"Test::Harness::Iterator\"\nRemoved from the Perl core. Prior version was 0.02.\n\n\"Test::Harness::Point\"\nRemoved from the Perl core. Prior version was 0.01.\n\n\"Test::Harness::Results\"\nRemoved from the Perl core. Prior version was 0.01.\n\n\"Test::Harness::Straps\"\nRemoved from the Perl core. Prior version was 0.2601.\n\n\"Test::Harness::Util\"\nRemoved from the Perl core. Prior version was 0.01.\n\n\"XSSymSet\"\nRemoved from the Perl core. Prior version was 1.1.\n" }, { "name": "Deprecated Modules and Pragmata", "content": "See \"Deprecated Modules\" above.\n" } ] }, "Documentation": { "content": "", "subsections": [ { "name": "New Documentation", "content": "• perlhaiku contains instructions on how to build perl for the Haiku platform.\n\n• perlmroapi describes the new interface for pluggable Method Resolution Orders.\n\n• perlperf, by Richard Foley, provides an introduction to the use of performance and\noptimization techniques which can be used with particular reference to perl programs.\n\n• perlrepository describes how to access the perl source using the git version control\nsystem.\n\n• perlpolicy extends the \"Social contract about contributed modules\" into the beginnings of\na document on Perl porting policies.\n" }, { "name": "Changes to Existing Documentation", "content": "• The various large Changes* files (which listed every change made to perl over the last 18\nyears) have been removed, and replaced by a small file, also called Changes, which just\nexplains how that same information may be extracted from the git version control system.\n\n• Porting/patching.pod has been deleted, as it mainly described interacting with the old\nPerforce-based repository, which is now obsolete. Information still relevant has been\nmoved to perlrepository.\n\n• The syntax \"unless (EXPR) BLOCK else BLOCK\" is now documented as valid, as is the syntax\n\"unless (EXPR) BLOCK elsif (EXPR) BLOCK ... else BLOCK\", although actually using the\nlatter may not be the best idea for the readability of your source code.\n\n• Documented -X overloading.\n\n• Documented that when() treats specially most of the filetest operators\n\n• Documented \"when\" as a syntax modifier.\n\n• Eliminated \"Old Perl threads tutorial\", which described 5005 threads.\n\npod/perlthrtut.pod is the same material reworked for ithreads.\n\n• Correct previous documentation: v-strings are not deprecated\n\nWith version objects, we need them to use MODULE VERSION syntax. This patch removes the\ndeprecation notice.\n\n• Security contact information is now part of perlsec.\n\n• A significant fraction of the core documentation has been updated to clarify the behavior\nof Perl's Unicode handling.\n\nMuch of the remaining core documentation has been reviewed and edited for clarity,\nconsistent use of language, and to fix the spelling of Tom Christiansen's name.\n\n• The Pod specification (perlpodspec) has been updated to bring the specification in line\nwith modern usage already supported by most Pod systems. A parameter string may now\nfollow the format name in a \"begin/end\" region. Links to URIs with a text description are\nnow allowed. The usage of \"L<\"section\">\" has been marked as deprecated.\n\n• if.pm has been documented in \"use\" in perlfunc as a means to get conditional loading of\nmodules despite the implicit BEGIN block around \"use\".\n\n• The documentation for $1 in perlvar.pod has been clarified.\n\n• \"\\N{U+code point}\" is now documented.\n" } ] }, "Selected Performance Enhancements": { "content": "• A new internal cache means that isa() will often be faster.\n\n• The implementation of \"C3\" Method Resolution Order has been optimised - linearisation for\nclasses with single inheritance is 40% faster. Performance for multiple inheritance is\nunchanged.\n\n• Under \"use locale\", the locale-relevant information is now cached on read-only values,\nsuch as the list returned by \"keys %hash\". This makes operations such as \"sort keys\n%hash\" in the scope of \"use locale\" much faster.\n\n• Empty \"DESTROY\" methods are no longer called.\n\n• Perlsvutf8upgrade() is now faster.\n\n• \"keys\" on empty hash is now faster.\n\n• \"if (%foo)\" has been optimized to be faster than \"if (keys %foo)\".\n\n• The string repetition operator (\"$str x $num\") is now several times faster when $str has\nlength one or $num is large.\n\n• Reversing an array to itself (as in \"@a = reverse @a\") in void context now happens in-\nplace and is several orders of magnitude faster than it used to be. It will also preserve\nnon-existent elements whenever possible, i.e. for non magical arrays or tied arrays with\n\"EXISTS\" and \"DELETE\" methods.\n", "subsections": [] }, "Installation and Configuration Improvements": { "content": "• perlapi, perlintern, perlmodlib and perltoc are now all generated at build time, rather\nthan being shipped as part of the release.\n\n• If \"vendorlib\" and \"vendorarch\" are the same, then they are only added to @INC once.\n\n• $Config{usedevel} and the C-level \"PERLUSEDEVEL\" are now defined if perl is built with\n\"-Dusedevel\".\n\n• Configure will enable use of \"-fstack-protector\", to provide protection against stack-\nsmashing attacks, if the compiler supports it.\n\n• Configure will now determine the correct prototypes for re-entrant functions and for\n\"gconvert\" if you are using a C++ compiler rather than a C compiler.\n\n• On Unix, if you build from a tree containing a git repository, the configuration process\nwill note the commit hash you have checked out, for display in the output of \"perl -v\"\nand \"perl -V\". Unpushed local commits are automatically added to the list of local\npatches displayed by \"perl -V\".\n\n• Perl now supports SystemTap's \"dtrace\" compatibility layer and an issue with linking\n\"miniperl\" has been fixed in the process.\n\n• perldoc now uses \"less -R\" instead of \"less\" for improved behaviour in the face of\n\"groff\"'s new usage of ANSI escape codes.