{
"mode": "man",
"parameter": "PERL5240DELTA",
"section": "1",
"url": "https://www.chedong.com/phpMan.php/man/PERL5240DELTA/1/json",
"generated": "2026-10-09T08:49:11Z",
"sections": {
"NAME": {
"content": "perl5240delta - what is new for perl v5.24.0\n",
"subsections": []
},
"DESCRIPTION": {
"content": "This document describes the differences between the 5.22.0 release and the 5.24.0 release.\n",
"subsections": []
},
"Core Enhancements": {
"content": "",
"subsections": [
{
"name": "Postfix dereferencing is no longer experimental",
"content": "Using the \"postderef\" and \"postderefqq\" features no longer emits a warning. Existing code\nthat disables the \"experimental::postderef\" warning category that they previously used will\ncontinue to work. The \"postderef\" feature has no effect; all Perl code can use postfix\ndereferencing, regardless of what feature declarations are in scope. The 5.24 feature bundle\nnow includes the \"postderefqq\" feature.\n"
},
{
"name": "Unicode 8.0 is now supported",
"content": "For details on what is in this release, see .\n"
},
{
"name": "perl will now croak when closing an in-place output file fails",
"content": "Until now, failure to close the output file for an in-place edit was not detected, meaning\nthat the input file could be clobbered without the edit being successfully completed. Now,\nwhen the output file cannot be closed successfully, an exception is raised.\n"
},
{
"name": "New \"\\b{lb}\" boundary in regular expressions",
"content": "\"lb\" stands for Line Break. It is a Unicode property that determines where a line of text is\nsuitable to break (typically so that it can be output without overflowing the available\nhorizontal space). This capability has long been furnished by the Unicode::LineBreak module,\nbut now a light-weight, non-customizable version that is suitable for many purposes is in\ncore Perl.\n"
},
{
"name": "\"qr/(?[ ])/\" now works in UTF-8 locales",
"content": "Extended Bracketed Character Classes now will successfully compile when \"use locale\" is in\neffect. The compiled pattern will use standard Unicode rules. If the runtime locale is not\na UTF-8 one, a warning is raised and standard Unicode rules are used anyway. No tainting is\ndone since the outcome does not actually depend on the locale.\n"
},
{
"name": "Integer shift (\"<<\" and \">>\") now more explicitly defined",
"content": "Negative shifts are reverse shifts: left shift becomes right shift, and right shift becomes\nleft shift.\n\nShifting by the number of bits in a native integer (or more) is zero, except when the\n\"overshift\" is right shifting a negative value under \"use integer\", in which case the result\nis -1 (arithmetic shift).\n\nUntil now negative shifting and overshifting have been undefined because they have relied on\nwhatever the C implementation happens to do. For example, for the overshift a common C\nbehavior is \"modulo shift\":\n\n1 >> 64 == 1 >> (64 % 64) == 1 >> 0 == 1 # Common C behavior.\n\n# And the same for <<, while Perl now produces 0 for both.\n\nNow these behaviors are well-defined under Perl, regardless of what the underlying C\nimplementation does. Note, however, that you are still constrained by the native integer\nwidth: you need to know how far left you can go. You can use for example:\n\nuse Config;\nmy $wordbits = $Config{uvsize} * 8; # Or $Config{uvsize} << 3.\n\nIf you need a more bits on the left shift, you can use for example the \"bigint\" pragma, or\nthe \"Bit::Vector\" module from CPAN.\n"
},
{
"name": "printf and sprintf now allow reordered precision arguments",
"content": "That is, \"sprintf '|%.*2$d|', 2, 3\" now returns \"|002|\". This extends the existing reordering\nmechanism (which allows reordering for arguments that are used as format fields, widths, and\nvector separators).\n"
},
{
"name": "More fields provided to \"sigaction\" callback with \"SA_SIGINFO\"",
"content": "When passing the \"SASIGINFO\" flag to sigaction, the \"errno\", \"status\", \"uid\", \"pid\", \"addr\"\nand \"band\" fields are now included in the hash passed to the handler, if supported by the\nplatform.\n"
},
{
"name": "Hashbang redirection to Perl 6",
"content": "Previously perl would redirect to another interpreter if it found a hashbang path unless the\npath contains \"perl\" (see perlrun). To improve compatibility with Perl 6 this behavior has\nbeen extended to also redirect if \"perl\" is followed by \"6\".\n"
}
]
},
"Security": {
"content": "",
"subsections": [
{
"name": "Set proper umask before calling mkstemp(3)",
"content": "In 5.22 perl started setting umask to 0600 before calling mkstemp(3) and restoring it\nafterwards. This wrongfully tells open(2) to strip the owner read and write bits from the\ngiven mode before applying it, rather than the intended negation of leaving only those bits\nin place.\n\nSystems that use mode 0666 in mkstemp(3) (like old versions of glibc) create a file with\npermissions 0066, leaving world read and write permissions regardless of current umask.\n\nThis has been fixed by using umask 0177 instead. [perl #127322]\n"
},
{
"name": "Fix out of boundary access in Win32 path handling",
"content": "This is CVE-2015-8608. For more information see [GH #15067]\n\n"
},
{
"name": "Fix loss of taint in canonpath",
"content": "This is CVE-2015-8607. For more information see [GH #15084]\n\n"
},
{
"name": "Avoid accessing uninitialized memory in win32 crypt()",
"content": "Added validation that will detect both a short salt and invalid characters in the salt. [GH\n#15091] \n"
},
{
"name": "Remove duplicate environment variables from \"environ\"",
"content": "Previously, if an environment variable appeared more than once in \"environ[]\", %ENV would\ncontain the last entry for that name, while a typical getenv() would return the first entry.\nWe now make sure %ENV contains the same as what \"getenv\" returns.\n\nSecond, we remove duplicates from \"environ[]\", so if a setting with that name is set in %ENV,\nwe won't pass an unsafe value to a child process.\n\n[CVE-2016-2381]\n"
}
]
},
"Incompatible Changes": {
"content": "",
"subsections": [
{
"name": "The \"autoderef\" feature has been removed",
