{ "mode": "man", "parameter": "perl5180delta", "section": "1", "url": "https://www.chedong.com/phpMan.php/man/perl5180delta/1/json", "generated": "2026-06-13T20:05:02Z", "sections": { "NAME": { "content": "perl5180delta - what is new for perl v5.18.0\n", "subsections": [] }, "DESCRIPTION": { "content": "This document describes differences between the v5.16.0 release and the v5.18.0 release.\n\nIf you are upgrading from an earlier release such as v5.14.0, first read perl5160delta, which\ndescribes differences between v5.14.0 and v5.16.0.\n", "subsections": [ { "name": "Core Enhancements", "content": "" }, { "name": "New mechanism for experimental features", "content": "Newly-added experimental features will now require this incantation:\n\nno warnings \"experimental::featurename\";\nuse feature \"featurename\"; # would warn without the prev line\n\nThere is a new warnings category, called \"experimental\", containing warnings that the feature\npragma emits when enabling experimental features.\n\nNewly-added experimental features will also be given special warning IDs, which consist of\n\"experimental::\" followed by the name of the feature. (The plan is to extend this mechanism\neventually to all warnings, to allow them to be enabled or disabled individually, and not\njust by category.)\n\nBy saying\n\nno warnings \"experimental::featurename\";\n\nyou are taking responsibility for any breakage that future changes to, or removal of, the\nfeature may cause.\n\nSince some features (like \"~~\" or \"my $\") now emit experimental warnings, and you may want\nto disable them in code that is also run on perls that do not recognize these warning\ncategories, consider using the \"if\" pragma like this:\n\nno if $] >= 5.018, warnings => \"experimental::featurename\";\n\nExisting experimental features may begin emitting these warnings, too. Please consult\nperlexperiment for information on which features are considered experimental.\n" }, { "name": "Hash overhaul", "content": "Changes to the implementation of hashes in perl v5.18.0 will be one of the most visible\nchanges to the behavior of existing code.\n\nBy default, two distinct hash variables with identical keys and values may now provide their\ncontents in a different order where it was previously identical.\n\nWhen encountering these changes, the key to cleaning up from them is to accept that hashes\nare unordered collections and to act accordingly.\n\nHash randomization\n\nThe seed used by Perl's hash function is now random. This means that the order which\nkeys/values will be returned from functions like \"keys()\", \"values()\", and \"each()\" will\ndiffer from run to run.\n\nThis change was introduced to make Perl's hashes more robust to algorithmic complexity\nattacks, and also because we discovered that it exposes hash ordering dependency bugs and\nmakes them easier to track down.\n\nToolchain maintainers might want to invest in additional infrastructure to test for things\nlike this. Running tests several times in a row and then comparing results will make it\neasier to spot hash order dependencies in code. Authors are strongly encouraged not to\nexpose the key order of Perl's hashes to insecure audiences.\n\nFurther, every hash has its own iteration order, which should make it much more difficult to\ndetermine what the current hash seed is.\n\nNew hash functions\n\nPerl v5.18 includes support for multiple hash functions, and changed the default (to\nONEATATIMEHARD), you can choose a different algorithm by defining a symbol at compile\ntime. For a current list, consult the INSTALL document. Note that as of Perl v5.18 we can\nonly recommend use of the default or SIPHASH. All the others are known to have security\nissues and are for research purposes only.\n\nPERLHASHSEED environment variable now takes a hex value\n\n\"PERLHASHSEED\" no longer accepts an integer as a parameter; instead the value is expected\nto be a binary value encoded in a hex string, such as \"0xf5867c55039dc724\". This is to make\nthe infrastructure support hash seeds of arbitrary lengths, which might exceed that of an\ninteger. (SipHash uses a 16 byte seed.)\n\nPERLPERTURBKEYS environment variable added\n\nThe \"PERLPERTURBKEYS\" environment variable allows one to control the level of randomization\napplied to \"keys\" and friends.\n\nWhen \"PERLPERTURBKEYS\" is 0, perl will not randomize the key order at all. The chance that\n\"keys\" changes due to an insert will be the same as in previous perls, basically only when\nthe bucket size is changed.\n\nWhen \"PERLPERTURBKEYS\" is 1, perl will randomize keys in a non-repeatable way. The chance\nthat \"keys\" changes due to an insert will be very high. This is the most secure and default\nmode.\n\nWhen \"PERLPERTURBKEYS\" is 2, perl will randomize keys in a repeatable way. Repeated runs\nof the same program should produce the same output every time.\n\n\"PERLHASHSEED\" implies a non-default \"PERLPERTURBKEYS\" setting. Setting\n\"PERLHASHSEED=0\" (exactly one 0) implies \"PERLPERTURBKEYS=0\" (hash key randomization\ndisabled); setting \"PERLHASHSEED\" to any other value implies \"PERLPERTURBKEYS=2\"\n(deterministic and repeatable hash key randomization). Specifying \"PERLPERTURBKEYS\"\nexplicitly to a different level overrides this behavior.\n\nHHaasshh::::UUttiill::::hhaasshhsseeeedd(()) now returns a string\n\nHash::Util::hashseed() now returns a string instead of an integer. This is to make the\ninfrastructure support hash seeds of arbitrary lengths which might exceed that of an integer.\n(SipHash uses a 16 byte seed.)\n\nOutput of PERLHASHSEEDDEBUG has been changed\n\nThe environment variable PERLHASHSEEDDEBUG now makes perl show both the hash function perl\nwas built with, and the seed, in hex, in use for that process. Code parsing this output,\nshould it exist, must change to accommodate the new format. Example of the new format:\n\n$ PERLHASHSEEDDEBUG=1 ./perl -e1\nHASHFUNCTION = MURMUR3 HASHSEED = 0x1476bb9f\n" }, { "name": "Upgrade to Unicode 6.2", "content": "Perl now supports Unicode 6.2. A list of changes from Unicode 6.1 is at\n.\n" }, { "name": "Character name aliases may now include non-Latin1-range characters", "content": "It is possible to define your own names for characters for use in \"\\N{...}\",\n\"charnames::vianame()\", etc. These names can now be comprised of characters from the whole\nUnicode range. This allows for names to be in your native language, and not just English.\nCertain restrictions apply to the characters that may be used (you can't define a name that\nhas punctuation in it, for example). See \"CUSTOM ALIASES\" in charnames.\n" }, { "name": "New DTrace probes", "content": "The following new DTrace probes have been added:\n\n• \"op-entry\"\n\n• \"loading-file\"\n\n• \"loaded-file\"\n\n\"${^LASTFH}\"\nThis new variable provides access to the filehandle that was last read. This is the handle\nused by $. and by \"tell\" and \"eof\" without arguments.\n" }, { "name": "Regular Expression Set Operations", "content": "This is an experimental feature to allow matching against the union, intersection, etc., of\nsets of code points, similar to Unicode::Regex::Set. It can also be used to extend \"/x\"\nprocessing to [bracketed] character classes, and as a replacement of user-defined properties,\nallowing more complex expressions than they do. See \"Extended Bracketed Character Classes\"\nin perlrecharclass.\n" }, { "name": "Lexical subroutines", "content": "This new feature is still considered experimental. To enable it:\n\nuse 5.018;\nno warnings \"experimental::lexicalsubs\";\nuse feature \"lexicalsubs\";\n\nYou can now declare subroutines with \"state sub foo\", \"my sub foo\", and \"our sub foo\".\n(\"state sub\" requires that the \"state\" feature be enabled, unless you write it as\n\"CORE::state sub foo\".)\n\n\"state sub\" creates a subroutine visible within the lexical scope in which it is declared.\nThe subroutine is shared between calls to the outer sub.\n\n\"my sub\" declares a lexical subroutine that is created each time the enclosing block is\nentered. \"state sub\" is generally slightly faster than \"my sub\".\n\n\"our sub\" declares a lexical alias to the package subroutine of the same name.\n\nFor more information, see \"Lexical Subroutines\" in perlsub.\n" }, { "name": "Computed Labels", "content": "The loop controls \"next\", \"last\" and \"redo\", and the special \"dump\" operator, now allow\narbitrary expressions to be used to compute labels at run time. Previously, any argument\nthat was not a constant was treated as the empty string.\n" }, { "name": "More CORE:: subs", "content": "Several more built-in functions have been added as subroutines to the CORE:: namespace -\nnamely, those non-overridable keywords that can be implemented without custom parsers:\n\"defined\", \"delete\", \"exists\", \"glob\", \"pos\", \"prototype\", \"scalar\", \"split\", \"study\", and\n\"undef\".\n\nAs some of these have prototypes, \"prototype('CORE::...')\" has been changed to not make a\ndistinction between overridable and non-overridable keywords. This is to make\n\"prototype('CORE::pos')\" consistent with \"prototype(&CORE::pos)\".\n" }, { "name": "\"kill\" with negative signal names", "content": "\"kill\" has always allowed a negative signal number, which kills the process group instead of\na single process. It has also allowed signal names. But it did not behave consistently,\nbecause negative signal names were treated as 0. Now negative signals names like \"-INT\" are\nsupported and treated the same way as -2 [perl #112990].\n" } ] }, "Security": { "content": "", "subsections": [ { "name": "See also: hash overhaul", "content": "Some of the changes in the hash overhaul were made to enhance security. Please read that\nsection.\n" }, { "name": "\"Storable\" security warning in documentation", "content": "The documentation for \"Storable\" now includes a section which warns readers of the danger of\naccepting Storable documents from untrusted sources. The short version is that deserializing\ncertain types of data can lead to loading modules and other code execution. This is\ndocumented behavior and wanted behavior, but this opens an attack vector for malicious\nentities.\n" }, { "name": "\"Locale::Maketext\" allowed code injection via a malicious template", "content": "If users could provide a translation string to Locale::Maketext, this could be used to invoke\narbitrary Perl subroutines available in the current process.\n\nThis has been fixed, but it is still possible to invoke any method provided by\n\"Locale::Maketext\" itself or a subclass that you are using. One of these methods in turn will\ninvoke the Perl core's \"sprintf\" subroutine.\n\nIn summary, allowing users to provide translation strings without auditing them is a bad\nidea.\n\nThis vulnerability is documented in CVE-2012-6329.\n" }, { "name": "Avoid calling memset with a negative count", "content": "Poorly written perl code that allows an attacker to specify the count to perl's \"x\" string\nrepeat operator can already cause a memory exhaustion denial-of-service attack. A flaw in\nversions of perl before v5.15.5 can escalate that into a heap buffer overrun; coupled with\nversions of glibc before 2.16, it possibly allows the execution of arbitrary code.\n\nThe flaw addressed to this commit has been assigned identifier CVE-2012-5195 and was\nresearched by Tim Brown.\n" }, { "name": "Incompatible Changes", "content": "" }, { "name": "See also: hash overhaul", "content": "Some of the changes in the hash overhaul are not fully compatible with previous versions of\nperl. Please read that section.\n" }, { "name": "An unknown character name in \"\\N{...}\" is now a syntax error", "content": "Previously, it warned, and the Unicode REPLACEMENT CHARACTER was substituted. Unicode now\nrecommends that this situation be a syntax error. Also, the previous behavior led to some\nconfusing warnings and behaviors, and since the REPLACEMENT CHARACTER has no use other than\nas a stand-in for some unknown character, any code that has this problem is buggy.\n" }, { "name": "Formerly deprecated characters in \"\\N{}\" character name aliases are now errors.", "content": "Since v5.12.0, it has been deprecated to use certain characters in user-defined \"\\N{...}\"\ncharacter names. These now cause a syntax error. For example, it is now an error to begin a\nname with a digit, such as in\n\nmy $undraftable = \"\\N{4F}\"; # Syntax error!\n\nor to have commas anywhere in the name. See \"CUSTOM ALIASES\" in charnames.\n" }, { "name": "\"\\N{BELL}\" now refers to U+1F514 instead of U+0007", "content": "Unicode 6.0 reused the name \"BELL\" for a different code point than it traditionally had\nmeant. Since Perl v5.14, use of this name still referred to U+0007, but would raise a\ndeprecation warning. Now, \"BELL\" refers to U+1F514, and the name for U+0007 is \"ALERT\". All\nthe functions in charnames have been correspondingly updated.