\n\n• \"perl -V\" now reports use of the compile-time options \"USEPERLATOF\" and\n\"USEATTRIBUTESFORPERLIO\".\n\n• As part of the flattening of ext, all extensions on all platforms are built by\nmakeext.pl. This replaces the Unix-specific ext/util/makeext, VMS-specific makeext.com\nand Win32-specific win32/buildext.pl.\n", "subsections": [] }, "Internal Changes": { "content": "Each release of Perl sees numerous internal changes which shouldn't affect day to day usage\nbut may still be notable for developers working with Perl's source code.\n\n• The J.R.R. Tolkien quotes at the head of C source file have been checked and proper\ncitations added, thanks to a patch from Tom Christiansen.\n\n• The internal structure of the dual-life modules traditionally found in the lib/ and ext/\ndirectories in the perl source has changed significantly. Where possible, dual-lifed\nmodules have been extracted from lib/ and ext/.\n\nDual-lifed modules maintained by Perl's developers as part of the Perl core now live in\ndist/. Dual-lifed modules maintained primarily on CPAN now live in cpan/. When\nreporting a bug in a module located under cpan/, please send your bug report directly to\nthe module's bug tracker or author, rather than Perl's bug tracker.\n\n• \"\\N{...}\" now compiles better, always forces UTF-8 internal representation\n\nPerl's developers have fixed several problems with the recognition of \"\\N{...}\"\nconstructs. As part of this, perl will store any scalar or regex containing \"\\N{name}\"\nor \"\\N{U+code point}\" in its definition in UTF-8 format. (This was true previously for\nall occurrences of \"\\N{name}\" that did not use a custom translator, but now it's always\ntrue.)\n\n• Perlmagicsetmglob now knows about globs, fixing RT #71254.\n\n• \"SVtRV\" no longer exists. RVs are now stored in IVs.\n\n• Perlvcroak() now accepts a null first argument. In addition, a full audit was made of\nthe \"not NULL\" compiler annotations, and those for several other internal functions were\ncorrected.\n\n• New macros \"dSAVEDERRNO\", \"dSAVEERRNO\", \"SAVEERRNO\", \"RESTOREERRNO\" have been added to\nformalise the temporary saving of the \"errno\" variable.\n\n• The function \"Perlsvinsertflags\" has been added to augment \"Perlsvinsert\".\n\n• The function PerlnewSVtype(type) has been added, equivalent to PerlnewSV() followed by\nPerlsvupgrade(type).\n\n• The function PerlnewSVpvnflags() has been added, equivalent to PerlnewSVpvn() and then\nperforming the action relevant to the flag.\n\nTwo flag bits are currently supported.\n\n• \"SVfUTF8\" will call SvUTF8on() for you. (Note that this does not convert a sequence\nof ISO 8859-1 characters to UTF-8). A wrapper, newSVpvnutf8() is available for this.\n\n• \"SVsTEMP\" now calls Perlsv2mortal() on the new SV.\n\nThere is also a wrapper that takes constant strings, newSVpvsflags().\n\n• The function \"Perlcroakxsusage\" has been added as a wrapper to \"Perlcroak\".\n\n• Perl now exports the functions \"PerlIOfindlayer\" and \"PerlIOlistalloc\".\n\n• \"PLna\" has been exterminated from the core code, replaced by local STRLEN temporaries,\nor *nolen() calls. Either approach is faster than \"PLna\", which is a pointer\ndereference into the interpreter structure under ithreads, and a global variable\notherwise.\n\n• Perlmgfree() used to leave freed memory accessible via SvMAGIC() on the scalar. It now\nupdates the linked list to remove each piece of magic as it is freed.\n\n• Under ithreads, the regex in \"PLregcurpm\" is now reference counted. This eliminates a\nlot of hackish workarounds to cope with it not being reference counted.\n\n• Perlmgmagical() would sometimes incorrectly turn on SvRMAGICAL(). This has been fixed.\n\n• The public IV and NV flags are now not set if the string value has trailing \"garbage\".\nThis behaviour is consistent with not setting the public IV or NV flags if the value is\nout of range for the type.\n\n• Uses of \"Nullav\", \"Nullcv\", \"Nullhv\", \"Nullop\", \"Nullsv\" etc have been replaced by \"NULL\"\nin the core code, and non-dual-life modules, as \"NULL\" is clearer to those unfamiliar\nwith the core code.\n\n• A macro MUTABLEPTR(p) has been added, which on (non-pedantic) gcc will not cast away\n\"const\", returning a \"void *\". Macros MUTABLESV(av), MUTABLESV(cv) etc build on this,\ncasting to \"AV *\" etc without casting away \"const\". This allows proper compile-time\nauditing of \"const\" correctness in the core, and helped picked up some errors (now\nfixed).\n\n• Macros mPUSHs() and mXPUSHs() have been added, for pushing SVs on the stack and\nmortalizing them.\n\n• Use of the private structure \"mrometa\" has changed slightly. Nothing outside the core\nshould be accessing this directly anyway.\n\n• A new tool, Porting/expand-macro.pl has been added, that allows you to view how a C\npreprocessor macro would be expanded when compiled. This is handy when trying to decode\nthe macro hell that is the perl guts.\n", "subsections": [] }, "Testing": { "content": "", "subsections": [ { "name": "Testing improvements", "content": "Parallel tests\nThe core distribution can now run its regression tests in parallel on Unix-like\nplatforms. Instead of running \"make test\", set \"TESTJOBS\" in your environment to the\nnumber of tests to run in parallel, and run \"make testharness\". On a Bourne-like shell,\nthis can be done as\n\nTESTJOBS=3 make testharness # Run 3 tests in parallel\n\nAn environment variable is used, rather than parallel make itself, because TAP::Harness\nneeds to be able to schedule individual non-conflicting test scripts itself, and there is\nno standard interface to \"make\" utilities to interact with their job schedulers.\n\nNote that currently some test scripts may fail when run in parallel (most notably\n\"ext/IO/t/iodir.t\"). If necessary run just the failing scripts again sequentially and\nsee if the failures go away.