"content": "The experimental \"autoderef\" feature (which allowed calling \"push\", \"pop\", \"shift\",\n\"unshift\", \"splice\", \"keys\", \"values\", and \"each\" on a scalar argument) has been deemed\nunsuccessful. It has now been removed; trying to use the feature (or to disable the\n\"experimental::autoderef\" warning it previously triggered) now yields an exception.\n"
},
{
"name": "Lexical $_ has been removed",
"content": "\"my $\" was introduced in Perl 5.10, and subsequently caused much confusion with no obvious\nsolution. In Perl 5.18.0, it was made experimental on the theory that it would either be\nremoved or redesigned in a less confusing (but backward-incompatible) way. Over the\nfollowing years, no alternatives were proposed. The feature has now been removed and will\nfail to compile.\n"
},
{
"name": "\"qr/\\b{wb}/\" is now tailored to Perl expectations",
"content": "This is now more suited to be a drop-in replacement for plain \"\\b\", but giving better results\nfor parsing natural language. Previously it strictly followed the current Unicode rules\nwhich calls for it to match between each white space character. Now it doesn't generally\nmatch within spans of white space, behaving like \"\\b\" does. See \"\\b{wb}\" in perlrebackslash\n"
},
{
"name": "Regular expression compilation errors",
"content": "Some regular expression patterns that had runtime errors now don't compile at all.\n\nAlmost all Unicode properties using the \"\\p{}\" and \"\\P{}\" regular expression pattern\nconstructs are now checked for validity at pattern compilation time, and invalid ones will\ncause the program to not compile. In earlier releases, this check was often deferred until\nrun time. Whenever an error check is moved from run- to compile time, erroneous code is\ncaught 100% of the time, whereas before it would only get caught if and when the offending\nportion actually gets executed, which for unreachable code might be never.\n"
},
{
"name": "\"qr/\\N{}/\" now disallowed under \"use re \"strict\"\"",
"content": "An empty \"\\N{}\" makes no sense, but for backwards compatibility is accepted as doing nothing,\nthough a deprecation warning is raised by default. But now this is a fatal error under the\nexperimental feature \"'strict' mode\" in re.\n"
},
{
"name": "Nested declarations are now disallowed",
"content": "A \"my\", \"our\", or \"state\" declaration is no longer allowed inside of another \"my\", \"our\", or\n\"state\" declaration.\n\nFor example, these are now fatal:\n\nmy ($x, my($y));\nour (my $x);\n\n[GH #14799] \n\n[GH #13548] \n"
},
{
"name": "The \"/\\C/\" character class has been removed.",
"content": "This regular expression character class was deprecated in v5.20.0 and has produced a\ndeprecation warning since v5.22.0. It is now a compile-time error. If you need to examine the\nindividual bytes that make up a UTF8-encoded character, then use utf8::encode() on the string\n(or a copy) first.\n"
},
{
"name": "chdir('') no longer chdirs home",
"content": "Using chdir('') or chdir(undef) to chdir home has been deprecated since perl v5.8, and will\nnow fail. Use chdir() instead.\n"
},
{
"name": "ASCII characters in variable names must now be all visible",
"content": "It was legal until now on ASCII platforms for variable names to contain non-graphical ASCII\ncontrol characters (ordinals 0 through 31, and 127, which are the C0 controls and \"DELETE\").\nThis usage has been deprecated since v5.20, and as of now causes a syntax error. The\nvariables these names referred to are special, reserved by Perl for whatever use it may\nchoose, now, or in the future. Each such variable has an alternative way of spelling it.\nInstead of the single non-graphic control character, a two character sequence beginning with\na caret is used, like $^] and \"${^GLOBALPHASE}\". Details are at perlvar. It remains\nlegal, though unwise and deprecated (raising a deprecation warning), to use certain non-\ngraphic non-ASCII characters in variables names when not under \"use utf8\". No code should do\nthis, as all such variables are reserved by Perl, and Perl doesn't currently define any of\nthem (but could at any time, without notice).\n"
},
{
"name": "An off by one issue in $Carp::MaxArgNums has been fixed",
"content": "$Carp::MaxArgNums is supposed to be the number of arguments to display. Prior to this\nversion, it was instead showing $Carp::MaxArgNums + 1 arguments, contrary to the\ndocumentation.\n"
},
{
"name": "Only blanks and tabs are now allowed within \"[...]\" within \"(?[...])\".",
"content": "The experimental Extended Bracketed Character Classes can contain regular bracketed character\nclasses within them. These differ from regular ones in that white space is generally\nignored, unless escaped by preceding it with a backslash. The white space that is ignored is\nnow limited to just tab \"\\t\" and SPACE characters. Previously, it was any white space. See\n\"Extended Bracketed Character Classes\" in perlrecharclass.\n"
}
]
},
"Deprecations": {
"content": "",
"subsections": [
{
"name": "Using code points above the platform's \"IV_MAX\" is now deprecated",
"content": "Unicode defines code points in the range \"0..0x10FFFF\". Some standards at one time defined\nthem up to 231 - 1, but Perl has allowed them to be as high as anything that will fit in a\nword on the platform being used. However, use of those above the platform's \"IVMAX\" is\nbroken in some constructs, notably \"tr///\", regular expression patterns involving\nquantifiers, and in some arithmetic and comparison operations, such as being the upper limit\nof a loop. Now the use of such code points raises a deprecation warning, unless that warning\ncategory is turned off. \"IVMAX\" is typically 231 -1 on 32-bit platforms, and 263-1 on\n64-bit ones.\n"
},
{
"name": "Doing bitwise operations on strings containing code points above 0xFF is deprecated",
"content": "The string bitwise operators treat their operands as strings of bytes, and values beyond 0xFF\nare nonsensical in this context. To operate on encoded bytes, first encode the strings. To\noperate on code points' numeric values, use \"split\" and \"map ord\". In the future, this\nwarning will be replaced by an exception.\n"
},
{