\n" }, { "name": "New Restrictions in Multi-Character Case-Insensitive Matching in Regular Expression Bracketed", "content": "" }, { "name": "Character Classes", "content": "Unicode has now withdrawn their previous recommendation for regular expressions to\nautomatically handle cases where a single character can match multiple characters case-\ninsensitively, for example, the letter LATIN SMALL LETTER SHARP S and the sequence \"ss\".\nThis is because it turns out to be impracticable to do this correctly in all circumstances.\nBecause Perl has tried to do this as best it can, it will continue to do so. (We are\nconsidering an option to turn it off.) However, a new restriction is being added on such\nmatches when they occur in [bracketed] character classes. People were specifying things such\nas \"/[\\0-\\xff]/i\", and being surprised that it matches the two character sequence \"ss\" (since\nLATIN SMALL LETTER SHARP S occurs in this range). This behavior is also inconsistent with\nusing a property instead of a range: \"\\p{Block=Latin1}\" also includes LATIN SMALL LETTER\nSHARP S, but \"/[\\p{Block=Latin1}]/i\" does not match \"ss\". The new rule is that for there to\nbe a multi-character case-insensitive match within a bracketed character class, the character\nmust be explicitly listed, and not as an end point of a range. This more closely obeys the\nPrinciple of Least Astonishment. See \"Bracketed Character Classes\" in perlrecharclass. Note\nthat a bug [perl #89774], now fixed as part of this change, prevented the previous behavior\nfrom working fully.\n" }, { "name": "Explicit rules for variable names and identifiers", "content": "Due to an oversight, single character variable names in v5.16 were completely unrestricted.\nThis opened the door to several kinds of insanity. As of v5.18, these now follow the rules\nof other identifiers, in addition to accepting characters that match the \"\\p{POSIXPunct}\"\nproperty.\n\nThere is no longer any difference in the parsing of identifiers specified by using braces\nversus without braces. For instance, perl used to allow \"${foo:bar}\" (with a single colon)\nbut not $foo:bar. Now that both are handled by a single code path, they are both treated the\nsame way: both are forbidden. Note that this change is about the range of permissible\nliteral identifiers, not other expressions.\n" }, { "name": "Vertical tabs are now whitespace", "content": "No one could recall why \"\\s\" didn't match \"\\cK\", the vertical tab. Now it does. Given the\nextreme rarity of that character, very little breakage is expected. That said, here's what\nit means:\n\n\"\\s\" in a regex now matches a vertical tab in all circumstances.\n\nLiteral vertical tabs in a regex literal are ignored when the \"/x\" modifier is used.\n\nLeading vertical tabs, alone or mixed with other whitespace, are now ignored when\ninterpreting a string as a number. For example:\n\n$dec = \" \\cK \\t 123\";\n$hex = \" \\cK \\t 0xF\";\n\nsay 0 + $dec; # was 0 with warning, now 123\nsay int $dec; # was 0, now 123\nsay oct $hex; # was 0, now 15\n" }, { "name": "\"/(?{})/\" and \"/(??{})/\" have been heavily reworked", "content": "The implementation of this feature has been almost completely rewritten. Although its main\nintent is to fix bugs, some behaviors, especially related to the scope of lexical variables,\nwill have changed. This is described more fully in the \"Selected Bug Fixes\" section.\n" }, { "name": "Stricter parsing of substitution replacement", "content": "It is no longer possible to abuse the way the parser parses \"s///e\" like this:\n\n%=(,\"Just another \");\n$=\"Perl hacker,\\n\";\ns//}->{/e;print\n\n\"given\" now aliases the global $\nInstead of assigning to an implicit lexical $, \"given\" now makes the global $ an alias for\nits argument, just like \"foreach\". However, it still uses lexical $ if there is lexical $\nin scope (again, just like \"foreach\") [perl #114020].\n" }, { "name": "The smartmatch family of features are now experimental", "content": "Smart match, added in v5.10.0 and significantly revised in v5.10.1, has been a regular point\nof complaint. Although there are a number of ways in which it is useful, it has also proven\nproblematic and confusing for both users and implementors of Perl. There have been a number\nof proposals on how to best address the problem. It is clear that smartmatch is almost\ncertainly either going to change or go away in the future. Relying on its current behavior\nis not recommended.\n\nWarnings will now be issued when the parser sees \"~~\", \"given\", or \"when\". To disable these\nwarnings, you can add this line to the appropriate scope:\n\nno if $] >= 5.018, warnings => \"experimental::smartmatch\";\n\nConsider, though, replacing the use of these features, as they may change behavior again\nbefore becoming stable.\n\nLexical $ is now experimental\nSince it was introduced in Perl v5.10, it has caused much confusion with no obvious solution:\n\n• Various modules (e.g., List::Util) expect callback routines to use the global $. \"use\nList::Util 'first'; my $; first { $ == 1 } @list\" does not work as one would expect.\n\n• A \"my $\" declaration earlier in the same file can cause confusing closure warnings.\n\n• The \"\" subroutine prototype character allows called subroutines to access your lexical\n$, so it is not really private after all.\n\n• Nevertheless, subroutines with a \"(@)\" prototype and methods cannot access the caller's\nlexical $, unless they are written in XS.\n\n• But even XS routines cannot access a lexical $ declared, not in the calling subroutine,\nbut in an outer scope, iff that subroutine happened not to mention $ or use any\noperators that default to $.\n\nIt is our hope that lexical $ can be rehabilitated, but this may cause changes in its\nbehavior. Please use it with caution until it becomes stable.\n" }, { "name": "readline() with \"$/ = \\N\" now reads N characters, not N bytes", "content": "Previously, when reading from a stream with I/O layers such as \"encoding\", the readline()\nfunction, otherwise known as the \"<>\" operator, would read N bytes from the top-most layer.\n[perl #79960]\n\nNow, N characters are read instead.\n\nThere is no change in behaviour when reading from streams with no extra layers, since bytes\nmap exactly to characters.\n" }, { "name": "Overridden \"glob\" is now passed one argument", "content": "\"glob\" overrides used to be passed a magical undocumented second argument that identified the\ncaller. Nothing on CPAN was using this, and it got in the way of a bug fix, so it was\nremoved. If you really need to identify the caller, see Devel::Callsite on CPAN.\n" }, { "name": "Here doc parsing", "content": "The body of a here document inside a quote-like operator now always begins on the line after\nthe \"<\n\nPerl has supported the idiom of swapping $< and $> (and likewise $( and $)) to\ntemporarily drop permissions since 5.0, like this:\n\n($<, $>) = ($>, $<);\n\nHowever, this idiom modifies the real user/group id, which can have undesirable side-\neffects, is no longer useful on any platform perl supports and complicates the\nimplementation of these variables and list assignment in general.\n\nAs an alternative, assignment only to $> is recommended:\n\nlocal $> = $<;\n\nSee also: Setuid Demystified .\n\n• \"microperl\", long broken and of unclear present purpose, will be removed.\n\n• Revamping \"\\Q\" semantics in double-quotish strings when combined with other escapes.\n\nThere are several bugs and inconsistencies involving combinations of \"\\Q\" and escapes\nlike \"\\x\", \"\\L\", etc., within a \"\\Q...\\E\" pair. These need to be fixed, and doing so\nwill necessarily change current behavior. The changes have not yet been settled.\n\n• Use of $x, where \"x\" stands for any actual (non-printing) C0 control character will be\ndisallowed in a future Perl version. Use \"${x}\" instead (where again \"x\" stands for a\ncontrol character), or better, $^A , where \"^\" is a caret (CIRCUMFLEX ACCENT), and \"A\"\nstands for any of the characters listed at the end of \"OPERATOR DIFFERENCES\" in\nperlebcdic.\n" }, { "name": "Performance Enhancements", "content": "• Lists of lexical variable declarations (\"my($x, $y)\") are now optimised down to a single\nop and are hence faster than before.\n\n• A new C preprocessor define \"NOTAINTSUPPORT\" was added that, if set, disables Perl's\ntaint support altogether. Using the -T or -t command line flags will cause a fatal\nerror. Beware that both core tests as well as many a CPAN distribution's tests will fail\nwith this change. On the upside, it provides a small performance benefit due to reduced\nbranching.\n\nDo not enable this unless you know exactly what you are getting yourself into.\n\n• \"pack\" with constant arguments is now constant folded in most cases [perl #113470].\n\n• Speed up in regular expression matching against Unicode properties. The largest gain is\nfor \"\\X\", the Unicode \"extended grapheme cluster.\" The gain for it is about 35% - 40%.\nBracketed character classes, e.g., \"[0-9\\x{100}]\" containing code points above 255 are\nalso now faster.\n\n• On platforms supporting it, several former macros are now implemented as static inline\nfunctions. This should speed things up slightly on non-GCC platforms.\n\n• The optimisation of hashes in boolean context has been extended to affect\n\"scalar(%hash)\", \"%hash ? ... : ...\", and \"sub { %hash || ... }\".\n\n• Filetest operators manage the stack in a fractionally more efficient manner.\n\n• Globs used in a numeric context are now numified directly in most cases, rather than\nbeing numified via stringification.\n\n• The \"x\" repetition operator is now folded to a single constant at compile time if called\nin scalar context with constant operands and no parentheses around the left operand.\n" }, { "name": "Modules and Pragmata", "content": "" }, { "name": "New Modules and Pragmata", "content": "• Config::Perl::V version 0.16 has been added as a dual-lifed module. It provides\nstructured data retrieval of \"perl -V\" output including information only known to the\n\"perl\" binary and not available via Config.\n" }, { "name": "Updated Modules and Pragmata", "content": "For a complete list of updates, run:\n\n$ corelist --diff 5.16.0 5.18.0\n\nYou can substitute your favorite version in place of 5.16.0, too.\n\n• Archive::Extract has been upgraded to 0.68.\n\nWork around an edge case on Linux with Busybox's unzip.\n\n• Archive::Tar has been upgraded to 1.90.\n\nptar now supports the -T option as well as dashless options [rt.cpan.org #75473],\n[rt.cpan.org #75475].\n\nAuto-encode filenames marked as UTF-8 [rt.cpan.org #75474].\n\nDon't use \"tell\" on IO::Zlib handles [rt.cpan.org #64339].\n\nDon't try to \"chown\" on symlinks.\n\n• autodie has been upgraded to 2.13.\n\n\"autodie\" now plays nicely with the 'open' pragma.\n\n• B has been upgraded to 1.42.\n\nThe \"stashoff\" method of COPs has been added. This provides access to an internal field\nadded in perl 5.16 under threaded builds [perl #113034].\n\n\"B::COP::stashpv\" now supports UTF-8 package names and embedded NULs.\n\nAll \"CVf*\" and \"GVf*\" and more SV-related flag values are now provided as constants in\nthe \"B::\" namespace and available for export. The default export list has not changed.\n\nThis makes the module work with the new pad API.\n\n• B::Concise has been upgraded to 0.95.\n\nThe \"-nobanner\" option has been fixed, and \"format\"s can now be dumped. When passed a\nsub name to dump, it will check also to see whether it is the name of a format. If a sub\nand a format share the same name, it will dump both.\n\nThis adds support for the new \"OpMAYBETRUEBOOL\" and \"OPpTRUEBOOL\" flags.\n\n• B::Debug has been upgraded to 1.18.\n\nThis adds support (experimentally) for \"B::PADLIST\", which was added in Perl 5.17.4.\n\n• B::Deparse has been upgraded to 1.20.\n\nAvoid warning when run under \"perl -w\".\n\nIt now deparses loop controls with the correct precedence, and multiple statements in a\n\"format\" line are also now deparsed correctly.\n\nThis release suppresses trailing semicolons in formats.\n\nThis release adds stub deparsing for lexical subroutines.\n\nIt no longer dies when deparsing \"sort\" without arguments. It now correctly omits the\ncomma for \"system $prog @args\" and \"exec $prog @args\".\n\n• bignum, bigint and bigrat have been upgraded to 0.33.\n\nThe overrides for \"hex\" and \"oct\" have been rewritten, eliminating several problems, and\nmaking one incompatible change:\n\n• Formerly, whichever of \"use bigint\" or \"use bigrat\" was compiled later would take\nprecedence over the other, causing \"hex\" and \"oct\" not to respect the other pragma\nwhen in scope.\n\n• Using any of these three pragmata would cause \"hex\" and \"oct\" anywhere else in the\nprogram to evaluate their arguments in list context and prevent them from inferring\n$ when called without arguments.\n\n• Using any of these three pragmata would make \"oct(\"1234\")\" return 1234 (for any\nnumber not beginning with 0) anywhere in the program. Now \"1234\" is translated from\noctal to decimal, whether within the pragma's scope or not.\n\n• The global overrides that facilitate lexical use of \"hex\" and \"oct\" now respect any\nexisting overrides that were in place before the new overrides were installed,\nfalling back to them outside of the scope of \"use bignum\".\n\n• \"use bignum \"hex\"\", \"use bignum \"oct\"\" and similar invocations for bigint and bigrat\nnow export a \"hex\" or \"oct\" function, instead of providing a global override.\n\n• Carp has been upgraded to 1.29.\n\nCarp is no longer confused when \"caller\" returns undef for a package that has been\ndeleted.\n\nThe \"longmess()\" and \"shortmess()\" functions are now documented.\n\n• CGI has been upgraded to 3.63.\n\nUnrecognized HTML escape sequences are now handled better, problematic trailing newlines\nare no longer inserted after