\n\nTest harness flexibility\nIt's now possible to override \"PERL5OPT\" and friends in t/TEST\n\nTest watchdog\nSeveral tests that have the potential to hang forever if they fail now incorporate a\n\"watchdog\" functionality that will kill them after a timeout, which helps ensure that\n\"make test\" and \"make testharness\" run to completion automatically.\n" }, { "name": "New Tests", "content": "Perl's developers have added a number of new tests to the core. In addition to the items\nlisted below, many modules updated from CPAN incorporate new tests.\n\n• Significant cleanups to core tests to ensure that language and interpreter features are\nnot used before they're tested.\n\n• \"make testporting\" now runs a number of important pre-commit checks which might be of\nuse to anyone working on the Perl core.\n\n• t/porting/podcheck.t automatically checks the well-formedness of POD found in all .pl,\n.pm and .pod files in the MANIFEST, other than in dual-lifed modules which are primarily\nmaintained outside the Perl core.\n\n• t/porting/manifest.t now tests that all files listed in MANIFEST are present.\n\n• t/op/whilereaddir.t tests that a bare readdir in while loop sets $.\n\n• t/comp/retainedlines.t checks that the debugger can retain source lines from \"eval\".\n\n• t/io/perliofail.t checks that bad layers fail.\n\n• t/io/perlioleaks.t checks that PerlIO layers are not leaking.\n\n• t/io/perlioopen.t checks that certain special forms of open work.\n\n• t/io/perlio.t includes general PerlIO tests.\n\n• t/io/pvbm.t checks that there is no unexpected interaction between the internal types\n\"PVBM\" and \"PVGV\".\n\n• t/mro/packagealiases.t checks that mro works properly in the presence of aliased\npackages.\n\n• t/op/dbm.t tests \"dbmopen\" and \"dbmclose\".\n\n• t/op/indexthr.t tests the interaction of \"index\" and threads.\n\n• t/op/patthr.t tests the interaction of esoteric patterns and threads.\n\n• t/op/qrgc.t tests that \"qr\" doesn't leak.\n\n• t/op/regemailthr.t tests the interaction of regex recursion and threads.\n\n• t/op/regexpqrembedthr.t tests the interaction of patterns with embedded \"qr//\" and\nthreads.\n\n• t/op/regexpunicodeprop.t tests Unicode properties in regular expressions.\n\n• t/op/regexpunicodepropthr.t tests the interaction of Unicode properties and threads.\n\n• t/op/regnctie.t tests the tied methods of \"Tie::Hash::NamedCapture\".\n\n• t/op/regposixcc.t checks that POSIX character classes behave consistently.\n\n• t/op/re.t checks that exportable \"re\" functions in universal.c work.\n\n• t/op/setpgrpstack.t checks that \"setpgrp\" works.\n\n• t/op/substrthr.t tests the interaction of \"substr\" and threads.\n\n• t/op/upgrade.t checks that upgrading and assigning scalars works.\n\n• t/uni/lexutf8.t checks that Unicode in the lexer works.\n\n• t/uni/tie.t checks that Unicode and \"tie\" work.\n\n• t/comp/finallinenum.t tests whether line numbers are correct at EOF\n\n• t/comp/formscope.t tests format scoping.\n\n• t/comp/linedebug.t tests whether \"@{\"<$file\"}\" works.\n\n• t/op/filetestt.t tests if -t file test works.\n\n• t/op/qr.t tests \"qr\".\n\n• t/op/utf8cache.t tests malfunctions of the utf8 cache.\n\n• t/re/uniprops.t test unicodes \"\\p{}\" regex constructs.\n\n• t/op/filehandle.t tests some suitably portable filetest operators to check that they work\nas expected, particularly in the light of some internal changes made in how filehandles\nare blessed.\n\n• t/op/timeloop.t tests that unix times greater than \"263\", which can now be handed to\n\"gmtime\" and \"localtime\", do not cause an internal overflow or an excessively long loop.\n" } ] }, "New or Changed Diagnostics": { "content": "", "subsections": [ { "name": "New Diagnostics", "content": "• SV allocation tracing has been added to the diagnostics enabled by \"-Dm\". The tracing\ncan alternatively output via the \"PERLMEMLOG\" mechanism, if that was enabled when the\nperl binary was compiled.\n\n• Smartmatch resolution tracing has been added as a new diagnostic. Use \"-DM\" to enable it.\n\n• A new debugging flag \"-DB\" now dumps subroutine definitions, leaving \"-Dx\" for its\noriginal purpose of dumping syntax trees.\n\n• Perl 5.12 provides a number of new diagnostic messages to help you write better code.\nSee perldiag for details of these new messages.\n\n• \"Bad plugin affecting keyword '%s'\"\n\n• \"gmtime(%.0f) too large\"\n\n• \"Lexing code attempted to stuff non-Latin-1 character into Latin-1 input\"\n\n• \"Lexing code internal error (%s)\"\n\n• \"localtime(%.0f) too large\"\n\n• \"Overloaded dereference did not return a reference\"\n\n• \"Overloaded qr did not return a REGEXP\"\n\n• \"Perlpmflag() is deprecated, and will be removed from the XS API\"\n\n• \"lvalue attribute ignored after the subroutine has been defined\"\n\nThis new warning is issued when one attempts to mark a subroutine as lvalue after it\nhas been defined.\n\n• Perl now warns you if \"++\" or \"--\" are unable to change the value because it's beyond\nthe limit of representation.\n\nThis uses a new warnings category: \"imprecision\".\n\n• \"lc\", \"uc\", \"lcfirst\", and \"ucfirst\" warn when passed undef.\n\n• \"Show constant in \"Useless use of a constant in void context\"\"\n\n• \"Prototype after '%s'\"\n\n• \"panic: svchop %s\"\n\nThis new fatal error occurs when the C routine Perlsvchop() was passed a position\nthat is not within the scalar's string buffer. This could be caused by buggy XS code,\nand at this point recovery is not possible.\n\n• The fatal error \"Malformed UTF-8 returned by \\N\" is now produced if the \"charnames\"\nhandler returns malformed UTF-8.\n\n• If an unresolved named character or sequence was encountered when compiling a regex\npattern then the fatal error \"\\N{NAME} must be resolved by the lexer\" is now\nproduced. This can happen, for example, when using a single-quotish context like \"$re\n= '\\N{SPACE}'; /$re/;\". See perldiag for more examples of how the lexer can get\nbypassed.