"name": "sysread(), syswrite(), recv() and send() are deprecated on :utf8 handles",
"content": "The sysread(), recv(), syswrite() and send() operators are deprecated on handles that have\nthe \":utf8\" layer, either explicitly, or implicitly, eg., with the :encoding(UTF-16LE) layer.\n\nBoth sysread() and recv() currently use only the \":utf8\" flag for the stream, ignoring the\nactual layers. Since sysread() and recv() do no UTF-8 validation they can end up creating\ninvalidly encoded scalars.\n\nSimilarly, syswrite() and send() use only the \":utf8\" flag, otherwise ignoring any layers.\nIf the flag is set, both write the value UTF-8 encoded, even if the layer is some different\nencoding, such as the example above.\n\nIdeally, all of these operators would completely ignore the \":utf8\" state, working only with\nbytes, but this would result in silently breaking existing code. To avoid this a future\nversion of perl will throw an exception when any of sysread(), recv(), syswrite() or send()\nare called on handle with the \":utf8\" layer.\n"
}
]
},
"Performance Enhancements": {
"content": "• The overhead of scope entry and exit has been considerably reduced, so for example\nsubroutine calls, loops and basic blocks are all faster now. This empty function call\nnow takes about a third less time to execute:\n\nsub f{} f();\n\n• Many languages, such as Chinese, are caseless. Perl now knows about most common ones,\nand skips much of the work when a program tries to change case in them (like ucfirst())\nor match caselessly (\"qr//i\"). This will speed up a program, such as a web server, that\ncan operate on multiple languages, while it is operating on a caseless one.\n\n• \"/fixed-substr/\" has been made much faster.\n\nOn platforms with a libc memchr() implementation which makes good use of underlying\nhardware support, patterns which include fixed substrings will now often be much faster;\nfor example with glibc on a recent x8664 CPU, this:\n\n$s = \"a\" x 1000 . \"wxyz\";\n$s =~ /wxyz/ for 1..30000\n\nis now about 7 times faster. On systems with slow memchr(), e.g. 32-bit ARM Raspberry\nPi, there will be a small or little speedup. Conversely, some pathological cases, such\nas \"\"ab\" x 1000 =~ /aa/\" will be slower now; up to 3 times slower on the rPi, 1.5x slower\non x8664.\n\n• Faster addition, subtraction and multiplication.\n\nSince 5.8.0, arithmetic became slower due to the need to support 64-bit integers. To deal\nwith 64-bit integers, a lot more corner cases need to be checked, which adds time. We now\ndetect common cases where there is no need to check for those corner cases, and special-\ncase them.\n\n• Preincrement, predecrement, postincrement, and postdecrement have been made faster by\ninternally splitting the functions which handled multiple cases into different functions.\n\n• Creating Perl debugger data structures (see \"Debugger Internals\" in perldebguts) for\nXSUBs and const subs has been removed. This removed one glob/scalar combo for each\nunique \".c\" file that XSUBs and const subs came from. On startup (\"perl -e\"0\"\") about\nhalf a dozen glob/scalar debugger combos were created. Loading XS modules created more\nglob/scalar combos. These things were being created regardless of whether the perl\ndebugger was being used, and despite the fact that it can't debug C code anyway\n\n• On Win32, \"stat\"ing or \"-X\"ing a path, if the file or directory does not exist, is now\n3.5x faster than before.\n\n• Single arguments in list assign are now slightly faster:\n\n($x) = (...);\n(...) = ($x);\n\n• Less peak memory is now used when compiling regular expression patterns.\n",
"subsections": []
},
"Modules and Pragmata": {
"content": "",
"subsections": [
{
"name": "Updated Modules and Pragmata",
"content": "• arybase has been upgraded from version 0.10 to 0.11.\n\n• Attribute::Handlers has been upgraded from version 0.97 to 0.99.\n\n• autodie has been upgraded from version 2.26 to 2.29.\n\n• autouse has been upgraded from version 1.08 to 1.11.\n\n• B has been upgraded from version 1.58 to 1.62.\n\n• B::Deparse has been upgraded from version 1.35 to 1.37.\n\n• base has been upgraded from version 2.22 to 2.23.\n\n• Benchmark has been upgraded from version 1.2 to 1.22.\n\n• bignum has been upgraded from version 0.39 to 0.42.\n\n• bytes has been upgraded from version 1.04 to 1.05.\n\n• Carp has been upgraded from version 1.36 to 1.40.\n\n• Compress::Raw::Bzip2 has been upgraded from version 2.068 to 2.069.\n\n• Compress::Raw::Zlib has been upgraded from version 2.068 to 2.069.\n\n• Config::Perl::V has been upgraded from version 0.24 to 0.25.\n\n• CPAN::Meta has been upgraded from version 2.150001 to 2.150005.\n\n• CPAN::Meta::Requirements has been upgraded from version 2.132 to 2.140.\n\n• CPAN::Meta::YAML has been upgraded from version 0.012 to 0.018.\n\n• Data::Dumper has been upgraded from version 2.158 to 2.160.\n\n• Devel::Peek has been upgraded from version 1.22 to 1.23.\n\n• Devel::PPPort has been upgraded from version 3.31 to 3.32.\n\n• Dumpvalue has been upgraded from version 1.17 to 1.18.\n\n• DynaLoader has been upgraded from version 1.32 to 1.38.\n\n• Encode has been upgraded from version 2.72 to 2.80.\n\n• encoding has been upgraded from version 2.14 to 2.17.\n\n• encoding::warnings has been upgraded from version 0.11 to 0.12.\n\n• English has been upgraded from version 1.09 to 1.10.\n\n• Errno has been upgraded from version 1.23 to 1.25.\n\n• experimental has been upgraded from version 0.013 to 0.016.\n\n• ExtUtils::CBuilder has been upgraded from version 0.280221 to 0.280225.\n\n• ExtUtils::Embed has been upgraded from version 1.32 to 1.33.\n\n• ExtUtils::MakeMaker has been upgraded from version 7.0401 to 7.1001.\n\n• ExtUtils::ParseXS has been upgraded from version 3.28 to 3.31.\n\n• ExtUtils::Typemaps has been upgraded from version 3.28 to 3.31.\n\n• feature has been upgraded from version 1.40 to 1.42.\n\n• fields has been upgraded from version 2.17 to 2.23.\n\n• File::Find has been upgraded from version 1.29 to 1.34.\n\n• File::Glob has been upgraded from version 1.24 to 1.26.\n\n• File::Path has been upgraded from version 2.09 to 2.1201.\n\n• File::Spec has been upgraded from version 3.56 to 3.63.