tags by \"startform()\" or \"startform()\", and bogus\n\"Insecure Dependency\" warnings appearing with some versions of perl are now worked\naround.\n\n• Class::Struct has been upgraded to 0.64.\n\nThe constructor now respects overridden accessor methods [perl #29230].\n\n• Compress::Raw::Bzip2 has been upgraded to 2.060.\n\nThe misuse of Perl's \"magic\" API has been fixed.\n\n• Compress::Raw::Zlib has been upgraded to 2.060.\n\nUpgrade bundled zlib to version 1.2.7.\n\nFix build failures on Irix, Solaris, and Win32, and also when building as C++\n[rt.cpan.org #69985], [rt.cpan.org #77030], [rt.cpan.org #75222].\n\nThe misuse of Perl's \"magic\" API has been fixed.\n\n\"compress()\", \"uncompress()\", \"memGzip()\" and \"memGunzip()\" have been speeded up by\nmaking parameter validation more efficient.\n\n• CPAN::Meta::Requirements has been upgraded to 2.122.\n\nTreat undef requirements to \"fromstringhash\" as 0 (with a warning).\n\nAdded \"requirementsformodule\" method.\n\n• CPANPLUS has been upgraded to 0.9135.\n\nAllow adding blib/script to PATH.\n\nSave the history between invocations of the shell.\n\nHandle multiple \"makemakerargs\" and \"makeflags\" arguments better.\n\nThis resolves issues with the SQLite source engine.\n\n• Data::Dumper has been upgraded to 2.145.\n\nIt has been optimized to only build a seen-scalar hash as necessary, thereby speeding up\nserialization drastically.\n\nAdditional tests were added in order to improve statement, branch, condition and\nsubroutine coverage. On the basis of the coverage analysis, some of the internals of\nDumper.pm were refactored. Almost all methods are now documented.\n\n• DBFile has been upgraded to 1.827.\n\nThe main Perl module no longer uses the \"@\" construct.\n\n• Devel::Peek has been upgraded to 1.11.\n\nThis fixes compilation with C++ compilers and makes the module work with the new pad API.\n\n• Digest::MD5 has been upgraded to 2.52.\n\nFix \"Digest::Perl::MD5\" OO fallback [rt.cpan.org #66634].\n\n• Digest::SHA has been upgraded to 5.84.\n\nThis fixes a double-free bug, which might have caused vulnerabilities in some cases.\n\n• DynaLoader has been upgraded to 1.18.\n\nThis is due to a minor code change in the XS for the VMS implementation.\n\nThis fixes warnings about using \"CODE\" sections without an \"OUTPUT\" section.\n\n• Encode has been upgraded to 2.49.\n\nThe Mac alias x-mac-ce has been added, and various bugs have been fixed in\nEncode::Unicode, Encode::UTF7 and Encode::GSM0338.\n\n• Env has been upgraded to 1.04.\n\nIts SPLICE implementation no longer misbehaves in list context.\n\n• ExtUtils::CBuilder has been upgraded to 0.280210.\n\nManifest files are now correctly embedded for those versions of VC++ which make use of\nthem. [perl #111782, #111798].\n\nA list of symbols to export can now be passed to \"link()\" when on Windows, as on other\nOSes [perl #115100].\n\n• ExtUtils::ParseXS has been upgraded to 3.18.\n\nThe generated C code now avoids unnecessarily incrementing \"PLamagicgeneration\" on Perl\nversions where it's done automatically (or on current Perl where the variable no longer\nexists).\n\nThis avoids a bogus warning for initialised XSUB non-parameters [perl #112776].\n\n• File::Copy has been upgraded to 2.26.\n\n\"copy()\" no longer zeros files when copying into the same directory, and also now fails\n(as it has long been documented to do) when attempting to copy a file over itself.\n\n• File::DosGlob has been upgraded to 1.10.\n\nThe internal cache of file names that it keeps for each caller is now freed when that\ncaller is freed. This means \"use File::DosGlob 'glob'; eval 'scalar <*>'\" no longer\nleaks memory.\n\n• File::Fetch has been upgraded to 0.38.\n\nAdded the 'filedefault' option for URLs that do not have a file component.\n\nUse \"File::HomeDir\" when available, and provide \"PERL5CPANPLUSHOME\" to override the\nautodetection.\n\nAlways re-fetch CHECKSUMS if \"fetchdir\" is set.\n\n• File::Find has been upgraded to 1.23.\n\nThis fixes inconsistent unixy path handling on VMS.\n\nIndividual files may now appear in list of directories to be searched [perl #59750].\n\n• File::Glob has been upgraded to 1.20.\n\nFile::Glob has had exactly the same fix as File::DosGlob. Since it is what Perl's own\n\"glob\" operator itself uses (except on VMS), this means \"eval 'scalar <*>'\" no longer\nleaks.\n\nA space-separated list of patterns return long lists of results no longer results in\nmemory corruption or crashes. This bug was introduced in Perl 5.16.0. [perl #114984]\n\n• File::Spec::Unix has been upgraded to 3.40.\n\n\"abs2rel\" could produce incorrect results when given two relative paths or the root\ndirectory twice [perl #111510].\n\n• File::stat has been upgraded to 1.07.\n\n\"File::stat\" ignores the filetest pragma, and warns when used in combination therewith.\nBut it was not warning for \"-r\". This has been fixed [perl #111640].\n\n\"-p\" now works, and does not return false for pipes [perl #111638].\n\nPreviously \"File::stat\"'s overloaded \"-x\" and \"-X\" operators did not give the correct\nresults for directories or executable files when running as root. They had been treating\nexecutable permissions for root just like for any other user, performing group membership\ntests etc for files not owned by root. They now follow the correct Unix behaviour - for a\ndirectory they are always true, and for a file if any of the three execute permission\nbits are set then they report that root can execute the file. Perl's builtin \"-x\" and\n\"-X\" operators have always been correct.\n\n• File::Temp has been upgraded to 0.23\n\nFixes various bugs involving directory removal. Defers unlinking tempfiles if the\ninitial unlink fails, which fixes problems on NFS.\n\n• GDBMFile has been upgraded to 1.15.\n\nThe undocumented optional fifth parameter to \"TIEHASH\" has been removed. This was\nintended to provide control of the callback used by \"gdbm*\" functions in case of fatal\nerrors (such as filesystem problems), but did not work (and could never have worked). No\ncode on CPAN even attempted to use it. The callback is now always the previous default,\n\"croak\". Problems on some platforms with how the \"C\" \"croak\" function is called have also\nbeen resolved.\n\n• Hash::Util has been upgraded to 0.15.\n\n\"hashunlocked\" and \"hashrefunlocked\" now returns true if the hash is unlocked, instead\nof always returning false [perl #112126].\n\n\"hashunlocked\", \"hashrefunlocked\", \"lockhashrecurse\" and \"unlockhashrecurse\" are\nnow exportable [perl #112126].\n\nTwo new functions, \"hashlocked\" and \"hashreflocked\", have been added. Oddly enough,\nthese two functions were already exported, even though they did not exist [perl #112126].\n\n• HTTP::Tiny has been upgraded to 0.025.\n\nAdd SSL verification features [github #6], [github #9].\n\nInclude the final URL in the response hashref.\n\nAdd \"localaddress\" option.\n\nThis improves SSL support.\n\n• IO has been upgraded to 1.28.\n\n\"sync()\" can now be called on read-only file handles [perl #64772].\n\nIO::Socket tries harder to cache or otherwise fetch socket information.\n\n• IPC::Cmd has been upgraded to 0.80.\n\nUse \"POSIX::exit\" instead of \"exit\" in \"runforked\" [rt.cpan.org #76901].\n\n• IPC::Open3 has been upgraded to 1.13.\n\nThe \"open3()\" function no longer uses \"POSIX::close()\" to close file descriptors since\nthat breaks the ref-counting of file descriptors done by PerlIO in cases where the file\ndescriptors are shared by PerlIO streams, leading to attempts to close the file\ndescriptors a second time when any such PerlIO streams are closed later on.\n\n• Locale::Codes has been upgraded to 3.25.\n\nIt includes some new codes.\n\n• Memoize has been upgraded to 1.03.\n\nFix the \"MERGE\" cache option.\n\n• Module::Build has been upgraded to 0.4003.\n\nFixed bug where modules without $VERSION might have a version of '0' listed in 'provides'\nmetadata, which will be rejected by PAUSE.\n\nFixed bug in PodParser to allow numerals in module names.\n\nFixed bug where giving arguments twice led to them becoming arrays, resulting in install\npaths like AARRRRAAYY(0xdeadbeef)/lib/Foo.pm.\n\nA minor bug fix allows markup to be used around the leading \"Name\" in a POD \"abstract\"\nline, and some documentation improvements have been made.\n\n• Module::CoreList has been upgraded to 2.90\n\nVersion information is now stored as a delta, which greatly reduces the size of the\nCoreList.pm file.\n\nThis restores compatibility with older versions of perl and cleans up the corelist data\nfor various modules.\n\n• Module::Load::Conditional has been upgraded to 0.54.\n\nFix use of \"requires\" on perls installed to a path with spaces.\n\nVarious enhancements include the new use of Module::Metadata.\n\n• Module::Metadata has been upgraded to 1.000011.\n\nThe creation of a Module::Metadata object for a typical module file has been sped up by\nabout 40%, and some spurious warnings about $VERSIONs have been suppressed.\n\n• Module::Pluggable has been upgraded to 4.7.\n\nAmongst other changes, triggers are now allowed on events, which gives a powerful way to\nmodify behaviour.\n\n• Net::Ping has been upgraded to 2.41.\n\nThis fixes some test failures on Windows.\n\n• Opcode has been upgraded to 1.25.\n\nReflect the removal of the boolkeys opcode and the addition of the clonecv, introcv and\npadcv opcodes.\n\n• overload has been upgraded to 1.22.\n\n\"no overload\" now warns for invalid arguments, just like \"use overload\".\n\n• PerlIO::encoding has been upgraded to 0.16.\n\nThis is the module implementing the \":encoding(...)\" I/O layer. It no longer corrupts\nmemory or crashes when the encoding back-end reallocates the buffer or gives it a\ntypeglob or shared hash key scalar.\n\n• PerlIO::scalar has been upgraded to 0.16.\n\nThe buffer scalar supplied may now only contain code points 0xFF or lower. [perl #109828]\n\n• Perl::OSType has been upgraded to 1.003.\n\nThis fixes a bug detecting the VOS operating system.\n\n• Pod::Html has been upgraded to 1.18.\n\nThe option \"--libpods\" has been reinstated. It is deprecated, and its use does nothing\nother than issue a warning that it is no longer supported.\n\nSince the HTML files generated by pod2html claim to have a UTF-8 charset, actually write\nthe files out using UTF-8 [perl #111446].\n\n• Pod::Simple has been upgraded to 3.28.\n\nNumerous improvements have been made, mostly to Pod::Simple::XHTML, which also has a\ncompatibility change: the \"codesinverbatim\" option is now disabled by default. See\ncpan/Pod-Simple/ChangeLog for the full details.\n\n• re has been upgraded to 0.23\n\nSingle character [class]es like \"/[s]/\" or \"/[s]/i\" are now optimized as if they did not\nhave the brackets, i.e. \"/s/\" or \"/s/i\".\n\nSee note about \"opcomp\" in the \"Internal Changes\" section below.\n\n• Safe has been upgraded to 2.35.\n\nFix interactions with \"Devel::Cover\".\n\nDon't eval code under \"no strict\".\n\n• Scalar::Util has been upgraded to version 1.27.\n\nFix an overloading issue with \"sum\".\n\n\"first\" and \"reduce\" now check the callback first (so &first(1) is disallowed).\n\nFix \"tainted\" on magical values [rt.cpan.org #55763].\n\nFix \"sum\" on previously magical values [rt.cpan.org #61118].\n\nFix reading past the end of a fixed buffer [rt.cpan.org #72700].\n\n• Search::Dict has been upgraded to 1.07.\n\nNo longer require \"stat\" on filehandles.\n\nUse \"fc\" for casefolding.\n\n• Socket has been upgraded to 2.009.\n\nConstants and functions required for IP multicast source group membership have been\nadded.\n\n\"unpacksockaddrin()\" and \"unpacksockaddrin6()\" now return just the IP address in\nscalar context, and \"inetntop()\" now guards against incorrect length scalars being\npassed in.\n\nThis fixes an uninitialized memory read.\n\n• Storable has been upgraded to 2.41.\n\nModifying $[0] within \"STORABLEfreeze\" no longer results in crashes [perl #112358].