\n\n• \"Invalid hexadecimal number in \\N{U+...}\" is a new fatal error triggered when the\ncharacter constant represented by \"...\" is not a valid hexadecimal number.\n\n• The new meaning of \"\\N\" as \"[^\\n]\" is not valid in a bracketed character class, just\nlike \".\" in a character class loses its special meaning, and will cause the fatal\nerror \"\\N in a character class must be a named character: \\N{...}\".\n\n• The rules on what is legal for the \"...\" in \"\\N{...}\" have been tightened up so that\nunless the \"...\" begins with an alphabetic character and continues with a combination\nof alphanumerics, dashes, spaces, parentheses or colons then the warning \"Deprecated\ncharacter(s) in \\N{...} starting at '%s'\" is now issued.\n\n• The warning \"Using just the first characters returned by \\N{}\" will be issued if the\n\"charnames\" handler returns a sequence of characters which exceeds the limit of the\nnumber of characters that can be used. The message will indicate which characters\nwere used and which were discarded.\n" }, { "name": "Changed Diagnostics", "content": "A number of existing diagnostic messages have been improved or corrected:\n\n• A new warning category \"illegalproto\" allows finer-grained control of warnings around\nfunction prototypes.\n\nThe two warnings:\n\n\"Illegal character in prototype for %s : %s\"\n\"Prototype after '%c' for %s : %s\"\n\nhave been moved from the \"syntax\" top-level warnings category into a new first-level\ncategory, \"illegalproto\". These two warnings are currently the only ones emitted during\nparsing of an invalid/illegal prototype, so one can now use\n\nno warnings 'illegalproto';\n\nto suppress only those, but not other syntax-related warnings. Warnings where prototypes\nare changed, ignored, or not met are still in the \"prototype\" category as before.\n\n• \"Deep recursion on subroutine \"%s\"\"\n\nIt is now possible to change the depth threshold for this warning from the default of\n100, by recompiling the perl binary, setting the C pre-processor macro\n\"PERLSUBDEPTHWARN\" to the desired value.\n\n• \"Illegal character in prototype\" warning is now more precise when reporting illegal\ncharacters after\n\n• mro merging error messages are now very similar to those produced by Algorithm::C3.\n\n• Amelioration of the error message \"Unrecognized character %s in column %d\"\n\nChanges the error message to \"Unrecognized character %s; marked by <-- HERE after %s<--\nHERE near column %d\". This should make it a little simpler to spot and correct the\nsuspicious character.\n\n• Perl now explicitly points to $. when it causes an uninitialized warning for ranges in\nscalar context.\n\n• \"split\" now warns when called in void context.\n\n• \"printf\"-style functions called with too few arguments will now issue the warning\n\"Missing argument in %s\" [perl #71000]\n\n• Perl now properly returns a syntax error instead of segfaulting if \"each\", \"keys\", or\n\"values\" is used without an argument.\n\n• tell() now fails properly if called without an argument and when no previous file was\nread.\n\ntell() now returns -1, and sets errno to \"EBADF\", thus restoring the 5.8.x behaviour.\n\n• \"overload\" no longer implicitly unsets fallback on repeated 'use overload' lines.\n\n• POSIX::strftime() can now handle Unicode characters in the format string.\n\n• The \"syntax\" category was removed from 5 warnings that should only be in \"deprecated\".\n\n• Three fatal \"pack\"/\"unpack\" error messages have been normalized to \"panic: %s\"\n\n• \"Unicode character is illegal\" has been rephrased to be more accurate\n\nIt now reads \"Unicode non-character is illegal in interchange\" and the perldiag\ndocumentation has been expanded a bit.\n\n• Currently, all but the first of the several characters that the \"charnames\" handler may\nreturn are discarded when used in a regular expression pattern bracketed character class.\nIf this happens then the warning \"Using just the first character returned by \\N{} in\ncharacter class\" will be issued.\n\n• The warning \"Missing right brace on \\N{} or unescaped left brace after \\N. Assuming the\nlatter\" will be issued if Perl encounters a \"\\N{\" but doesn't find a matching \"}\". In\nthis case Perl doesn't know if it was mistakenly omitted, or if \"match non-newline\"\nfollowed by \"match a \"{\"\" was desired. It assumes the latter because that is actually a\nvalid interpretation as written, unlike the other case. If you meant the former, you\nneed to add the matching right brace. If you did mean the latter, you can silence this\nwarning by writing instead \"\\N\\{\".\n\n• \"gmtime\" and \"localtime\" called with numbers smaller than they can reliably handle will\nnow issue the warnings \"gmtime(%.0f) too small\" and \"localtime(%.0f) too small\".\n\nThe following diagnostic messages have been removed:\n\n• \"Runaway format\"\n\n• \"Can't locate package %s for the parents of %s\"\n\nIn general this warning it only got produced in conjunction with other warnings, and\nremoving it allowed an ISA lookup optimisation to be added.\n\n• \"v-string in use/require is non-portable\"\n" } ] }, "Utility Changes": { "content": "• h2ph now looks in \"include-fixed\" too, which is a recent addition to gcc's search path.\n\n• h2xs no longer incorrectly treats enum values like macros. It also now handles C++ style\ncomments (\"//\") properly in enums.\n\n• perl5db.pl now supports \"LVALUE\" subroutines. Additionally, the debugger now correctly\nhandles proxy constant subroutines, and subroutine stubs.\n\n• perlbug now uses %Module::CoreList::bugtracker to print out upstream bug tracker URLs.