\n\n• Filter::Util::Call has been upgraded from version 1.54 to 1.55.\n\n• Getopt::Long has been upgraded from version 2.45 to 2.48.\n\n• Hash::Util has been upgraded from version 0.18 to 0.19.\n\n• Hash::Util::FieldHash has been upgraded from version 1.15 to 1.19.\n\n• HTTP::Tiny has been upgraded from version 0.054 to 0.056.\n\n• I18N::Langinfo has been upgraded from version 0.12 to 0.13.\n\n• if has been upgraded from version 0.0604 to 0.0606.\n\n• IO has been upgraded from version 1.35 to 1.36.\n\n• IO-Compress has been upgraded from version 2.068 to 2.069.\n\n• IPC::Open3 has been upgraded from version 1.18 to 1.20.\n\n• IPC::SysV has been upgraded from version 2.04 to 2.0601.\n\n• List::Util has been upgraded from version 1.41 to 1.4202.\n\n• locale has been upgraded from version 1.06 to 1.08.\n\n• Locale::Codes has been upgraded from version 3.34 to 3.37.\n\n• Math::BigInt has been upgraded from version 1.9997 to 1.999715.\n\n• Math::BigInt::FastCalc has been upgraded from version 0.31 to 0.40.\n\n• Math::BigRat has been upgraded from version 0.2608 to 0.260802.\n\n• Module::CoreList has been upgraded from version 5.20150520 to 5.20160320.\n\n• Module::Metadata has been upgraded from version 1.000026 to 1.000031.\n\n• mro has been upgraded from version 1.17 to 1.18.\n\n• ODBMFile has been upgraded from version 1.12 to 1.14.\n\n• Opcode has been upgraded from version 1.32 to 1.34.\n\n• parent has been upgraded from version 0.232 to 0.234.\n\n• Parse::CPAN::Meta has been upgraded from version 1.4414 to 1.4417.\n\n• Perl::OSType has been upgraded from version 1.008 to 1.009.\n\n• perlfaq has been upgraded from version 5.021009 to 5.021010.\n\n• PerlIO::encoding has been upgraded from version 0.21 to 0.24.\n\n• PerlIO::mmap has been upgraded from version 0.014 to 0.016.\n\n• PerlIO::scalar has been upgraded from version 0.22 to 0.24.\n\n• PerlIO::via has been upgraded from version 0.15 to 0.16.\n\n• Pod::Functions has been upgraded from version 1.09 to 1.10.\n\n• Pod::Perldoc has been upgraded from version 3.25 to 3.2502.\n\n• Pod::Simple has been upgraded from version 3.29 to 3.32.\n\n• Pod::Usage has been upgraded from version 1.64 to 1.68.\n\n• POSIX has been upgraded from version 1.53 to 1.65.\n\n• Scalar::Util has been upgraded from version 1.41 to 1.4202.\n\n• SDBMFile has been upgraded from version 1.13 to 1.14.\n\n• SelfLoader has been upgraded from version 1.22 to 1.23.\n\n• Socket has been upgraded from version 2.018 to 2.02003.\n\n• Storable has been upgraded from version 2.53 to 2.56.\n\n• strict has been upgraded from version 1.09 to 1.11.\n\n• Term::ANSIColor has been upgraded from version 4.03 to 4.04.\n\n• Term::Cap has been upgraded from version 1.15 to 1.17.\n\n• Test has been upgraded from version 1.26 to 1.28.\n\n• Test::Harness has been upgraded from version 3.35 to 3.36.\n\n• Thread::Queue has been upgraded from version 3.05 to 3.08.\n\n• threads has been upgraded from version 2.01 to 2.06.\n\n• threads::shared has been upgraded from version 1.48 to 1.50.\n\n• Tie::File has been upgraded from version 1.01 to 1.02.\n\n• Tie::Scalar has been upgraded from version 1.03 to 1.04.\n\n• Time::HiRes has been upgraded from version 1.9726 to 1.9732.\n\n• Time::Piece has been upgraded from version 1.29 to 1.31.\n\n• Unicode::Collate has been upgraded from version 1.12 to 1.14.\n\n• Unicode::Normalize has been upgraded from version 1.18 to 1.25.\n\n• Unicode::UCD has been upgraded from version 0.61 to 0.64.\n\n• UNIVERSAL has been upgraded from version 1.12 to 1.13.\n\n• utf8 has been upgraded from version 1.17 to 1.19.\n\n• version has been upgraded from version 0.9909 to 0.9916.\n\n• warnings has been upgraded from version 1.32 to 1.36.\n\n• Win32 has been upgraded from version 0.51 to 0.52.\n\n• Win32API::File has been upgraded from version 0.1202 to 0.1203.\n\n• XS::Typemap has been upgraded from version 0.13 to 0.14.\n\n• XSLoader has been upgraded from version 0.20 to 0.21.\n"
}
]
},
"Documentation": {
"content": "",
"subsections": [
{
"name": "Changes to Existing Documentation",
"content": "perlapi\n\n• The process of using undocumented globals has been documented, namely, that one should\nsend email to perl5-porters@perl.org first to get the go-\nahead for documenting and using an undocumented function or global variable.\n\nperlcall\n\n• A number of cleanups have been made to perlcall, including:\n\n• use \"EXTEND(SP, n)\" and PUSHs() instead of XPUSHs() where applicable and update prose\nto match\n\n• add POPu, POPul and POPpbytex to the \"complete list of POP macros\" and clarify the\ndocumentation for some of the existing entries, and a note about side-effects\n\n• add API documentation for POPu and POPul\n\n• use ERRSV more efficiently\n\n• approaches to thread-safety storage of SVs.\n\nperlfunc\n\n• The documentation of \"hex\" has been revised to clarify valid inputs.\n\n• Better explain meaning of negative PIDs in \"waitpid\". [GH #15108]\n\n\n• General cleanup: there's more consistency now (in POD usage, grammar, code examples),\nbetter practices in code examples (use of \"my\", removal of bareword filehandles, dropped\nusage of \"&\" when calling subroutines, ...), etc.\n\nperlguts\n\n• A new section has been added, \"Dynamic Scope and the Context Stack\" in perlguts, which\nexplains how the perl context stack works.\n\nperllocale\n\n• A stronger caution about using locales in threaded applications is given. Locales are\nnot thread-safe, and you can get wrong results or even segfaults if you use them there.\n\nperlmodlib\n\n• We now recommend contacting the module-authors list or PAUSE in seeking guidance on the\nnaming of modules.\n\nperlop\n\n• The documentation of \"qx//\" now describes how $? is affected.\n\nperlpolicy\n\n• This note has been added to perlpolicy:\n\nWhile civility is required, kindness is encouraged; if you have any\ndoubt about whether you are being civil, simply ask yourself, \"Am I\nbeing kind?\" and aspire to that.\n\nperlreftut\n\n• Fix some examples to be strict clean.\n\nperlrebackslash\n\n• Clarify that in languages like Japanese and Thai, dictionary lookup is required to\ndetermine word boundaries.\n\nperlsub\n\n• Updated to note that anonymous subroutines can have signatures.\n\nperlsyn\n\n• Fixed a broken example where \"=\" was used instead of \"==\" in conditional in do/while\nexample.\n\nperltie\n\n• The usage of \"FIRSTKEY\" and \"NEXTKEY\" has been clarified.\n\nperlunicode\n\n• Discourage use of 'In' as a prefix signifying the Unicode Block property.\n\nperlvar\n\n• The documentation of $@ was reworded to clarify that it is not just for syntax errors in\n\"eval\". [GH #14572] \n\n• The specific true value of $!{E...} is now documented, noting that it is subject to\nchange and not guaranteed.\n\n• Use of $OLDPERLVERSION is now discouraged.\n\nperlxs\n\n• The documentation of \"PROTOTYPES\" has been corrected; they are disabled by default, not\nenabled.\n"