\n\nAn object whose class implements \"STORABLEattach\" is now thawed only once when there are\nmultiple references to it in the structure being thawed [perl #111918].\n\nRestricted hashes were not always thawed correctly [perl #73972].\n\nStorable would croak when freezing a blessed REF object with a \"STORABLEfreeze()\" method\n[perl #113880].\n\nIt can now freeze and thaw vstrings correctly. This causes a slight incompatible change\nin the storage format, so the format version has increased to 2.9.\n\nThis contains various bugfixes, including compatibility fixes for older versions of Perl\nand vstring handling.\n\n• Sys::Syslog has been upgraded to 0.32.\n\nThis contains several bug fixes relating to \"getservbyname()\", \"setlogsock()\"and log\nlevels in \"syslog()\", together with fixes for Windows, Haiku-OS and GNU/kFreeBSD. See\ncpan/Sys-Syslog/Changes for the full details.\n\n• Term::ANSIColor has been upgraded to 4.02.\n\nAdd support for italics.\n\nImprove error handling.\n\n• Term::ReadLine has been upgraded to 1.10. This fixes the use of the cpan and cpanp\nshells on Windows in the event that the current drive happens to contain a \\dev\\tty file.\n\n• Test::Harness has been upgraded to 3.26.\n\nFix glob semantics on Win32 [rt.cpan.org #49732].\n\nDon't use \"Win32::GetShortPathName\" when calling perl [rt.cpan.org #47890].\n\nIgnore -T when reading shebang [rt.cpan.org #64404].\n\nHandle the case where we don't know the wait status of the test more gracefully.\n\nMake the test summary 'ok' line overridable so that it can be changed to a plugin to make\nthe output of prove idempotent.\n\nDon't run world-writable files.\n\n• Text::Tabs and Text::Wrap have been upgraded to 2012.0818. Support for Unicode combining\ncharacters has been added to them both.\n\n• threads::shared has been upgraded to 1.31.\n\nThis adds the option to warn about or ignore attempts to clone structures that can't be\ncloned, as opposed to just unconditionally dying in that case.\n\nThis adds support for dual-valued values as created by Scalar::Util::dualvar.\n\n• Tie::StdHandle has been upgraded to 4.3.\n\n\"READ\" now respects the offset argument to \"read\" [perl #112826].\n\n• Time::Local has been upgraded to 1.2300.\n\nSeconds values greater than 59 but less than 60 no longer cause \"timegm()\" and\n\"timelocal()\" to croak.\n\n• Unicode::UCD has been upgraded to 0.53.\n\nThis adds a function allcasefolds() that returns all the casefolds.\n\n• Win32 has been upgraded to 0.47.\n\nNew APIs have been added for getting and setting the current code page.\n" }, { "name": "Removed Modules and Pragmata", "content": "• Version::Requirements has been removed from the core distribution. It is available under\na different name: CPAN::Meta::Requirements.\n" } ] }, "Documentation": { "content": "", "subsections": [ { "name": "Changes to Existing Documentation", "content": "perlcheat\n\n• perlcheat has been reorganized, and a few new sections were added.\n\nperldata\n\n• Now explicitly documents the behaviour of hash initializer lists that contain duplicate\nkeys.\n\nperldiag\n\n• The explanation of symbolic references being prevented by \"strict refs\" now doesn't\nassume that the reader knows what symbolic references are.\n\nperlfaq\n\n• perlfaq has been synchronized with version 5.0150040 from CPAN.\n\nperlfunc\n\n• The return value of \"pipe\" is now documented.\n\n• Clarified documentation of \"our\".\n\nperlop\n\n• Loop control verbs (\"dump\", \"goto\", \"next\", \"last\" and \"redo\") have always had the same\nprecedence as assignment operators, but this was not documented until now.\n\nDiagnostics\n\nThe following additions or changes have been made to diagnostic output, including warnings\nand fatal error messages. For the complete list of diagnostic messages, see perldiag.\n" }, { "name": "New Diagnostics", "content": "New Errors\n\n• Unterminated delimiter for here document\n\nThis message now occurs when a here document label has an initial quotation mark but the\nfinal quotation mark is missing.\n\nThis replaces a bogus and misleading error message about not finding the label itself\n[perl #114104].\n\n• panic: child pseudo-process was never scheduled\n\nThis error is thrown when a child pseudo-process in the ithreads implementation on\nWindows was not scheduled within the time period allowed and therefore was not able to\ninitialize properly [perl #88840].\n\n• Group name must start with a non-digit word character in regex; marked by <-- HERE in\nm/%s/\n\nThis error has been added for \"(?&0)\", which is invalid. It used to produce an\nincomprehensible error message [perl #101666].\n\n• Can't use an undefined value as a subroutine reference\n\nCalling an undefined value as a subroutine now produces this error message. It used to,\nbut was accidentally disabled, first in Perl 5.004 for non-magical variables, and then in\nPerl v5.14 for magical (e.g., tied) variables. It has now been restored. In the mean\ntime, undef was treated as an empty string [perl #113576].\n\n• Experimental \"%s\" subs not enabled\n\nTo use lexical subs, you must first enable them:\n\nno warnings 'experimental::lexicalsubs';\nuse feature 'lexicalsubs';\nmy sub foo { ... }\n\nNew Warnings\n\n• 'Strings with code points over 0xFF may not be mapped into in-memory file handles'\n\n• '%s' resolved to '\\o{%s}%d'\n\n• 'Trailing white-space in a charnames alias definition is deprecated'\n\n• 'A sequence of multiple spaces in a charnames alias definition is deprecated'\n\n• 'Passing malformed UTF-8 to \"%s\" is deprecated'\n\n• Subroutine \"&%s\" is not available\n\n(W closure) During compilation, an inner named subroutine or eval is attempting to\ncapture an outer lexical subroutine that is not currently available. This can happen for\none of two reasons. First, the lexical subroutine may be declared in an outer anonymous\nsubroutine that has not yet been created. (Remember that named subs are created at\ncompile time, while anonymous subs are created at run-time.) For example,\n\nsub { my sub a {...} sub f { \\&a } }\n\nAt the time that f is created, it can't capture the current the \"a\" sub, since the\nanonymous subroutine hasn't been created yet. Conversely, the following won't give a\nwarning since the anonymous subroutine has by now been created and is live:\n\nsub { my sub a {...} eval 'sub f { \\&a }' }->();\n\nThe second situation is caused by an eval accessing a variable that has gone out of\nscope, for example,\n\nsub f {\nmy sub a {...}\nsub { eval '\\&a' }\n}\nf()->();\n\nHere, when the '\\&a' in the eval is being compiled, f() is not currently being executed,\nso its &a is not available for capture.\n\n• \"%s\" subroutine &%s masks earlier declaration in same %s\n\n(W misc) A \"my\" or \"state\" subroutine has been redeclared in the current scope or\nstatement, effectively eliminating all access to the previous instance. This is almost\nalways a typographical error. Note that the earlier subroutine will still exist until\nthe end of the scope or until all closure references to it are destroyed.\n\n• The %s feature is experimental\n\n(S experimental) This warning is emitted if you enable an experimental feature via \"use\nfeature\". Simply suppress the warning if you want to use the feature, but know that in\ndoing so you are taking the risk of using an experimental feature which may change or be\nremoved in a future Perl version:\n\nno warnings \"experimental::lexicalsubs\";\nuse feature \"lexicalsubs\";\n\n• sleep(%u) too large\n\n(W overflow) You called \"sleep\" with a number that was larger than it can reliably handle\nand \"sleep\" probably slept for less time than requested.\n\n• Wide character in setenv\n\nAttempts to put wide characters into environment variables via %ENV now provoke this\nwarning.\n\n• \"Invalid negative number (%s) in chr\"\n\n\"chr()\" now warns when passed a negative value [perl #83048].\n\n• \"Integer overflow in srand\"\n\n\"srand()\" now warns when passed a value that doesn't fit in a \"UV\" (since the value will\nbe truncated rather than overflowing) [perl #40605].\n\n• \"-i used with no filenames on the command line, reading from STDIN\"\n\nRunning perl with the \"-i\" flag now warns if no input files are provided on the command\nline [perl #113410].\n" }, { "name": "Changes to Existing Diagnostics", "content": "• $* is no longer supported\n\nThe warning that use of $* and $# is no longer supported is now generated for every\nlocation that references them. Previously it would fail to be generated if another\nvariable using the same typeglob was seen first (e.g. \"@*\" before $*), and would not be\ngenerated for the second and subsequent uses. (It's hard to fix the failure to generate\nwarnings at all without also generating them every time, and warning every time is\nconsistent with the warnings that $[ used to generate.)\n\n• The warnings for \"\\b{\" and \"\\B{\" were added. They are a deprecation warning which should\nbe turned off by that category. One should not have to turn off regular regexp warnings\nas well to get rid of these.\n\n• Constant(%s): Call to &{$^H{%s}} did not return a defined value\n\nConstant overloading that returns \"undef\" results in this error message. For numeric\nconstants, it used to say \"Constant(undef)\". \"undef\" has been replaced with the number\nitself.\n\n• The error produced when a module cannot be loaded now includes a hint that the module may\nneed to be installed: \"Can't locate hopping.pm in @INC (you may need to install the\nhopping module) (@INC contains: ...)\"\n\n• vector argument not supported with alpha versions\n\nThis warning was not suppressible, even with \"no warnings\". Now it is suppressible, and\nhas been moved from the \"internal\" category to the \"printf\" category.\n\n• \"Can't do {n,m} with n > m in regex; marked by <-- HERE in m/%s/\"\n\nThis fatal error has been turned into a warning that reads:\n\nQuantifier {n,m} with n > m can't match in regex\n\n(W regexp) Minima should be less than or equal to maxima. If you really want your regexp\nto match something 0 times, just put {0}.\n\n• The \"Runaway prototype\" warning that occurs in bizarre cases has been removed as being\nunhelpful and inconsistent.\n\n• The \"Not a format reference\" error has been removed, as the only case in which it could\nbe triggered was a bug.\n\n• The \"Unable to create sub named %s\" error has been removed for the same reason.\n\n• The 'Can't use \"my %s\" in sort comparison' error has been downgraded to a warning, '\"my\n%s\" used in sort comparison' (with 'state' instead of 'my' for state variables). In\naddition, the heuristics for guessing whether lexical $a or $b has been misused have been\nimproved to generate fewer false positives. Lexical $a and $b are no longer disallowed\nif they are outside the sort block. Also, a named unary or list operator inside the sort\nblock no longer causes the $a or $b to be ignored [perl #86136].\n" }, { "name": "Utility Changes", "content": "h2xs\n\n• h2xs no longer produces invalid code for empty defines. [perl #20636]\n" }, { "name": "Configuration and Compilation", "content": "• Added \"useversionedarchname\" option to Configure\n\nWhen set, it includes 'apiversionstring' in 'archname'. E.g.\nx8664-linux-5.13.6-thread-multi. It is unset by default.\n\nThis feature was requested by Tim Bunce, who observed that \"INSTALLBASE\" creates a\nlibrary structure that does not differentiate by perl version. Instead, it places\narchitecture specific files in \"$installbase/lib/perl5/$archname\". This makes it\ndifficult to use a common \"INSTALLBASE\" library path with multiple versions of perl.\n\nBy setting \"-Duseversionedarchname\", the $archname will be distinct for architecture and\nAPI version, allowing mixed use of \"INSTALLBASE\".\n\n• Add a \"PERLNOINLINEFUNCTIONS\" option\n\nIf \"PERLNOINLINEFUNCTIONS\" is defined, don't include \"inline.h\"\n\nThis permits test code to include the perl headers for definitions without creating a\nlink dependency on the perl library (which may not exist yet).\n\n• Configure will honour the external \"MAILDOMAIN\" environment variable, if set.\n\n• \"installman\" no longer ignores the silent option\n\n• Both \"META.yml\" and \"META.json\" files are now included in the distribution.\n\n• Configure will now correctly detect \"isblank()\" when compiling with a C++ compiler.\n\n• The pager detection in Configure has been improved to allow responses which specify\noptions after the program name, e.g. /usr/bin/less -R, if the user accepts the default\nvalue. This helps perldoc when handling ANSI escapes [perl #72156].\n" } ] }, "Testing": { "content": "• The test suite now has a section for tests that require very large amounts of memory.\nThese tests won't run by default; they can be enabled by setting the \"PERLTESTMEMORY\"\nenvironment variable to the number of gibibytes of memory that may be safely used.