\nIf a user identifies a particular module as the topic of their bug report and we're able\nto divine the URL for its upstream bug tracker, perlbug now provide a message to the user\nexplaining that the core copies the CPAN version directly, and provide the URL for\nreporting the bug directly to the upstream author.\n\nperlbug no longer reports \"Message sent\" when it hasn't actually sent the message\n\n• perlthanks is a new utility for sending non-bug-reports to the authors and maintainers of\nPerl. Getting nothing but bug reports can become a bit demoralising. If Perl 5.12 works\nwell for you, please try out perlthanks. It will make the developers smile.\n\n• Perl's developers have fixed bugs in a2p having to do with the match() operator in list\ncontext. Additionally, a2p no longer generates code that uses the $[ variable.\n", "subsections": [] }, "Selected Bug Fixes": { "content": "• U+0FFFF is now a legal character in regular expressions.\n\n• ppqr now always returns a new regexp SV. Resolves RT #69852.\n\nInstead of returning a(nother) reference to the (pre-compiled) regexp in the optree, use\nregtempcopy() to create a copy of it, and return a reference to that. This resolves\nissues about Regexp::DESTROY not being called in a timely fashion (the original bug\ntracked by RT #69852), as well as bugs related to blessing regexps, and of assigning to\nregexps, as described in correspondence added to the ticket.\n\nIt transpires that we also need to undo the SvPVX() sharing when ithreads cloning a\nRegexp SV, because motherre is set to NULL, instead of a cloned copy of the motherre.\nThis change might fix bugs with regexps and threads in certain other situations, but as\nyet neither tests nor bug reports have indicated any problems, so it might not actually\nbe an edge case that it's possible to reach.\n\n• Several compilation errors and segfaults when perl was built with \"-Dmad\" were fixed.\n\n• Fixes for lexer API changes in 5.11.2 which broke NYTProf's savesrc option.\n\n• \"-t\" should only return TRUE for file handles connected to a TTY\n\nThe Microsoft C version of isatty() returns TRUE for all character mode devices,\nincluding the /dev/null-style \"nul\" device and printers like \"lpt1\".\n\n• Fixed a regression caused by commit fafafbaf which caused a panic during parameter\npassing [perl #70171]\n\n• On systems which in-place edits without backup files, -i'*' now works as the\ndocumentation says it does [perl #70802]\n\n• Saving and restoring magic flags no longer loses readonly flag.\n\n• The malformed syntax \"grep EXPR LIST\" (note the missing comma) no longer causes abrupt\nand total failure.\n\n• Regular expressions compiled with \"qr{}\" literals properly set \"$'\" when matching again.\n\n• Using named subroutines with \"sort\" should no longer lead to bus errors [perl #71076]\n\n• Numerous bugfixes catch small issues caused by the recently-added Lexer API.\n\n• Smart match against @ sometimes gave false negatives. [perl #71078]\n\n• $@ may now be assigned a read-only value (without error or busting the stack).\n\n• \"sort\" called recursively from within an active comparison subroutine no longer causes a\nbus error if run multiple times. [perl #71076]\n\n• Tie::Hash::NamedCapture::* will not abort if passed bad input (RT #71828)\n\n• @ and $ no longer leak under threads (RT #34342 and #41138, also #70602, #70974)\n\n• \"-I\" on shebang line now adds directories in front of @INC as documented, and as does\n\"-I\" when specified on the command-line.\n\n• \"kill\" is now fatal when called on non-numeric process identifiers. Previously, an\n\"undef\" process identifier would be interpreted as a request to kill process 0, which\nwould terminate the current process group on POSIX systems. Since process identifiers are\nalways integers, killing a non-numeric process is now fatal.\n\n• 5.10.0 inadvertently disabled an optimisation, which caused a measurable performance drop\nin list assignment, such as is often used to assign function parameters from @. The\noptimisation has been re-instated, and the performance regression fixed. (This fix is\nalso present in 5.10.1)\n\n• Fixed memory leak on \"while (1) { map 1, 1 }\" [RT #53038].\n\n• Some potential coredumps in PerlIO fixed [RT #57322,54828].\n\n• The debugger now works with lvalue subroutines.\n\n• The debugger's \"m\" command was broken on modules that defined constants [RT #61222].\n\n• \"crypt\" and string complement could return tainted values for untainted arguments [RT\n#59998].\n\n• The \"-i\".suffix command-line switch now recreates the file using restricted permissions,\nbefore changing its mode to match the original file. This eliminates a potential race\ncondition [RT #60904].\n\n• On some Unix systems, the value in $? would not have the top bit set (\"$? & 128\") even if\nthe child core dumped.\n\n• Under some circumstances, $^R could incorrectly become undefined [RT #57042].\n\n• In the XS API, various hash functions, when passed a pre-computed hash where the key is\nUTF-8, might result in an incorrect lookup.\n\n• XS code including XSUB.h before perl.h gave a compile-time error [RT #57176].\n\n• \"$object->isa('Foo')\" would report false if the package \"Foo\" didn't exist, even if the\nobject's @ISA contained \"Foo\".\n\n• Various bugs in the new-to 5.10.0 mro code, triggered by manipulating @ISA, have been\nfound and fixed.\n\n• Bitwise operations on references could crash the interpreter, e.g. \"$x=\\$y; $x |= \"foo\"\"\n[RT #54956].\n\n• Patterns including alternation might be sensitive to the internal UTF-8 representation,\ne.g.\n\nmy $byte = chr(192);\nmy $utf8 = chr(192); utf8::upgrade($utf8);\n$utf8 =~ /$byte|X}/i; # failed in 5.10.0\n\n• Within UTF8-encoded Perl source files (i.e. where \"use utf8\" is in effect), double-quoted\nliteral strings could be corrupted where a \"\\xNN\", \"\\0NNN\" or \"\\N{}\" is followed by a\nliteral character with ordinal value greater than 255 [RT #59908].