}
]
},
"Diagnostics": {
"content": "The 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",
"subsections": [
{
"name": "New Diagnostics",
"content": "New Errors\n\n• %s must not be a named sequence in transliteration operator\n\n• Can't find Unicode property definition \"%s\" in regex;\n\n• Can't redeclare \"%s\" in \"%s\"\n\n• Character following \\p must be '{' or a single-character Unicode property name in regex;\n\n• Empty \\%c in regex; marked by <-- HERE in m/%s/\n\n• Illegal user-defined property name\n\n• Invalid number '%s' for -C option.\n\n• Sequence (?... not terminated in regex; marked by <-- HERE in m/%s/\n\n• Sequence (?P<... not terminated in regex; marked by <-- HERE in m/%s/\n\n• Sequence (?P>... not terminated in regex; marked by <-- HERE in m/%s/\n\nNew Warnings\n\n• Assuming NOT a POSIX class since %s in regex; marked by <-- HERE in m/%s/\n\n• %s() is deprecated on :utf8 handles\n"
},
{
"name": "Changes to Existing Diagnostics",
"content": "• Accessing the \"IO\" part of a glob as \"FILEHANDLE\" instead of \"IO\" is no longer\ndeprecated. It is discouraged to encourage uniformity (so that, for example, one can\ngrep more easily) but it will not be removed. [GH #15105]\n\n\n• The diagnostic \"Hexadecimal float: internal error\" has been changed to \"Hexadecimal\nfloat: internal error (%s)\" to include more information.\n\n• Can't modify non-lvalue subroutine call of &%s\n\nThis error now reports the name of the non-lvalue subroutine you attempted to use as an\nlvalue.\n\n• When running out of memory during an attempt the increase the stack size, previously,\nperl would die using the cryptic message \"panic: avextendguts() negative count\n(-9223372036854775681)\". This has been fixed to show the prettier message: Out of memory\nduring stack extend\n"
}
]
},
"Configuration and Compilation": {
"content": "• \"Configure\" now acts as if the \"-O\" option is always passed, allowing command line\noptions to override saved configuration. This should eliminate confusion when command\nline options are ignored for no obvious reason. \"-O\" is now permitted, but ignored.\n\n• Bison 3.0 is now supported.\n\n• Configure no longer probes for libnm by default. Originally this was the \"New Math\"\nlibrary, but the name has been re-used by the GNOME NetworkManager. [GH #15115]\n\n\n• Added Configure probes for \"newlocale\", \"freelocale\", and \"uselocale\".\n\n• \"PPPort.so/PPPort.dll\" no longer get installed, as they are not used by \"PPPort.pm\", only\nby its test files.\n\n• It is now possible to specify which compilation date to show on \"perl -V\" output, by\nsetting the macro \"PERLBUILDDATE\".\n\n• Using the \"NOHASHSEED\" define in combination with the default hash algorithm\n\"PERLHASHFUNCONEATATIMEHARD\" resulted in a fatal error while compiling the\ninterpreter, since Perl 5.17.10. This has been fixed.\n\n• Configure should handle spaces in paths a little better.\n\n• No longer generate EBCDIC POSIX-BC tables. We don't believe anyone is using Perl and\nPOSIX-BC at this time, and by not generating these tables it saves time during\ndevelopment, and makes the resulting tar ball smaller.\n\n• The GNU Make makefile for Win32 now supports parallel builds. [perl #126632]\n\n• You can now build perl with MSVC++ on Win32 using GNU Make. [perl #126632]\n\n• The Win32 miniperl now has a real \"getcwd\" which increases build performance resulting in\ngetcwd() being 605x faster in Win32 miniperl.\n\n• Configure now takes \"-Dusequadmath\" into account when calculating the \"alignbytes\"\nconfiguration variable. Previously the mis-calculated \"alignbytes\" could cause alignment\nerrors on debugging builds. [perl #127894]\n",
"subsections": []
},
"Testing": {
"content": "• A new test (t/op/aassign.t) has been added to test the list assignment operator\n\"OPAASSIGN\".\n\n• Parallel building has been added to the dmake \"makefile.mk\" makefile. All Win32 compilers\nare supported.\n",
"subsections": []
},
"Platform Support": {
"content": "",
"subsections": [
{
"name": "Platform-Specific Notes",
"content": "AmigaOS\n• The AmigaOS port has been reintegrated into the main tree, based off of Perl 5.22.1.\n\nCygwin\n• Tests are more robust against unusual cygdrive prefixes. [GH #15076]\n\n\nEBCDIC\nUTF-EBCDIC extended\nUTF-EBCDIC is like UTF-8, but for EBCDIC platforms. It now has been extended so that\nit can represent code points up to 2 64 - 1 on platforms with 64-bit words. This\nbrings it into parity with UTF-8. This enhancement requires an incompatible change\nto the representation of code points in the range 2 30 to 2 31 -1 (the latter\nwas the previous maximum representable code point). This means that a file that\ncontains one of these code points, written out with previous versions of perl cannot\nbe read in, without conversion, by a perl containing this change. We do not believe\nany such files are in existence, but if you do have one, submit a ticket at\nperlbug@perl.org , and we will write a conversion script for\nyou.\n\nEBCDIC cmp() and sort() fixed for UTF-EBCDIC strings\nComparing two strings that were both encoded in UTF-8 (or more precisely, UTF-EBCDIC)\ndid not work properly until now. Since sort() uses cmp(), this fixes that as well.\n\nEBCDIC \"tr///\" and \"y///\" fixed for \"\\N{}\", and \"use utf8\" ranges\nPerl v5.22 introduced the concept of portable ranges to regular expression patterns.\nA portable range matches the same set of characters no matter what platform is being\nrun on. This concept is now extended to \"tr///\". See \"tr///\".