\n", "subsections": [ { "name": "Platform Support", "content": "" }, { "name": "Discontinued Platforms", "content": "BeOS\nBeOS was an operating system for personal computers developed by Be Inc, initially for\ntheir BeBox hardware. The OS Haiku was written as an open source replacement\nfor/continuation of BeOS, and its perl port is current and actively maintained.\n\nUTS Global\nSupport code relating to UTS global has been removed. UTS was a mainframe version of\nSystem V created by Amdahl, subsequently sold to UTS Global. The port has not been\ntouched since before Perl v5.8.0, and UTS Global is now defunct.\n\nVM/ESA\nSupport for VM/ESA has been removed. The port was tested on 2.3.0, which IBM ended\nservice on in March 2002. 2.4.0 ended service in June 2003, and was superseded by Z/VM.\nThe current version of Z/VM is V6.2.0, and scheduled for end of service on 2015/04/30.\n\nMPE/IX\nSupport for MPE/IX has been removed.\n\nEPOC\nSupport code relating to EPOC has been removed. EPOC was a family of operating systems\ndeveloped by Psion for mobile devices. It was the predecessor of Symbian. The port was\nlast updated in April 2002.\n\nRhapsody\nSupport for Rhapsody has been removed.\n" }, { "name": "Platform-Specific Notes", "content": "AIX\n\nConfigure now always adds \"-qlanglvl=extc99\" to the CC flags on AIX when using xlC. This\nwill make it easier to compile a number of XS-based modules that assume C99 [perl #113778].\n\nclang++\n\nThere is now a workaround for a compiler bug that prevented compiling with clang++ since Perl\nv5.15.7 [perl #112786].\n\nC++\n\nWhen compiling the Perl core as C++ (which is only semi-supported), the mathom functions are\nnow compiled as \"extern \"C\"\", to ensure proper binary compatibility. (However, binary\ncompatibility isn't generally guaranteed anyway in the situations where this would matter.)\n\nDarwin\n\nStop hardcoding an alignment on 8 byte boundaries to fix builds using -Dusemorebits.\n\nHaiku\n\nPerl should now work out of the box on Haiku R1 Alpha 4.\n\nMidnightBSD\n\n\"libcr\" was removed from recent versions of MidnightBSD and older versions work better with\n\"pthread\". Threading is now enabled using \"pthread\" which corrects build errors with\nthreading enabled on 0.4-CURRENT.\n\nSolaris\n\nIn Configure, avoid running sed commands with flags not supported on Solaris.\n\nVMS\n\n• Where possible, the case of filenames and command-line arguments is now preserved by\nenabling the CRTL features \"DECC$EFSCASEPRESERVE\" and \"DECC$ARGVPARSESTYLE\" at start-\nup time. The latter only takes effect when extended parse is enabled in the process from\nwhich Perl is run.\n\n• The character set for Extended Filename Syntax (EFS) is now enabled by default on VMS.\nAmong other things, this provides better handling of dots in directory names, multiple\ndots in filenames, and spaces in filenames. To obtain the old behavior, set the logical\nname \"DECC$EFSCHARSET\" to \"DISABLE\".\n\n• Fixed linking on builds configured with \"-Dusemymalloc=y\".\n\n• Experimental support for building Perl with the HP C++ compiler is available by\nconfiguring with \"-Dusecxx\".\n\n• All C header files from the top-level directory of the distribution are now installed on\nVMS, providing consistency with a long-standing practice on other platforms. Previously\nonly a subset were installed, which broke non-core extension builds for extensions that\ndepended on the missing include files.\n\n• Quotes are now removed from the command verb (but not the parameters) for commands\nspawned via \"system\", backticks, or a piped \"open\". Previously, quotes on the verb were\npassed through to DCL, which would fail to recognize the command. Also, if the verb is\nactually a path to an image or command procedure on an ODS-5 volume, quoting it now\nallows the path to contain spaces.\n\n• The a2p build has been fixed for the HP C++ compiler on OpenVMS.\n\nWin32\n\n• Perl can now be built using Microsoft's Visual C++ 2012 compiler by specifying\nCCTYPE=MSVC110 (or MSVC110FREE if you are using the free Express edition for Windows\nDesktop) in win32/Makefile.\n\n• The option to build without \"USESOCKETSASHANDLES\" has been removed.\n\n• Fixed a problem where perl could crash while cleaning up threads (including the main\nthread) in threaded debugging builds on Win32 and possibly other platforms [perl\n#114496].\n\n• A rare race condition that would lead to sleep taking more time than requested, and\npossibly even hanging, has been fixed [perl #33096].\n\n• \"link\" on Win32 now attempts to set $! to more appropriate values based on the Win32 API\nerror code. [perl #112272]\n\nPerl no longer mangles the environment block, e.g. when launching a new sub-process, when\nthe environment contains non-ASCII characters. Known problems still remain, however, when\nthe environment contains characters outside of the current ANSI codepage (e.g. see the\nitem about Unicode in %ENV in\n). [perl #113536]\n\n• Building perl with some Windows compilers used to fail due to a problem with miniperl's\n\"glob\" operator (which uses the \"perlglob\" program) deleting the PATH environment\nvariable [perl #113798].\n\n• A new makefile option, \"USE64BITINT\", has been added to the Windows makefiles. Set\nthis to \"define\" when building a 32-bit perl if you want it to use 64-bit integers.\n\nMachine code size reductions, already made to the DLLs of XS modules in Perl v5.17.2,\nhave now been extended to the perl DLL itself.\n\nBuilding with VC++ 6.0 was inadvertently broken in Perl v5.17.2 but has now been fixed\nagain.\n\nWinCE\n\nBuilding on WinCE is now possible once again, although more work is required to fully restore\na clean build.\n" }, { "name": "Internal Changes", "content": "• Synonyms for the misleadingly named \"avlen()\" have been created: \"avtopindex()\" and\n\"avtindex\". All three of these return the number of the highest index in the array, not\nthe number of elements it contains.\n\n• SvUPGRADE() is no longer an expression. Originally this macro (and its underlying\nfunction, svupgrade()) were documented as boolean, although in reality they always\ncroaked on error and never returned false. In 2005 the documentation was updated to\nspecify a void return value, but SvUPGRADE() was left always returning 1 for backwards\ncompatibility. This has now been removed, and SvUPGRADE() is now a statement with no\nreturn value.\n\nSo this is now a syntax error:\n\nif (!SvUPGRADE(sv)) { croak(...); }\n\nIf you have code like that, simply replace it with\n\nSvUPGRADE(sv);\n\nor to avoid compiler warnings with older perls, possibly\n\n(void)SvUPGRADE(sv);\n\n• Perl has a new copy-on-write mechanism that allows any SvPOK scalar to be upgraded to a\ncopy-on-write scalar. A reference count on the string buffer is stored in the string\nbuffer itself. This feature is not enabled by default.\n\nIt can be enabled in a perl build by running Configure with\n-Accflags=-DPERLNEWCOPYONWRITE, and we would encourage XS authors to try their code\nwith such an enabled perl, and provide feedback. Unfortunately, there is not yet a good\nguide to updating XS code to cope with COW. Until such a document is available, consult\nthe perl5-porters mailing list.\n\nIt breaks a few XS modules by allowing copy-on-write scalars to go through code paths\nthat never encountered them before.\n\n• Copy-on-write no longer uses the SvFAKE and SvREADONLY flags. Hence, SvREADONLY\nindicates a true read-only SV.\n\nUse the SvIsCOW macro (as before) to identify a copy-on-write scalar.\n\n• \"PLglobindex\" is gone.\n\n• The private Perlcroaknomodify has had its context parameter removed. It is now has a\nvoid prototype. Users of the public API croaknomodify remain unaffected.\n\n• Copy-on-write (shared hash key) scalars are no longer marked read-only. \"SvREADONLY\"\nreturns false on such an SV, but \"SvIsCOW\" still returns true.\n\n• A new op type, \"OPPADRANGE\" has been introduced. The perl peephole optimiser will,\nwhere possible, substitute a single padrange op for a pushmark followed by one or more\npad ops, and possibly also skipping list and nextstate ops. In addition, the op can\ncarry out the tasks associated with the RHS of a \"my(...) = @\" assignment, so those ops\nmay be optimised away too.\n\n• Case-insensitive matching inside a [bracketed] character class with a multi-character\nfold no longer excludes one of the possibilities in the circumstances that it used to.\n[perl #89774].\n\n• \"PLformfeed\" has been removed.\n\n• The regular expression engine no longer reads one byte past the end of the target string.\nWhile for all internally well-formed scalars this should never have been a problem, this\nchange facilitates clever tricks with string buffers in CPAN modules. [perl #73542]\n\n• Inside a BEGIN block, \"PLcompcv\" now points to the currently-compiling subroutine,\nrather than the BEGIN block itself.\n\n• \"mglength\" has been deprecated.\n\n• \"svlen\" now always returns a byte count and \"svlenutf8\" a character count.\nPreviously, \"svlen\" and \"svlenutf8\" were both buggy and would sometimes returns bytes\nand sometimes characters. \"svlenutf8\" no longer assumes that its argument is in UTF-8.\nNeither of these creates UTF-8 caches for tied or overloaded values or for non-PVs any\nmore.\n\n• \"svmortalcopy\" now copies string buffers of shared hash key scalars when called from XS\nmodules [perl #79824].\n\n• The new \"RXfMODIFIESVARS\" flag can be set by custom regular expression engines to\nindicate that the execution of the regular expression may cause variables to be modified.\nThis lets \"s///\" know to skip certain optimisations. Perl's own regular expression\nengine sets this flag for the special backtracking verbs that set $REGMARK and $REGERROR.\n\n• The APIs for accessing lexical pads have changed considerably.\n\n\"PADLIST\"s are now longer \"AV\"s, but their own type instead. \"PADLIST\"s now contain a\n\"PAD\" and a \"PADNAMELIST\" of \"PADNAME\"s, rather than \"AV\"s for the pad and the list of\npad names. \"PAD\"s, \"PADNAMELIST\"s, and \"PADNAME\"s are to be accessed as such through the\nnewly added pad API instead of the plain \"AV\" and \"SV\" APIs. See perlapi for details.\n\n• In the regex API, the numbered capture callbacks are passed an index indicating what\nmatch variable is being accessed. There are special index values for the \"$`, $&, $&\"\nvariables. Previously the same three values were used to retrieve \"${^PREMATCH},\n${^MATCH}, ${^POSTMATCH}\" too, but these have now been assigned three separate values.\nSee \"Numbered capture callbacks\" in perlreapi.\n\n• \"PLsawampersand\" was previously a boolean indicating that any of \"$`, $&, $&\" had been\nseen; it now contains three one-bit flags indicating the presence of each of the\nvariables individually.\n\n• The \"CV *\" typemap entry now supports \"&{}\" overloading and typeglobs, just like \"&{...}\"\n[perl #96872].\n\n• The \"SVfAMAGIC\" flag to indicate overloading is now on the stash, not the object. It is\nnow set automatically whenever a method or @ISA changes, so its meaning has changed, too.\nIt now means \"potentially overloaded\". When the overload table is calculated, the flag\nis automatically turned off if there is no overloading, so there should be no noticeable\nslowdown.\n\nThe staleness of the overload tables is now checked when overload methods are invoked,\nrather than during \"bless\".\n\n\"A\" magic is gone. The changes to the handling of the \"SVfAMAGIC\" flag eliminate the\nneed for it.\n\n\"PLamagicgeneration\" has been removed as no longer necessary. For XS modules, it is\nnow a macro alias to \"PLna\".\n\nThe fallback overload setting is now stored in a stash entry separate from overloadedness\nitself.\n\n• The character-processing code has been cleaned up in places. The changes should be\noperationally invisible.\n\n• The \"study\" function was made a no-op in v5.16. It was simply disabled via a \"return\"\nstatement; the code was left in place. Now the code supporting what \"study\" used to do\nhas been removed.\n\n• Under threaded perls, there is no longer a separate PV allocated for every COP to store\nits package name (\"cop->stashpv\"). Instead, there is an offset (\"cop->stashoff\") into\nthe new \"PLstashpad\" array, which holds stash pointers.\n\n• In the pluggable regex API, the \"regexpengine\" struct has acquired a new field\n\"opcomp\", which is currently just for perl's internal use, and should be initialized to\nNULL by other regex plugin modules.\n\n• A new function \"alloccopstash\" has been added to the API, but is considered experimental.\nSee perlapi.\n\n• Perl used to implement get magic in a way that would sometimes hide bugs in code that\ncould call mgget() too many times on magical values. This hiding of errors no longer\noccurs, so long-standing bugs may become visible now. If you see magic-related errors in\nXS code, check to make sure it, together with the Perl API functions it uses, calls\nmgget() only once on SvGMAGICAL() values.