\n\n• \"B::Deparse\" failed to correctly deparse various constructs: \"readpipe STRING\" [RT\n#62428], CORE::require(STRING) [RT #62488], \"sub foo()\" [RT #62484].\n\n• Using \"setpgrp\" with no arguments could corrupt the perl stack.\n\n• The block form of \"eval\" is now specifically trappable by \"Safe\" and \"ops\". Previously it\nwas erroneously treated like string \"eval\".\n\n• In 5.10.0, the two characters \"[~\" were sometimes parsed as the smart match operator\n(\"~~\") [RT #63854].\n\n• In 5.10.0, the \"*\" quantifier in patterns was sometimes treated as \"{0,32767}\" [RT\n#60034, #60464]. For example, this match would fail:\n\n(\"ab\" x 32768) =~ /^(ab)*$/\n\n• \"shmget\" was limited to a 32 bit segment size on a 64 bit OS [RT #63924].\n\n• Using \"next\" or \"last\" to exit a \"given\" block no longer produces a spurious warning like\nthe following:\n\nExiting given via last at foo.pl line 123\n\n• Assigning a format to a glob could corrupt the format; e.g.:\n\n*bar=*foo{FORMAT}; # foo format now bad\n\n• Attempting to coerce a typeglob to a string or number could cause an assertion failure.\nThe correct error message is now generated, \"Can't coerce GLOB to $type\".\n\n• Under \"use filetest 'access'\", \"-x\" was using the wrong access mode. This has been fixed\n[RT #49003].\n\n• \"length\" on a tied scalar that returned a Unicode value would not be correct the first\ntime. This has been fixed.\n\n• Using an array \"tie\" inside in array \"tie\" could SEGV. This has been fixed. [RT #51636]\n\n• A race condition inside PerlIOStdioclose() has been identified and fixed. This used to\ncause various threading issues, including SEGVs.\n\n• In \"unpack\", the use of \"()\" groups in scalar context was internally placing a list on\nthe interpreter's stack, which manifested in various ways, including SEGVs. This is now\nfixed [RT #50256].\n\n• Magic was called twice in \"substr\", \"\\&$x\", \"tie $x, $m\" and \"chop\". These have all been\nfixed.\n\n• A 5.10.0 optimisation to clear the temporary stack within the implicit loop of \"s///ge\"\nhas been reverted, as it turned out to be the cause of obscure bugs in seemingly\nunrelated parts of the interpreter [commit ef0d4e17921ee3de].\n\n• The line numbers for warnings inside \"elsif\" are now correct.\n\n• The \"..\" operator now works correctly with ranges whose ends are at or close to the\nvalues of the smallest and largest integers.\n\n• \"binmode STDIN, ':raw'\" could lead to segmentation faults on some platforms. This has\nbeen fixed [RT #54828].\n\n• An off-by-one error meant that \"index $str, ...\" was effectively being executed as \"index\n\"$str\\0\", ...\". This has been fixed [RT #53746].\n\n• Various leaks associated with named captures in regexes have been fixed [RT #57024].\n\n• A weak reference to a hash would leak. This was affecting \"DBI\" [RT #56908].\n\n• Using (?|) in a regex could cause a segfault [RT #59734].\n\n• Use of a UTF-8 \"tr//\" within a closure could cause a segfault [RT #61520].\n\n• Calling Perlsvchop() or otherwise upgrading an SV could result in an unaligned 64-bit\naccess on the SPARC architecture [RT #60574].\n\n• In the 5.10.0 release, \"incversionlist\" would incorrectly list \"5.10.*\" after \"5.8.*\";\nthis affected the @INC search order [RT #67628].\n\n• In 5.10.0, \"pack \"a*\", $taintedvalue\" returned a non-tainted value [RT #52552].\n\n• In 5.10.0, \"printf\" and \"sprintf\" could produce the fatal error \"panic:\nutf8mgposcacheupdate\" when printing UTF-8 strings [RT #62666].\n\n• In the 5.10.0 release, a dynamically created \"AUTOLOAD\" method might be missed (method\ncache issue) [RT #60220,60232].\n\n• In the 5.10.0 release, a combination of \"use feature\" and \"//ee\" could cause a memory\nleak [RT #63110].\n\n• \"-C\" on the shebang (\"#!\") line is once more permitted if it is also specified on the\ncommand line. \"-C\" on the shebang line used to be a silent no-op if it was not also on\nthe command line, so perl 5.10.0 disallowed it, which broke some scripts. Now perl checks\nwhether it is also on the command line and only dies if it is not [RT #67880].\n\n• In 5.10.0, certain types of re-entrant regular expression could crash, or cause the\nfollowing assertion failure [RT #60508]:\n\nAssertion rx->sublen >= (s - rx->subbeg) + i failed\n\n• Perl now includes previously missing files from the Unicode Character Database.\n\n• Perl now honors \"TMPDIR\" when opening an anonymous temporary file.\n", "subsections": [] }, "Platform Specific Changes": { "content": "Perl is incredibly portable. In general, if a platform has a C compiler, someone has ported\nPerl to it (or will soon). We're happy to announce that Perl 5.12 includes support for\nseveral new platforms. At the same time, it's time to bid farewell to some (very) old\nfriends.\n", "subsections": [ { "name": "New Platforms", "content": "Haiku\nPerl's developers have merged patches from Haiku's maintainers. Perl should now build on\nHaiku.\n\nMirOS BSD\nPerl should now build on MirOS BSD.\n" }, { "name": "Discontinued Platforms", "content": "Domain/OS\nMiNT\nTenon MachTen\n" }, { "name": "Updated Platforms", "content": "AIX\n• Removed libbsd for AIX 5L and 6.1. Only flock() was used from libbsd.\n\n• Removed libgdbm for AIX 5L and 6.1 if libgdbm < 1.8.3-5 is installed. The libgdbm is\ndelivered as an optional package with the AIX Toolbox. Unfortunately the versions\nbelow 1.8.3-5 are broken.\n\n• Hints changes mean that AIX 4.2 should work again.\n\nCygwin\n• Perl now supports IPv6 on Cygwin 1.7 and newer.\n\n• On Cygwin we now strip the last number from the DLL. This has been the behaviour in\nthe cygwin.com build for years. The hints files have been updated.\n\nDarwin (Mac OS X)\n• Skip testing the beBY.CP1131 locale on Darwin 10 (Mac OS X 10.6), as it's still\nbuggy.\n\n• Correct infelicities in the regexp used to identify buggy locales on Darwin 8 and 9\n(Mac OS X 10.4 and 10.5, respectively).