\n\nThere were also some problems with these operations under \"use utf8\", which are now\nfixed\n\nFreeBSD\n• Use the fdclose() function from FreeBSD if it is available. [GH #15082]\n\n\nIRIX\n• Under some circumstances IRIX stdio fgetc() and fread() set the errno to \"ENOENT\",\nwhich made no sense according to either IRIX or POSIX docs. Errno is now cleared in\nsuch cases. [GH #14557] \n\n• Problems when multiplying long doubles by infinity have been fixed. [GH #14993]\n\n\nMacOS X\n• Until now OS X builds of perl have specified a link target of 10.3 (Panther, 2003)\nbut have not specified a compiler target. From now on, builds of perl on OS X 10.6\nor later (Snow Leopard, 2008) by default capture the current OS X version and specify\nthat as the explicit build target in both compiler and linker flags, thus preserving\nbinary compatibility for extensions built later regardless of changes in OS X, SDK,\nor compiler and linker versions. To override the default value used in the build and\npreserved in the flags, specify \"export MACOSXDEPLOYMENTTARGET=10.N\" before\nconfiguring and building perl, where 10.N is the version of OS X you wish to target.\nIn OS X 10.5 or earlier there is no change to the behavior present when those systems\nwere current; the link target is still OS X 10.3 and there is no explicit compiler\ntarget.\n\n• Builds with both -DDEBUGGING and threading enabled would fail with a \"panic: free\nfrom wrong pool\" error when built or tested from Terminal on OS X. This was caused\nby perl's internal management of the environment conflicting with an atfork handler\nusing the libc setenv() function to update the environment.\n\nPerl now uses setenv()/unsetenv() to update the environment on OS X. [GH #14955]\n\n\nSolaris\n• All Solaris variants now build a shared libperl\n\nSolaris and variants like OpenIndiana now always build with the shared Perl library\n(Configure -Duseshrplib). This was required for the OpenIndiana builds, but this has\nalso been the setting for Oracle/Sun Perl builds for several years.\n\nTru64\n• Workaround where Tru64 balks when prototypes are listed as \"PERLSTATICINLINE\", but\nwhere the test is build with \"-DPERLNOINLINEFUNCTIONS\".\n\nVMS\n• On VMS, the math function prototypes in \"math.h\" are now visible under C++. Now\nbuilding the POSIX extension with C++ will no longer crash.\n\n• VMS has had \"setenv\"/\"unsetenv\" since v7.0 (released in 1996), \"Perlvmssetenv\" now\nalways uses \"setenv\"/\"unsetenv\".\n\n• Perl now implements its own \"killpg\" by scanning for processes in the specified\nprocess group, which may not mean exactly the same thing as a Unix process group, but\nallows us to send a signal to a parent (or master) process and all of its sub-\nprocesses. At the perl level, this means we can now send a negative pid like so:\n\nkill SIGKILL, -$pid;\n\nto signal all processes in the same group as $pid.\n\n• For those %ENV elements based on the CRTL environ array, we've always preserved case\nwhen setting them but did look-ups only after upcasing the key first, which made\nlower- or mixed-case entries go missing. This problem has been corrected by making\n%ENV elements derived from the environ array case-sensitive on look-up as well as\ncase-preserving on store.\n\n• Environment look-ups for \"PERL5LIB\" and \"PERLLIB\" previously only considered logical\nnames, but now consider all sources of %ENV as determined by \"PERLENVTABLES\" and as\ndocumented in \"%ENV\" in perlvms.\n\n• The minimum supported version of VMS is now v7.3-2, released in 2003. As a side\neffect of this change, VAX is no longer supported as the terminal release of OpenVMS\nVAX was v7.3 in 2001.\n\nWin32\n• A new build option \"USENOREGISTRY\" has been added to the makefiles. This option is\noff by default, meaning the default is to do Windows registry lookups. This option\nstops Perl from looking inside the registry for anything. For what values are looked\nup in the registry see perlwin32. Internally, in C, the name of this option is\n\"WIN32NOREGISTRY\".\n\n• The behavior of Perl using \"HKEYCURRENTUSER\\Software\\Perl\" and\n\"HKEYLOCALMACHINE\\Software\\Perl\" to lookup certain values, including %ENV vars\nstarting with \"PERL\" has changed. Previously, the 2 keys were checked for entries at\nall times through the perl process's life time even if they did not exist. For\nperformance reasons, now, if the root key (i.e. \"HKEYCURRENTUSER\\Software\\Perl\" or\n\"HKEYLOCALMACHINE\\Software\\Perl\") does not exist at process start time, it will not\nbe checked again for %ENV override entries for the remainder of the perl process's\nlife. This more closely matches Unix behavior in that the environment is copied or\ninherited on startup and changing the variable in the parent process or another\nprocess or editing .bashrc will not change the environmental variable in other\nexisting, running, processes.\n\n• One glob fetch was removed for each \"-X\" or \"stat\" call whether done from Perl code\nor internally from Perl's C code. The glob being looked up was\n\"${^WIN32SLOPPYSTAT}\" which is a special variable. This makes \"-X\" and \"stat\"\nslightly faster.\n\n• During miniperl's process startup, during the build process, 4 to 8 IO calls related\nto the process starting .pl and the buildcustomize.pl file were removed from the code\nopening and executing the first 1 or 2 .pl files.\n\n• Builds using Microsoft Visual C++ 2003 and earlier no longer produce an \"INTERNAL\nCOMPILER ERROR\" message. [perl #126045]\n\n• Visual C++ 2013 builds will now execute on XP and higher. Previously they would only\nexecute on Vista and higher.\n\n• You can now build perl with GNU Make and GCC. [perl #123440]\n\n• \"truncate($filename, $size)\" now works for files over 4GB in size. [perl #125347]\n\n• Parallel building has been added to the dmake \"makefile.mk\" makefile. All Win32\ncompilers are supported.\n\n• Building a 64-bit perl with a 64-bit GCC but a 32-bit gmake would result in an\ninvalid $Config{archname} for the resulting perl. [perl #127584]\n\n• Errors set by Winsock functions are now put directly into $^E, and the relevant\n\"WSAE*\" error codes are now exported from the Errno and POSIX modules for testing\nthis against.