\n\n• OP allocation for CVs now uses a slab allocator. This simplifies memory management for\nOPs allocated to a CV, so cleaning up after a compilation error is simpler and safer\n[perl #111462][perl #112312].\n\n• \"PERLDEBUGREADONLYOPS\" has been rewritten to work with the new slab allocator,\nallowing it to catch more violations than before.\n\n• The old slab allocator for ops, which was only enabled for \"PERLIMPLICITSYS\" and\n\"PERLDEBUGREADONLYOPS\", has been retired.\n" }, { "name": "Selected Bug Fixes", "content": "• Here document terminators no longer require a terminating newline character when they\noccur at the end of a file. This was already the case at the end of a string eval [perl\n#65838].\n\n• \"-DPERLGLOBALSTRUCT\" builds now free the global struct after they've finished using it.\n\n• A trailing '/' on a path in @INC will no longer have an additional '/' appended.\n\n• The \":crlf\" layer now works when unread data doesn't fit into its own buffer. [perl\n#112244].\n\n• \"ungetc()\" now handles UTF-8 encoded data. [perl #116322].\n\n• A bug in the core typemap caused any C types that map to the TBOOL core typemap entry to\nnot be set, updated, or modified when the TBOOL variable was used in an OUTPUT: section\nwith an exception for RETVAL. TBOOL in an INPUT: section was not affected. Using a\nTBOOL return type for an XSUB (RETVAL) was not affected. A side effect of fixing this\nbug is, if a TBOOL is specified in the OUTPUT: section (which previous did nothing to\nthe SV), and a read only SV (literal) is passed to the XSUB, croaks like \"Modification of\na read-only value attempted\" will happen. [perl #115796]\n\n• On many platforms, providing a directory name as the script name caused perl to do\nnothing and report success. It should now universally report an error and exit nonzero.\n[perl #61362]\n\n• \"sort {undef} ...\" under fatal warnings no longer crashes. It had begun crashing in Perl\nv5.16.\n\n• Stashes blessed into each other (\"bless \\%Foo::, 'Bar'; bless \\%Bar::, 'Foo'\") no longer\nresult in double frees. This bug started happening in Perl v5.16.\n\n• Numerous memory leaks have been fixed, mostly involving fatal warnings and syntax errors.\n\n• Some failed regular expression matches such as \"'f' =~ /../g\" were not resetting \"pos\".\nAlso, \"match-once\" patterns (\"m?...?g\") failed to reset it, too, when invoked a second\ntime [perl #23180].\n\n• Several bugs involving \"local *ISA\" and \"local *Foo::\" causing stale MRO caches have been\nfixed.\n\n• Defining a subroutine when its typeglob has been aliased no longer results in stale\nmethod caches. This bug was introduced in Perl v5.10.\n\n• Localising a typeglob containing a subroutine when the typeglob's package has been\ndeleted from its parent stash no longer produces an error. This bug was introduced in\nPerl v5.14.\n\n• Under some circumstances, \"local *method=...\" would fail to reset method caches upon\nscope exit.\n\n• \"/[.foo.]/\" is no longer an error, but produces a warning (as before) and is treated as\n\"/[.fo]/\" [perl #115818].\n\n• \"goto $tiedvar\" now calls FETCH before deciding what type of goto (subroutine or label)\nthis is.\n\n• Renaming packages through glob assignment (\"*Foo:: = *Bar::; *Bar:: = *Baz::\") in\ncombination with \"m?...?\" and \"reset\" no longer makes threaded builds crash.\n\n• A number of bugs related to assigning a list to hash have been fixed. Many of these\ninvolve lists with repeated keys like \"(1, 1, 1, 1)\".\n\n• The expression \"scalar(%h = (1, 1, 1, 1))\" now returns 4, not 2.\n\n• The return value of \"%h = (1, 1, 1)\" in list context was wrong. Previously this would\nreturn \"(1, undef, 1)\", now it returns \"(1, undef)\".\n\n• Perl now issues the same warning on \"($s, %h) = (1, {})\" as it does for \"(%h) =\n({})\", \"Reference found where even-sized list expected\".\n\n• A number of additional edge cases in list assignment to hashes were corrected. For\nmore details see commit 23b7025ebc.\n\n• Attributes applied to lexical variables no longer leak memory. [perl #114764]\n\n• \"dump\", \"goto\", \"last\", \"next\", \"redo\" or \"require\" followed by a bareword (or version)\nand then an infix operator is no longer a syntax error. It used to be for those infix\noperators (like \"+\") that have a different meaning where a term is expected. [perl\n#105924]\n\n• \"require a::b . 1\" and \"require a::b + 1\" no longer produce erroneous ambiguity warnings.\n[perl #107002]\n\n• Class method calls are now allowed on any string, and not just strings beginning with an\nalphanumeric character. [perl #105922]\n\n• An empty pattern created with \"qr//\" used in \"m///\" no longer triggers the \"empty pattern\nreuses last pattern\" behaviour. [perl #96230]\n\n• Tying a hash during iteration no longer results in a memory leak.\n\n• Freeing a tied hash during iteration no longer results in a memory leak.\n\n• List assignment to a tied array or hash that dies on STORE no longer results in a memory\nleak.\n\n• If the hint hash (\"%^H\") is tied, compile-time scope entry (which copies the hint hash)\nno longer leaks memory if FETCH dies. [perl #107000]\n\n• Constant folding no longer inappropriately triggers the special \"split \" \"\" behaviour.\n[perl #94490]\n\n• \"defined scalar(@array)\", \"defined do { &foo }\", and similar constructs now treat the\nargument to \"defined\" as a simple scalar. [perl #97466]\n\n• Running a custom debugging that defines no *DB::DB glob or provides a subroutine stub for\n&DB::DB no longer results in a crash, but an error instead. [perl #114990]\n\n• \"reset \"\"\" now matches its documentation. \"reset\" only resets \"m?...?\" patterns when\ncalled with no argument. An empty string for an argument now does nothing. (It used to\nbe treated as no argument.) [perl #97958]\n\n• \"printf\" with an argument returning an empty list no longer reads past the end of the\nstack, resulting in erratic behaviour. [perl #77094]\n\n• \"--subname\" no longer produces erroneous ambiguity warnings. [perl #77240]\n\n• \"v10\" is now allowed as a label or package name. This was inadvertently broken when\nv-strings were added in Perl v5.6. [perl #56880]\n\n• \"length\", \"pos\", \"substr\" and \"sprintf\" could be confused by ties, overloading,\nreferences and typeglobs if the stringification of such changed the internal\nrepresentation to or from UTF-8. [perl #114410]\n\n• utf8::encode now calls FETCH and STORE on tied variables. utf8::decode now calls STORE\n(it was already calling FETCH).\n\n• \"$tied =~ s/$nonutf8/$utf8/\" no longer loops infinitely if the tied variable returns a\nLatin-1 string, shared hash key scalar, or reference or typeglob that stringifies as\nASCII or Latin-1. This was a regression from v5.12.\n\n• \"s///\" without /e is now better at detecting when it needs to forego certain\noptimisations, fixing some buggy cases:\n\n• Match variables in certain constructs (\"&&\", \"||\", \"..\" and others) in the\nreplacement part; e.g., \"s/(.)/$l{$a||$1}/g\". [perl #26986]\n\n• Aliases to match variables in the replacement.\n\n• $REGERROR or $REGMARK in the replacement. [perl #49190]\n\n• An empty pattern (\"s//$foo/\") that causes the last-successful pattern to be used,\nwhen that pattern contains code blocks that modify the variables in the replacement.\n\n• The taintedness of the replacement string no longer affects the taintedness of the return\nvalue of \"s///e\".\n\n• The $| autoflush variable is created on-the-fly when needed. If this happened (e.g., if\nit was mentioned in a module or eval) when the currently-selected filehandle was a\ntypeglob with an empty IO slot, it used to crash. [perl #115206]\n\n• Line numbers at the end of a string eval are no longer off by one. [perl #114658]\n\n• @INC filters (subroutines returned by subroutines in @INC) that set $ to a copy-on-write\nscalar no longer cause the parser to modify that string buffer in place.\n\n• \"length($object)\" no longer returns the undefined value if the object has string\noverloading that returns undef. [perl #115260]\n\n• The use of \"PLstashcache\", the stash name lookup cache for method calls, has been\nrestored,\n\nCommit da6b625f78f5f133 in August 2011 inadvertently broke the code that looks up values\nin \"PLstashcache\". As it's only a cache, quite correctly everything carried on working\nwithout it.\n\n• The error \"Can't localize through a reference\" had disappeared in v5.16.0 when \"local\n%$ref\" appeared on the last line of an lvalue subroutine. This error disappeared for\n\"\\local %$ref\" in perl v5.8.1. It has now been restored.\n\n• The parsing of here-docs has been improved significantly, fixing several parsing bugs and\ncrashes and one memory leak, and correcting wrong subsequent line numbers under certain\nconditions.\n\n• Inside an eval, the error message for an unterminated here-doc no longer has a newline in\nthe middle of it [perl #70836].\n\n• A substitution inside a substitution pattern (\"s/${s|||}//\") no longer confuses the\nparser.\n\n• It may be an odd place to allow comments, but \"s//\"\" # hello/e\" has always worked, unless\nthere happens to be a null character before the first #. Now it works even in the\npresence of nulls.\n\n• An invalid range in \"tr///\" or \"y///\" no longer results in a memory leak.\n\n• String eval no longer treats a semicolon-delimited quote-like operator at the very end\n(\"eval 'q;;'\") as a syntax error.\n\n• \"warn {$ => 1} + 1\" is no longer a syntax error. The parser used to get confused with\ncertain list operators followed by an anonymous hash and then an infix operator that\nshares its form with a unary operator.\n\n• \"(caller $n)[6]\" (which gives the text of the eval) used to return the actual parser\nbuffer. Modifying it could result in crashes. Now it always returns a copy. The string\nreturned no longer has \"\\n;\" tacked on to the end. The returned text also includes here-\ndoc bodies, which used to be omitted.\n\n• The UTF-8 position cache is now reset when accessing magical variables, to avoid the\nstring buffer and the UTF-8 position cache getting out of sync [perl #114410].\n\n• Various cases of get magic being called twice for magical UTF-8 strings have been fixed.\n\n• This code (when not in the presence of $& etc)\n\n$ = 'x' x 1000000;\n1 while /(.)/;\n\nused to skip the buffer copy for performance reasons, but suffered from $1 etc changing\nif the original string changed. That's now been fixed.\n\n• Perl doesn't use PerlIO anymore to report out of memory messages, as PerlIO might attempt\nto allocate more memory.\n\n• In a regular expression, if something is quantified with \"{n,m}\" where \"n > m\", it can't\npossibly match. Previously this was a fatal error, but now is merely a warning (and that\nsomething won't match). [perl #82954].\n\n• It used to be possible for formats defined in subroutines that have subsequently been\nundefined and redefined to close over variables in the wrong pad (the newly-defined\nenclosing sub), resulting in crashes or \"Bizarre copy\" errors.\n\n• Redefinition of XSUBs at run time could produce warnings with the wrong line number.\n\n• The %vd sprintf format does not support version objects for alpha versions. It used to\noutput the format itself (%vd) when passed an alpha version, and also emit an \"Invalid\nconversion in printf\" warning. It no longer does, but produces the empty string in the\noutput. It also no longer leaks memory in this case.\n\n• \"$obj->SUPER::method\" calls in the main package could fail if the SUPER package had\nalready been accessed by other means.\n\n• Stash aliasing (\"*foo:: = *bar::\") no longer causes SUPER calls to ignore changes to\nmethods or @ISA or use the wrong package.\n\n• Method calls on packages whose names end in ::SUPER are no longer treated as SUPER method\ncalls, resulting in failure to find the method. Furthermore, defining subroutines in\nsuch packages no longer causes them to be found by SUPER method calls on the containing\npackage [perl #114924].\n\n• \"\\w\" now matches the code points U+200C (ZERO WIDTH NON-JOINER) and U+200D (ZERO WIDTH\nJOINER). \"\\W\" no longer matches these. This change is because Unicode corrected their\ndefinition of what \"\\w\" should match.\n\n• \"dump LABEL\" no longer leaks its label.\n\n• Constant folding no longer changes the behaviour of functions like \"stat()\" and\n\"truncate()\" that can take either filenames or handles. \"stat 1 ? foo : bar\" nows treats\nits argument as a file name (since it is an arbitrary expression), rather than the handle\n\"foo\".\n\n• \"truncate FOO, $len\" no longer falls back to treating \"FOO\" as a file name if the\nfilehandle has been deleted. This was broken in Perl v5.16.0.\n\n• Subroutine redefinitions after sub-to-glob and glob-to-glob assignments no longer cause\ndouble frees or panic messages.\n\n• \"s///\" now turns vstrings into plain strings when performing a substitution, even if the\nresulting string is the same (\"s/a/a/\").