\n\nDragonFly BSD\n• Fix thread library selection [perl #69686]\n\nFreeBSD\n• The hints files now identify the correct threading libraries on FreeBSD 7 and later.\n\nIrix\n• We now work around a bizarre preprocessor bug in the Irix 6.5 compiler: \"cc -E -\"\nunfortunately goes into K&R mode, but \"cc -E file.c\" doesn't.\n\nNetBSD\n• Hints now supports versions 5.*.\n\nOpenVMS\n• \"-UDEBUGGING\" is now the default on VMS.\n\nLike it has been everywhere else for ages and ages. Also make command-line selection\nof -UDEBUGGING and -DDEBUGGING work in configure.com; before the only way to turn it\noff was by saying no in answer to the interactive question.\n\n• The default pipe buffer size on VMS has been updated to 8192 on 64-bit systems.\n\n• Reads from the in-memory temporary files of \"PerlIO::scalar\" used to fail if $/ was\nset to a numeric reference (to indicate record-style reads). This is now fixed.\n\n• VMS now supports \"getgrgid\".\n\n• Many improvements and cleanups have been made to the VMS file name handling and\nconversion code.\n\n• Enabling the \"PERLVMSPOSIXEXIT\" logical name now encodes a POSIX exit status in a\nVMS condition value for better interaction with GNV's bash shell and other utilities\nthat depend on POSIX exit values. See \"$?\" in perlvms for details.\n\n• \"File::Copy\" now detects Unix compatibility mode on VMS.\n\nStratus VOS\n• Various changes from Stratus have been merged in.\n\nSymbian\n• There is now support for Symbian S60 3.2 SDK and S60 5.0 SDK.\n\nWindows\n• Perl 5.12 supports Windows 2000 and later. The supporting code for legacy versions of\nWindows is still included, but will be removed during the next development cycle.\n\n• Initial support for building Perl with MinGW-w64 is now available.\n\n• perl.exe now includes a manifest resource to specify the \"trustInfo\" settings for\nWindows Vista and later. Without this setting Windows would treat perl.exe as a\nlegacy application and apply various heuristics like redirecting access to protected\nfile system areas (like the \"Program Files\" folder) to the users \"VirtualStore\"\ninstead of generating a proper \"permission denied\" error.\n\nThe manifest resource also requests the Microsoft Common-Controls version 6.0 (themed\ncontrols introduced in Windows XP). Check out the Win32::VisualStyles module on CPAN\nto switch back to old style unthemed controls for legacy applications.\n\n• The \"-t\" filetest operator now only returns true if the filehandle is connected to a\nconsole window. In previous versions of Perl it would return true for all character\nmode devices, including NUL and LPT1.\n\n• The \"-p\" filetest operator now works correctly, and the Fcntl::SIFIFO constant is\ndefined when Perl is compiled with Microsoft Visual C. In previous Perl versions\n\"-p\" always returned a false value, and the Fcntl::SIFIFO constant was not defined.\n\nThis bug is specific to Microsoft Visual C and never affected Perl binaries built\nwith MinGW.\n\n• The socket error codes are now more widely supported: The POSIX module will define\nthe symbolic names, like POSIX::EWOULDBLOCK, and stringification of socket error\ncodes in $! works as well now;\n\nC:\\>perl -MPOSIX -E \"$!=POSIX::EWOULDBLOCK; say $!\"\nA non-blocking socket operation could not be completed immediately.\n\n• flock() will now set sensible error codes in $!. Previous Perl versions copied the\nvalue of $^E into $!, which caused much confusion.\n\n• select() now supports all empty \"fdset\"s more correctly.\n\n• '.\\foo' and '..\\foo' were treated differently than './foo' and '../foo' by \"do\" and\n\"require\" [RT #63492].\n\n• Improved message window handling means that \"alarm\" and \"kill\" messages will no\nlonger be dropped under race conditions.\n\n• Various bits of Perl's build infrastructure are no longer converted to win32 line\nendings at release time. If this hurts you, please report the problem with the\nperlbug program included with perl.\n" } ] }, "Known Problems": { "content": "This is a list of some significant unfixed bugs, which are regressions from either 5.10.x or\n5.8.x.\n\n• Some CPANPLUS tests may fail if there is a functioning file ../../cpanp-run-perl outside\nyour build directory. The failure shouldn't imply there's a problem with the actual\nfunctional software. The bug is already fixed in [RT #74188] and is scheduled for\ninclusion in perl-v5.12.1.\n\n• \"List::Util::first\" misbehaves in the presence of a lexical $ (typically introduced by\n\"my $\" or implicitly by \"given\"). The variable which gets set for each iteration is the\npackage variable $, not the lexical $ [RT #67694].\n\nA similar issue may occur in other modules that provide functions which take a block as\ntheir first argument, like\n\nfoo { ... $ ...} list\n\n• Some regexes may run much more slowly when run in a child thread compared with the thread\nthe pattern was compiled into [RT #55600].\n\n• Things like \"\"\\N{LATIN SMALL LIGATURE FF}\" =~ /\\N{LATIN SMALL LETTER F}+/\" will appear to\nhang as they get into a very long running loop [RT #72998].\n\n• Several porters have reported mysterious crashes when Perl's entire test suite is run\nafter a build on certain Windows 2000 systems. When run by hand, the individual tests\nreportedly work fine.\n", "subsections": [] }, "Errata": { "content": "• This one is actually a change introduced in 5.10.0, but it was missed from that release's\nperldelta, so it is mentioned here instead.\n\nA bugfix related to the handling of the \"/m\" modifier and \"qr\" resulted in a change of\nbehaviour between 5.8.x and 5.10.0:\n\n# matches in 5.8.x, doesn't match in 5.10.0\n$re = qr/^bar/; \"foo\\nbar\" =~ /$re/m;\n", "subsections": [] }, "Acknowledgements": { "content": "Perl 5.12.0 represents approximately two years of development since Perl 5.10.0 and contains\nover 750,000 lines of changes across over 3,000 files from over 200 authors and committers.