\n\nThe previous behavior of putting the errors (converted to POSIX-style \"E*\" error\ncodes since Perl 5.20.0) into $! was buggy due to the non-equivalence of like-named\nWinsock and POSIX error constants, a relationship between which has unfortunately\nbeen established in one way or another since Perl 5.8.0.\n\nThe new behavior provides a much more robust solution for checking Winsock errors in\nportable software without accidentally matching POSIX tests that were intended for\nother OSes and may have different meanings for Winsock.\n\nThe old behavior is currently retained, warts and all, for backwards compatibility,\nbut users are encouraged to change any code that tests $! against \"E*\" constants for\nWinsock errors to instead test $^E against \"WSAE*\" constants. After a suitable\ndeprecation period, the old behavior may be removed, leaving $! unchanged after\nWinsock function calls, to avoid any possible confusion over which error variable to\ncheck.\n\nppc64el\nfloating point\nThe floating point format of ppc64el (Debian naming for little-endian PowerPC) is now\ndetected correctly.\n"
}
]
},
"Internal Changes": {
"content": "• The implementation of perl's context stack system, and its internal API, have been\nheavily reworked. Note that no significant changes have been made to any external APIs,\nbut XS code which relies on such internal details may need to be fixed. The main changes\nare:\n\n• The PUSHBLOCK(), POPSUB() etc. macros have been replaced with static inline functions\nsuch as cxpushblock(), cxpopsub() etc. These use function args rather than\nimplicitly relying on local vars such as \"gimme\" and \"newsp\" being available. Also\ntheir functionality has changed: in particular, cxpopblock() no longer decrements\n\"cxstackix\". The ordering of the steps in the \"ppleave*\" functions involving\ncxpopblock(), cxpopsub() etc. has changed. See the new documentation, \"Dynamic\nScope and the Context Stack\" in perlguts, for details on how to use them.\n\n• Various macros, which now consistently have a CX prefix, have been added:\n\nCXCUR(), CXLEAVESCOPE(), CXPOP()\n\nor renamed:\n\nCXPOPSAVEARRAY(), CXDEBUG(), CXPUSHSUBST(), CXPOPSUBST()\n\n• cxpushblock() now saves \"PLsavestackix\" and \"PLtmpsfloor\", so \"ppenter*\" and\n\"ppleave*\" no longer do\n\nENTER; SAVETMPS; ....; LEAVE\n\n• cxpopblock() now also restores \"PLcurpm\".\n\n• In dounwind() for every context type, the current savestack frame is now processed\nbefore each context is popped; formerly this was only done for sub-like context\nframes. This action has been removed from cxpopsub() and placed into its own macro,\nCXLEAVESCOPE(cx), which must be called before cxpopsub() etc.\n\ndounwind() now also does a cxpopblock() on the last popped frame (formerly it only\ndid the cxpopsub() etc. actions on each frame).\n\n• The temps stack is now freed on scope exit; previously, temps created during the last\nstatement of a block wouldn't be freed until the next \"nextstate\" following the block\n(apart from an existing hack that did this for recursive subs in scalar context); and\nin something like \"f(g())\", the temps created by the last statement in g() would\nformerly not be freed until the statement following the return from f().\n\n• Most values that were saved on the savestack on scope entry are now saved in suitable\nnew fields in the context struct, and saved and restored directly by cxpushfoo() and\ncxpopfoo(), which is much faster.\n\n• Various context struct fields have been added, removed or modified.\n\n• The handling of @ in cxpushsub() and cxpopsub() has been considerably tidied up,\nincluding removing the \"argarray\" field from the context struct, and extracting out\nsome common (but rarely used) code into a separate function, cleardefarray(). Also,\nuseful subsets of cxpopsub() which had been unrolled in places like \"ppgoto\" have\nbeen gathered into the new functions cxpopsubargs() and cxpopsubcommon().\n\n• \"ppleavesub\" and \"ppleavesublv\" now use the same function as the rest of the\n\"ppleave*\"'s to process return args.\n\n• \"CXpFORPAD\" and \"CXpFORGV\" flags have been added, and \"CXtLOOPFOR\" has been\nsplit into \"CXtLOOPLIST\", \"CXtLOOPARY\".\n\n• Some variables formerly declared by \"dMULTICALL\" (but not documented) have been\nremoved.\n\n• The obscure \"PLtimesbuf\" variable, effectively a vestige of Perl 1, has been removed. It\nwas documented as deprecated in Perl 5.20, with a statement that it would be removed\nearly in the 5.21.x series; that has now finally happened. [GH #13632]\n\n\n• An unwarranted assertion in PerlnewATTRSUBx() has been removed. If a stub subroutine\ndefinition with a prototype has been seen, then any subsequent stub (or definition) of\nthe same subroutine with an attribute was causing an assertion failure because of a null\npointer. [GH #15081] \n\n• \"::\" has been replaced by \"\" in \"ExtUtils::ParseXS\", like it's done for\nparameters/return values. This is more consistent, and simplifies writing XS code\nwrapping C++ classes into a nested Perl namespace (it requires only a typedef for\n\"FooBar\" rather than two, one for \"FooBar\" and the other for \"Foo::Bar\").\n\n• The toutf8case() function is now deprecated. Instead use \"toUPPERutf8\",\n\"toTITLEutf8\", \"toLOWERutf8\", and \"toFOLDutf8\". (See\n.)\n\n• Perl core code and the threads extension have been annotated so that, if Perl is\nconfigured to use threads, then during compile-time clang (3.6 or later) will warn about\nsuspicious uses of mutexes. See \nfor more information.\n\n• The signbit() emulation has been enhanced. This will help older and/or more exotic\nplatforms or configurations.\n\n• Most EBCDIC-specific code in the core has been unified with non-EBCDIC code, to avoid\nrepetition and make maintenance easier.\n\n• MSWin32 code for $^X has been moved out of the win32 directory to caretx.c, where other\noperating systems set that variable.\n\n• svref() is now part of the API.\n\n• \"svbackoff\" in perlapi had its return type changed from \"int\" to \"void\". It previously\nhas always returned 0 since Perl 5.000 stable but that was undocumented. Although\n\"svbackoff\" is marked as public API, XS code is not expected to be impacted since the\nproper API call would be through public API \"svsetsv(sv, &PLsvundef)\", or quasi-public\n\"SvOOKoff\", or non-public \"SvOKoff\" calls, and the return value of \"svbackoff\" was\npreviously a meaningless constant that can be rewritten as \"(svbackoff(sv),0)\".\n\n• The \"EXTEND\" and \"MEXTEND\" macros have been improved to avoid various issues with integer\ntruncation and wrapping. In particular, some casts formerly used within the macros have\nbeen removed. This means for example that passing an unsigned \"nitems\" argument is\nlikely to raise a compiler warning now (it's always been documented to require a signed\nvalue; formerly int, lately SSizet).\n\n• \"PLsawalias\" and \"GPfALIASEDSV\" have been removed.\n\n• \"GvASSIGNGENERATION\" and \"GvASSIGNGENERATIONset\" have been removed.\n",