\n\n• Prototype mismatch warnings no longer erroneously treat constant subs as having no\nprototype when they actually have \"\".\n\n• Constant subroutines and forward declarations no longer prevent prototype mismatch\nwarnings from omitting the sub name.\n\n• \"undef\" on a subroutine now clears call checkers.\n\n• The \"ref\" operator started leaking memory on blessed objects in Perl v5.16.0. This has\nbeen fixed [perl #114340].\n\n• \"use\" no longer tries to parse its arguments as a statement, making \"use constant { ()\n};\" a syntax error [perl #114222].\n\n• On debugging builds, \"uninitialized\" warnings inside formats no longer cause assertion\nfailures.\n\n• On debugging builds, subroutines nested inside formats no longer cause assertion failures\n[perl #78550].\n\n• Formats and \"use\" statements are now permitted inside formats.\n\n• \"print $x\" and \"sub { print $x }->()\" now always produce the same output. It was\npossible for the latter to refuse to close over $x if the variable was not active; e.g.,\nif it was defined outside a currently-running named subroutine.\n\n• Similarly, \"print $x\" and \"print eval '$x'\" now produce the same output. This also\nallows \"my $x if 0\" variables to be seen in the debugger [perl #114018].\n\n• Formats called recursively no longer stomp on their own lexical variables, but each\nrecursive call has its own set of lexicals.\n\n• Attempting to free an active format or the handle associated with it no longer results in\na crash.\n\n• Format parsing no longer gets confused by braces, semicolons and low-precedence\noperators. It used to be possible to use braces as format delimiters (instead of \"=\" and\n\".\"), but only sometimes. Semicolons and low-precedence operators in format argument\nlines no longer confuse the parser into ignoring the line's return value. In format\nargument lines, braces can now be used for anonymous hashes, instead of being treated\nalways as \"do\" blocks.\n\n• Formats can now be nested inside code blocks in regular expressions and other quoted\nconstructs (\"/(?{...})/\" and \"qq/${...}/\") [perl #114040].\n\n• Formats are no longer created after compilation errors.\n\n• Under debugging builds, the -DA command line option started crashing in Perl v5.16.0. It\nhas been fixed [perl #114368].\n\n• A potential deadlock scenario involving the premature termination of a pseudo- forked\nchild in a Windows build with ithreads enabled has been fixed. This resolves the common\nproblem of the t/op/fork.t test hanging on Windows [perl #88840].\n\n• The code which generates errors from \"require()\" could potentially read one or two bytes\nbefore the start of the filename for filenames less than three bytes long and ending\n\"/\\.p?\\z/\". This has now been fixed. Note that it could never have happened with module\nnames given to \"use()\" or \"require()\" anyway.\n\n• The handling of pathnames of modules given to \"require()\" has been made thread-safe on\nVMS.\n\n• Non-blocking sockets have been fixed on VMS.\n\n• Pod can now be nested in code inside a quoted construct outside of a string eval. This\nused to work only within string evals [perl #114040].\n\n• \"goto ''\" now looks for an empty label, producing the \"goto must have label\" error\nmessage, instead of exiting the program [perl #111794].\n\n• \"goto \"\\0\"\" now dies with \"Can't find label\" instead of \"goto must have label\".\n\n• The C function \"hvstore\" used to result in crashes when used on \"%^H\" [perl #111000].\n\n• A call checker attached to a closure prototype via \"cvsetcallchecker\" is now copied to\nclosures cloned from it. So \"cvsetcallchecker\" now works inside an attribute handler\nfor a closure.\n\n• Writing to $^N used to have no effect. Now it croaks with \"Modification of a read-only\nvalue\" by default, but that can be overridden by a custom regular expression engine, as\nwith $1 [perl #112184].\n\n• \"undef\" on a control character glob (\"undef *^H\") no longer emits an erroneous warning\nabout ambiguity [perl #112456].\n\n• For efficiency's sake, many operators and built-in functions return the same scalar each\ntime. Lvalue subroutines and subroutines in the CORE:: namespace were allowing this\nimplementation detail to leak through. \"print &CORE::uc(\"a\"), &CORE::uc(\"b\")\" used to\nprint \"BB\". The same thing would happen with an lvalue subroutine returning the return\nvalue of \"uc\". Now the value is copied in such cases.\n\n• \"method {}\" syntax with an empty block or a block returning an empty list used to crash\nor use some random value left on the stack as its invocant. Now it produces an error.\n\n• \"vec\" now works with extremely large offsets (>2 GB) [perl #111730].\n\n• Changes to overload settings now take effect immediately, as do changes to inheritance\nthat affect overloading. They used to take effect only after \"bless\".\n\nObjects that were created before a class had any overloading used to remain non-\noverloaded even if the class gained overloading through \"use overload\" or @ISA changes,\nand even after \"bless\". This has been fixed [perl #112708].\n\n• Classes with overloading can now inherit fallback values.\n\n• Overloading was not respecting a fallback value of 0 if there were overloaded objects on\nboth sides of an assignment operator like \"+=\" [perl #111856].\n\n• \"pos\" now croaks with hash and array arguments, instead of producing erroneous warnings.\n\n• \"while(each %h)\" now implies \"while(defined($ = each %h))\", like \"readline\" and\n\"readdir\".\n\n• Subs in the CORE:: namespace no longer crash after \"undef *\" when called with no\nargument list (&CORE::time with no parentheses).\n\n• \"unpack\" no longer produces the \"'/' must follow a numeric type in unpack\" error when it\nis the data that are at fault [perl #60204].\n\n• \"join\" and \"@array\" now call FETCH only once on a tied $\" [perl #8931].\n\n• Some subroutine calls generated by compiling core ops affected by a \"CORE::GLOBAL\"\noverride had op checking performed twice. The checking is always idempotent for pure\nPerl code, but the double checking can matter when custom call checkers are involved.\n\n• A race condition used to exist around fork that could cause a signal sent to the parent\nto be handled by both parent and child. Signals are now blocked briefly around fork to\nprevent this from happening [perl #82580].\n\n• The implementation of code blocks in regular expressions, such as \"(?{})\" and \"(??{})\",\nhas been heavily reworked to eliminate a whole slew of bugs. The main user-visible\nchanges are:\n\n• Code blocks within patterns are now parsed in the same pass as the surrounding code;\nin particular it is no longer necessary to have balanced braces: this now works:\n\n/(?{ $x='{' })/\n\nThis means that this error message is no longer generated:\n\nSequence (?{...}) not terminated or not {}-balanced in regex\n\nbut a new error may be seen:\n\nSequence (?{...}) not terminated with ')'\n\nIn addition, literal code blocks within run-time patterns are only compiled once, at\nperl compile-time:\n\nfor my $p (...) {\n# this 'FOO' block of code is compiled once,\n# at the same time as the surrounding 'for' loop\n/$p{(?{FOO;})/;\n}\n\n• Lexical variables are now sane as regards scope, recursion and closure behavior. In\nparticular, \"/A(?{B})C/\" behaves (from a closure viewpoint) exactly like \"/A/ && do {\nB } && /C/\", while \"qr/A(?{B})C/\" is like \"sub {/A/ && do { B } && /C/}\". So this\ncode now works how you might expect, creating three regexes that match 0, 1, and 2:\n\nfor my $i (0..2) {\npush @r, qr/^(??{$i})$/;\n}\n\"1\" =~ $r[1]; # matches\n\n• The \"use re 'eval'\" pragma is now only required for code blocks defined at runtime;\nin particular in the following, the text of the $r pattern is still interpolated into\nthe new pattern and recompiled, but the individual compiled code-blocks within $r are\nreused rather than being recompiled, and \"use re 'eval'\" isn't needed any more:\n\nmy $r = qr/abc(?{....})def/;\n/xyz$r/;\n\n• Flow control operators no longer crash. Each code block runs in a new dynamic scope,\nso \"next\" etc. will not see any enclosing loops. \"return\" returns a value from the\ncode block, not from any enclosing subroutine.\n\n• Perl normally caches the compilation of run-time patterns, and doesn't recompile if\nthe pattern hasn't changed, but this is now disabled if required for the correct\nbehavior of closures. For example:\n\nmy $code = '(??{$x})';\nfor my $x (1..3) {\n# recompile to see fresh value of $x each time\n$x =~ /$code/;\n}\n\n• The \"/msix\" and \"(?msix)\" etc. flags are now propagated into the return value from\n\"(??{})\"; this now works:\n\n\"AB\" =~ /a(??{'b'})/i;\n\n• Warnings and errors will appear to come from the surrounding code (or for run-time\ncode blocks, from an eval) rather than from an \"reeval\":\n\nuse re 'eval'; $c = '(?{ warn \"foo\" })'; /$c/;\n/(?{ warn \"foo\" })/;\n\nformerly gave:\n\nfoo at (reeval 1) line 1.\nfoo at (reeval 2) line 1.\n\nand now gives:\n\nfoo at (eval 1) line 1.\nfoo at /some/prog line 2.\n\n• Perl now can be recompiled to use any Unicode version. In v5.16, it worked on Unicodes\n6.0 and 6.1, but there were various bugs if earlier releases were used; the older the\nrelease the more problems.\n\n• \"vec\" no longer produces \"uninitialized\" warnings in lvalue context [perl #9423].\n\n• An optimization involving fixed strings in regular expressions could cause a severe\nperformance penalty in edge cases. This has been fixed [perl #76546].\n\n• In certain cases, including empty subpatterns within a regular expression (such as \"(?:)\"\nor \"(?:|)\") could disable some optimizations. This has been fixed.\n\n• The \"Can't find an opnumber\" message that \"prototype\" produces when passed a string like\n\"CORE::nonexistentkeyword\" now passes UTF-8 and embedded NULs through unchanged [perl\n#97478].\n\n• \"prototype\" now treats magical variables like $1 the same way as non-magical variables\nwhen checking for the CORE:: prefix, instead of treating them as subroutine names.\n\n• Under threaded perls, a runtime code block in a regular expression could corrupt the\npackage name stored in the op tree, resulting in bad reads in \"caller\", and possibly\ncrashes [perl #113060].\n\n• Referencing a closure prototype (\"\\&{$[1]}\" in an attribute handler for a closure) no\nlonger results in a copy of the subroutine (or assertion failures on debugging builds).\n\n• \"eval 'PACKAGE'\" now returns the right answer on threaded builds if the current\npackage has been assigned over (as in \"*ThisPackage:: = *ThatPackage::\") [perl #78742].\n\n• If a package is deleted by code that it calls, it is possible for \"caller\" to see a stack\nframe belonging to that deleted package. \"caller\" could crash if the stash's memory\naddress was reused for a scalar and a substitution was performed on the same scalar [perl\n#113486].\n\n• \"UNIVERSAL::can\" no longer treats its first argument differently depending on whether it\nis a string or number internally.\n\n• \"open\" with \"<&\" for the mode checks to see whether the third argument is a number, in\ndetermining whether to treat it as a file descriptor or a handle name. Magical variables\nlike $1 were always failing the numeric check and being treated as handle names.\n\n• \"warn\"'s handling of magical variables ($1, ties) has undergone several fixes. \"FETCH\"\nis only called once now on a tied argument or a tied $@ [perl #97480]. Tied variables\nreturning objects that stringify as \"\" are no longer ignored. A tied $@ that happened to\nreturn a reference the previous time it was used is no longer ignored.\n\n• \"warn \"\"\" now treats $@ with a number in it the same way, regardless of whether it\nhappened via \"$@=3\" or \"$@=\"3\"\". It used to ignore the former. Now it appends\n\"\\t...caught\", as it has always done with \"$@=\"3\"\".\n\n• Numeric operators on magical variables (e.g., \"$1 + 1\") used to use floating point\noperations even where integer operations were more appropriate, resulting in loss of\naccuracy on 64-bit platforms [perl #109542].\n\n• Unary negation no longer treats a string as a number if the string happened to be used as\na number at some point. So, if $x contains the string \"dogs\", \"-$x\" returns \"-dogs\" even\nif \"$y=0+$x\" has happened at some point.\n\n• In Perl v5.14, \"-'-10'\" was fixed to return \"10\", not \"+10\". But magical variables ($1,\nties) were not fixed till now [perl #57706].\n\n• Unary negation now treats strings consistently, regardless of the internal \"UTF8\" flag.\n\n• A regression introduced in Perl v5.16.0 involving \"tr/SEARCHLIST/REPLACEMENTLIST/\" has\nbeen fixed. Only the first instance is supposed to be meaningful if a character appears\nmore than once in \"SEARCHLIST\". Under some circumstances, the final instance was\noverriding all earlier ones. [perl #113584]\n\n• Regular expressions like \"qr/\\87/\" previously silently inserted a NUL character, thus\nmatching as if it had been written \"qr/\\00087/\". Now it matches as if it had been\nwritten as \"qr/87/\", with a message that the sequence \"\\8\" is unrecognized.