\n\nPerl continues to flourish into its third decade thanks to a vibrant community of users and\ndevelopers. The following people are known to have contributed the improvements that became\nPerl 5.12.0:\n\nAaron Crane, Abe Timmerman, Abhijit Menon-Sen, Abigail, Adam Russell, Adriano Ferreira, Ævar\nArnfjörð Bjarmason, Alan Grover, Alexandr Ciornii, Alex Davies, Alex Vandiver, Andreas\nKoenig, Andrew Rodland, andrew@sundale.net, Andy Armstrong, Andy Dougherty, Jose AUGUSTE-\nETIENNE, Benjamin Smith, Ben Morrow, bharanee rathna, Bo Borgerson, Bo Lindbergh, Brad\nGilbert, Bram, Brendan O'Dea, brian d foy, Charles Bailey, Chip Salzenberg, Chris 'BinGOs'\nWilliams, Christoph Lamprecht, Chris Williams, chromatic, Claes Jakobsson, Craig A. Berry,\nDan Dascalescu, Daniel Frederick Crisman, Daniel M. Quinlan, Dan Jacobson, Dan Kogai, Dave\nMitchell, Dave Rolsky, David Cantrell, David Dick, David Golden, David Mitchell, David M.\nSyzdek, David Nicol, David Wheeler, Dennis Kaarsemaker, Dintelmann, Peter, Dominic Dunlop,\nDr.Ruud, Duke Leto, Enrico Sorcinelli, Eric Brine, Father Chrysostomos, Florian Ragwitz,\nFrank Wiegand, Gabor Szabo, Gene Sullivan, Geoffrey T. Dairiki, George Greer, Gerard Goossen,\nGisle Aas, Goro Fuji, Graham Barr, Green, Paul, Hans Dieter Pearcey, Harmen, H. Merijn Brand,\nHugo van der Sanden, Ian Goodacre, Igor Sutton, Ingo Weinhold, James Bence, James Mastros,\nJan Dubois, Jari Aalto, Jarkko Hietaniemi, Jay Hannah, Jerry Hedden, Jesse Vincent, Jim\nCromie, Jody Belka, John E. Malmberg, John Malmberg, John Peacock, John Peacock via RT, John\nP. Linderman, John Wright, Josh ben Jore, Jos I. Boumans, Karl Williamson, Kenichi Ishigaki,\nKen Williams, Kevin Brintnall, Kevin Ryde, Kurt Starsinic, Leon Brocard, Lubomir Rintel, Luke\nRoss, Marcel Grünauer, Marcus Holland-Moritz, Mark Jason Dominus, Marko Asplund, Martin\nHasch, Mashrab Kuvatov, Matt Kraai, Matt S Trout, Max Maischein, Michael Breen, Michael\nCartmell, Michael G Schwern, Michael Witten, Mike Giroux, Milosz Tanski, Moritz Lenz,\nNicholas Clark, Nick Cleaton, Niko Tyni, Offer Kaye, Osvaldo Villalon, Paul Fenwick, Paul\nGaborit, Paul Green, Paul Johnson, Paul Marquess, Philip Hazel, Philippe Bruhat, Rafael\nGarcia-Suarez, Rainer Tammer, Rajesh Mandalemula, Reini Urban, Renée Bäcker, Ricardo Signes,\nRicardo SIGNES, Richard Foley, Rich Rauenzahn, Rick Delaney, Risto Kankkunen, Robert May,\nRoberto C. Sanchez, Robin Barker, SADAHIRO Tomoyuki, Salvador Ortiz Garcia, Sam Vilain, Scott\nLanning, Sébastien Aperghis-Tramoni, Sérgio Durigan Júnior, Shlomi Fish, Simon 'corecode'\nSchubert, Sisyphus, Slaven Rezic, Smylers, Steffen Müller, Steffen Ullrich, Stepan Kasal,\nSteve Hay, Steven Schubiger, Steve Peters, Tels, The Doctor, Tim Bunce, Tim Jenness, Todd\nRinaldo, Tom Christiansen, Tom Hukins, Tom Wyant, Tony Cook, Torsten Schoenfeld, Tye McQueen,\nVadim Konovalov, Vincent Pit, Hio YAMASHINA, Yasuhiro Matsumoto, Yitzchak Scott-Thoennes,\nYuval Kogman, Yves Orton, Zefram, Zsban Ambrus\n\nThis is woefully incomplete as it's automatically generated from version control history. In\nparticular, it doesn't include the names of the (very much appreciated) contributors who\nreported issues in previous versions of Perl that helped make Perl 5.12.0 better. For a more\ncomplete list of all of Perl's historical contributors, please see the \"AUTHORS\" file in the\nPerl 5.12.0 distribution.\n\nOur \"retired\" pumpkings Nicholas Clark and Rafael Garcia-Suarez deserve special thanks for\ntheir brilliant and substantive ongoing contributions. Nicholas personally authored over 30%\nof the patches since 5.10.0. Rafael comes in second in patch authorship with 11%, but is\nfirst by a long shot in committing patches authored by others, pushing 44% of the commits\nsince 5.10.0 in this category, often after providing considerable coaching to the patch\nauthors. These statistics in no way comprise all of their contributions, but express in\nshorthand that we couldn't have done it without them.\n\nMany of the changes included in this version originated in the CPAN modules included in\nPerl's core. We're grateful to the entire CPAN community for helping Perl to flourish.\n", "subsections": [] }, "Reporting Bugs": { "content": "If you find what you think is a bug, you might check the articles recently posted to the\ncomp.lang.perl.misc newsgroup and the perl bug database at .\nThere may also be information at , the Perl Home Page.\n\nIf you believe you have an unreported bug, please run the perlbug program included with your\nrelease. Be sure to trim your bug down to a tiny but sufficient test case. Your bug report,\nalong with the output of \"perl -V\", will be sent off to perlbug@perl.org to be analyzed by\nthe Perl porting team.\n\nIf the bug you are reporting has security implications, which make it inappropriate to send\nto a publicly archived mailing list, then please send it to perl5-security-report@perl.org.\nThis points to a closed subscription unarchived mailing list, which includes all the core\ncommitters, who will be able to help assess the impact of issues, figure out a resolution,\nand help co-ordinate the release of patches to mitigate or fix the problem across all\nplatforms on which Perl is supported. Please only use this address for security issues in the\nPerl core, not for modules independently distributed on CPAN.\n", "subsections": [] }, "SEE ALSO": { "content": "The Changes file for an explanation of how to view exhaustive details on what changed.\n\nThe INSTALL file for how to build Perl.\n\nThe README file for general stuff.\n\nThe Artistic and Copying files for copyright information.\n\n for a list of issues found after this release, as\nwell as a list of CPAN modules known to be incompatible with this release.\n\nperl v5.38.2 2026-06-12 PERL5120DELTA(1)", "subsections": [] } }, "summary": "perl5120delta - what is new for perl v5.12.0", "flags": [], "examples": [], "see_also": [] }