\n\n• \"SUB\" now works in special blocks (\"BEGIN\", \"END\", etc.).\n\n• Thread creation on Windows could theoretically result in a crash if done inside a \"BEGIN\"\nblock. It still does not work properly, but it no longer crashes [perl #111610].\n\n• \"\\&{''}\" (with the empty string) now autovivifies a stub like any other sub name, and no\nlonger produces the \"Unable to create sub\" error [perl #94476].\n\n• A regression introduced in v5.14.0 has been fixed, in which some calls to the \"re\" module\nwould clobber $ [perl #113750].\n\n• \"do FILE\" now always either sets or clears $@, even when the file can't be read. This\nensures that testing $@ first (as recommended by the documentation) always returns the\ncorrect result.\n\n• The array iterator used for the \"each @array\" construct is now correctly reset when\n@array is cleared [perl #75596]. This happens, for example, when the array is globally\nassigned to, as in \"@array = (...)\", but not when its values are assigned to. In terms of\nthe XS API, it means that \"avclear()\" will now reset the iterator.\n\nThis mirrors the behaviour of the hash iterator when the hash is cleared.\n\n• \"$class->can\", \"$class->isa\", and \"$class->DOES\" now return correct results, regardless\nof whether that package referred to by $class exists [perl #47113].\n\n• Arriving signals no longer clear $@ [perl #45173].\n\n• Allow \"my ()\" declarations with an empty variable list [perl #113554].\n\n• During parsing, subs declared after errors no longer leave stubs [perl #113712].\n\n• Closures containing no string evals no longer hang on to their containing subroutines,\nallowing variables closed over by outer subroutines to be freed when the outer sub is\nfreed, even if the inner sub still exists [perl #89544].\n\n• Duplication of in-memory filehandles by opening with a \"<&=\" or \">&=\" mode stopped\nworking properly in v5.16.0. It was causing the new handle to reference a different\nscalar variable. This has been fixed [perl #113764].\n\n• \"qr//\" expressions no longer crash with custom regular expression engines that do not set\n\"offs\" at regular expression compilation time [perl #112962].\n\n• \"delete local\" no longer crashes with certain magical arrays and hashes [perl #112966].\n\n• \"local\" on elements of certain magical arrays and hashes used not to arrange to have the\nelement deleted on scope exit, even if the element did not exist before \"local\".\n\n• \"scalar(write)\" no longer returns multiple items [perl #73690].\n\n• String to floating point conversions no longer misparse certain strings under \"use\nlocale\" [perl #109318].\n\n• @INC filters that die no longer leak memory [perl #92252].\n\n• The implementations of overloaded operations are now called in the correct context. This\nallows, among other things, being able to properly override \"<>\" [perl #47119].\n\n• Specifying only the \"fallback\" key when calling \"use overload\" now behaves properly [perl\n#113010].\n\n• \"sub foo { my $a = 0; while ($a) { ... } }\" and \"sub foo { while (0) { ... } }\" now\nreturn the same thing [perl #73618].\n\n• String negation now behaves the same under \"use integer;\" as it does without [perl\n#113012].\n\n• \"chr\" now returns the Unicode replacement character (U+FFFD) for -1, regardless of the\ninternal representation. -1 used to wrap if the argument was tied or a string\ninternally.\n\n• Using a \"format\" after its enclosing sub was freed could crash as of perl v5.12.0, if the\nformat referenced lexical variables from the outer sub.\n\n• Using a \"format\" after its enclosing sub was undefined could crash as of perl v5.10.0, if\nthe format referenced lexical variables from the outer sub.\n\n• Using a \"format\" defined inside a closure, which format references lexical variables from\noutside, never really worked unless the \"write\" call was directly inside the closure. In\nv5.10.0 it even started crashing. Now the copy of that closure nearest the top of the\ncall stack is used to find those variables.\n\n• Formats that close over variables in special blocks no longer crash if a stub exists with\nthe same name as the special block before the special block is compiled.\n\n• The parser no longer gets confused, treating \"eval foo ()\" as a syntax error if preceded\nby \"print;\" [perl #16249].\n\n• The return value of \"syscall\" is no longer truncated on 64-bit platforms [perl #113980].\n\n• Constant folding no longer causes \"print 1 ? FOO : BAR\" to print to the FOO handle [perl\n#78064].\n\n• \"do subname\" now calls the named subroutine and uses the file name it returns, instead of\nopening a file named \"subname\".\n\n• Subroutines looked up by rv2cv check hooks (registered by XS modules) are now taken into\nconsideration when determining whether \"foo bar\" should be the sub call \"foo(bar)\" or the\nmethod call \"\"bar\"->foo\".\n\n• \"CORE::foo::bar\" is no longer treated specially, allowing global overrides to be called\ndirectly via \"CORE::GLOBAL::uc(...)\" [perl #113016].\n\n• Calling an undefined sub whose typeglob has been undefined now produces the customary\n\"Undefined subroutine called\" error, instead of \"Not a CODE reference\".\n\n• Two bugs involving @ISA have been fixed. \"*ISA = *globwithoutarray\" and \"undef *ISA;\n@{*ISA}\" would prevent future modifications to @ISA from updating the internal caches\nused to look up methods. The *globwithoutarray case was a regression from Perl v5.12.\n\n• Regular expression optimisations sometimes caused \"$\" with \"/m\" to produce failed or\nincorrect matches [perl #114068].\n\n• \"SUB\" now works in a \"sort\" block when the enclosing subroutine is predeclared with\n\"sub foo;\" syntax [perl #113710].\n\n• Unicode properties only apply to Unicode code points, which leads to some subtleties when\nregular expressions are matched against above-Unicode code points. There is a warning\ngenerated to draw your attention to this. However, this warning was being generated\ninappropriately in some cases, such as when a program was being parsed. Non-Unicode\nmatches such as \"\\w\" and \"[:word:]\" should not generate the warning, as their definitions\ndon't limit them to apply to only Unicode code points. Now the message is only generated\nwhen matching against \"\\p{}\" and \"\\P{}\". There remains a bug, [perl #114148], for the\nvery few properties in Unicode that match just a single code point. The warning is not\ngenerated if they are matched against an above-Unicode code point.\n\n• Uninitialized warnings mentioning hash elements would only mention the element name if it\nwas not in the first bucket of the hash, due to an off-by-one error.\n\n• A regular expression optimizer bug could cause multiline \"^\" to behave incorrectly in the\npresence of line breaks, such that \"\"/\\n\\n\" =~ m#\\A(?:^/$)#im\" would not match [perl\n#115242].\n\n• Failed \"fork\" in list context no longer corrupts the stack. \"@a = (1, 2, fork, 3)\" used\nto gobble up the 2 and assign \"(1, undef, 3)\" if the \"fork\" call failed.\n\n• Numerous memory leaks have been fixed, mostly involving tied variables that die, regular\nexpression character classes and code blocks, and syntax errors.\n\n• Assigning a regular expression (\"${qr//}\") to a variable that happens to hold a floating\npoint number no longer causes assertion failures on debugging builds.\n\n• Assigning a regular expression to a scalar containing a number no longer causes\nsubsequent numification to produce random numbers.\n\n• Assigning a regular expression to a magic variable no longer wipes away the magic. This\nwas a regression from v5.10.\n\n• Assigning a regular expression to a blessed scalar no longer results in crashes. This\nwas also a regression from v5.10.\n\n• Regular expression can now be assigned to tied hash and array elements with flattening\ninto strings.\n\n• Numifying a regular expression no longer results in an uninitialized warning.\n\n• Negative array indices no longer cause EXISTS methods of tied variables to be ignored.\nThis was a regression from v5.12.\n\n• Negative array indices no longer result in crashes on arrays tied to non-objects.\n\n• \"$byteoverload .= $utf8\" no longer results in doubly-encoded UTF-8 if the left-hand\nscalar happened to have produced a UTF-8 string the last time overloading was invoked.\n\n• \"goto &sub\" now uses the current value of @, instead of using the array the subroutine\nwas originally called with. This means \"local @ = (...); goto &sub\" now works [perl\n#43077].\n\n• If a debugger is invoked recursively, it no longer stomps on its own lexical variables.\nFormerly under recursion all calls would share the same set of lexical variables [perl\n#115742].\n\n• *{ARRAY} returned from a subroutine no longer spontaneously becomes empty.\n\n• When using \"say\" to print to a tied filehandle, the value of \"$\\\" is correctly localized,\neven if it was previously undef. [perl #119927]\n" }, { "name": "Known Problems", "content": "• UTF8-flagged strings in %ENV on HP-UX 11.00 are buggy\n\nThe interaction of UTF8-flagged strings and %ENV on HP-UX 11.00 is currently dodgy in\nsome not-yet-fully-diagnosed way. Expect test failures in t/op/magic.t, followed by\nunknown behavior when storing wide characters in the environment.\n" } ] }, "Obituary": { "content": "Hojung Yoon (AMORETTE), 24, of Seoul, South Korea, went to his long rest on May 8, 2013 with\nllama figurine and autographed TIMTOADY card. He was a brilliant young Perl 5 & 6 hacker and\na devoted member of Seoul.pm. He programmed Perl, talked Perl, ate Perl, and loved Perl. We\nbelieve that he is still programming in Perl with his broken IBM laptop somewhere. He will\nbe missed.\n", "subsections": [] }, "Acknowledgements": { "content": "Perl v5.18.0 represents approximately 12 months of development since Perl v5.16.0 and\ncontains approximately 400,000 lines of changes across 2,100 files from 113 authors.\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 v5.18.0:\n\nAaron Crane, Aaron Trevena, Abhijit Menon-Sen, Adrian M. Enache, Alan Haggai Alavi, Alexandr\nCiornii, Andrew Tam, Andy Dougherty, Anton Nikishaev, Aristotle Pagaltzis, Augustina Blair,\nBob Ernst, Brad Gilbert, Breno G. de Oliveira, Brian Carlson, Brian Fraser, Charlie Gonzalez,\nChip Salzenberg, Chris 'BinGOs' Williams, Christian Hansen, Colin Kuskie, Craig A. Berry,\nDagfinn Ilmari Mannsåker, Daniel Dragan, Daniel Perrett, Darin McBride, Dave Rolsky, David\nGolden, David Leadbeater, David Mitchell, David Nicol, Dominic Hargreaves, E. Choroba, Eric\nBrine, Evan Miller, Father Chrysostomos, Florian Ragwitz, François Perrad, George Greer, Goro\nFuji, H.Merijn Brand, Herbert Breunung, Hugo van der Sanden, Igor Zaytsev, James E Keenan,\nJan Dubois, Jasmine Ahuja, Jerry D. Hedden, Jess Robinson, Jesse Luehrs, Joaquin Ferrero,\nJoel Berger, John Goodyear, John Peacock, Karen Etheridge, Karl Williamson, Karthik\nRajagopalan, Kent Fredric, Leon Timmermans, Lucas Holt, Lukas Mai, Marcus Holland-Moritz,\nMarkus Jansen, Martin Hasch, Matthew Horsfall, Max Maischein, Michael G Schwern, Michael\nSchroeder, Moritz Lenz, Nicholas Clark, Niko Tyni, Oleg Nesterov, Patrik Hägglund, Paul\nGreen, Paul Johnson, Paul Marquess, Peter Martini, Rafael Garcia-Suarez, Reini Urban, Renee\nBaecker, Rhesa Rozendaal, Ricardo Signes, Robin Barker, Ronald J. Kimball, Ruslan Zakirov,\nSalvador Fandiño, Sawyer X, Scott Lanning, Sergey Alekseev, Shawn M Moore, Shirakata Kentaro,\nShlomi Fish, Sisyphus, Smylers, Steffen Müller, Steve Hay, Steve Peters, Steven Schubiger,\nSullivan Beck, Sven Strickroth, Sébastien Aperghis-Tramoni, Thomas Sibley, Tobias Leich, Tom\nWyant, Tony Cook, Vadim Konovalov, Vincent Pit, Volker Schatz, Walt Mankowski, Yves Orton,\nZefram.\n\nThe list above is almost certainly incomplete as it is automatically generated from version\ncontrol history. In particular, it does not include the names of the (very much appreciated)\ncontributors who reported issues to the Perl bug tracker.\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\nFor a more complete list of all of Perl's historical contributors, please see the AUTHORS\nfile in the Perl source distribution.\n", "subsections": [ { "name": "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 http://rt.perl.org/perlbug/ .\nThere may also be information at http://www.perl.org/ , 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 analysed 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\nthe Perl core, not for modules independently distributed on CPAN.\n" } ] }, "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\n\nperl v5.34.0 2025-07-25 PERL5180DELTA(1)", "subsections": [] } }, "summary": "perl5180delta - what is new for perl v5.18.0", "flags": [], "examples": [], "see_also": [] }