"subsections": []
},
"Selected Bug Fixes": {
"content": "• It now works properly to specify a user-defined property, such as\n\nqr/\\p{mypkg1::IsMyProperty}/i\n\nwith \"/i\" caseless matching, an explicit package name, and IsMyProperty not defined at\nthe time of the pattern compilation.\n\n• Perl's memcpy(), memmove(), memset() and memcmp() fallbacks are now more compatible with\nthe originals. [perl #127619]\n\n• Fixed the issue where a \"s///r\") with -DPERLNOCOW attempts to modify the source SV,\nresulting in the program dying. [perl #127635]\n\n• Fixed an EBCDIC-platform-only case where a pattern could fail to match. This occurred\nwhen matching characters from the set of C1 controls when the target matched string was\nin UTF-8.\n\n• Narrow the filename check in strict.pm and warnings.pm. Previously, it assumed that if\nthe filename (without the .pmc? extension) differed from the package name, if was a\nmisspelled use statement (i.e. \"use Strict\" instead of \"use strict\"). We now check\nwhether there's really a miscapitalization happening, and not some other issue.\n\n• Turn an assertion into a more user friendly failure when parsing regexes. [perl #127599]\n\n• Correctly raise an error when trying to compile patterns with unterminated character\nclasses while there are trailing backslashes. [perl #126141].\n\n• Line numbers larger than 231-1 but less than 232 are no longer returned by caller()\nas negative numbers. [perl #126991]\n\n• \"unless ( assignment )\" now properly warns when syntax warnings are enabled. [perl\n#127122]\n\n• Setting an \"ISA\" glob to an array reference now properly adds \"isaelem\" magic to any\nexisting elements. Previously modifying such an element would not update the ISA cache,\nso method calls would call the wrong function. Perl would also crash if the \"ISA\" glob\nwas destroyed, since new code added in 5.23.7 would try to release the \"isaelem\" magic\nfrom the elements. [perl #127351]\n\n• If a here-doc was found while parsing another operator, the parser had already read end\nof file, and the here-doc was not terminated, perl could produce an assertion or a\nsegmentation fault. This now reliably complains about the unterminated here-doc. [perl\n#125540]\n\n• untie() would sometimes return the last value returned by the UNTIE() handler as well as\nits normal value, messing up the stack. [perl #126621]\n\n• Fixed an operator precedence problem when \" castflags & 2\" is true. [perl #127474]\n\n• Caching of DESTROY methods could result in a non-pointer or a non-STASH stored in the\nSvSTASH() slot of a stash, breaking the B STASH() method. The DESTROY method is now\ncached in the MRO metadata for the stash. [perl #126410]\n\n• The AUTOLOAD method is now called when searching for a DESTROY method, and correctly sets\n$AUTOLOAD too. [perl #124387] [perl #127494]\n\n• Avoid parsing beyond the end of the buffer when processing a \"#line\" directive with no\nfilename. [perl #127334]\n\n• Perl now raises a warning when a regular expression pattern looks like it was supposed to\ncontain a POSIX class, like \"qr/[[:alpha:]]/\", but there was some slight defect in its\nspecification which causes it to instead be treated as a regular bracketed character\nclass. An example would be missing the second colon in the above like this:\n\"qr/[[:alpha]]/\". This compiles to match a sequence of two characters. The second is\n\"]\", and the first is any of: \"[\", \":\", \"a\", \"h\", \"l\", or \"p\". This is unlikely to be\nthe intended meaning, and now a warning is raised. No warning is raised unless the\nspecification is very close to one of the 14 legal POSIX classes. (See \"POSIX Character\nClasses\" in perlrecharclass.) [perl #8904]\n\n• Certain regex patterns involving a complemented POSIX class in an inverted bracketed\ncharacter class, and matching something else optionally would improperly fail to match.\nAn example of one that could fail is \"qr/?[^\\Wbar]\\x{100}/\". This has been fixed.\n[perl #127537]\n\n• Perl 5.22 added support to the C99 hexadecimal floating point notation, but sometimes\nmisparses hex floats. This has been fixed. [perl #127183]\n\n• A regression that allowed undeclared barewords in hash keys to work despite strictures\nhas been fixed. [GH #15099] \n\n• Calls to the placeholder &PLsvyes used internally when an import() or unimport() method\nisn't found now correctly handle scalar context. [GH #14902]\n\n\n• Report more context when we see an array where we expect to see an operator and avoid an\nassertion failure. [GH #14472] \n\n• Modifying an array that was previously a package @ISA no longer causes assertion failures\nor crashes. [GH #14492] \n\n• Retain binary compatibility across plain and DEBUGGING perl builds. [GH #15122]\n\n\n• Avoid leaking memory when setting $ENV{foo} on darwin. [GH #14955]\n\n\n• \"/...\\G/\" no longer crashes on utf8 strings. When \"\\G\" is a fixed number of characters\nfrom the start of the regex, perl needs to count back that many characters from the\ncurrent pos() position and start matching from there. However, it was counting back bytes\nrather than characters, which could lead to panics on utf8 strings.\n\n• In some cases operators that return integers would return negative integers as large\npositive integers. [GH #15049] \n\n• The pipe() operator would assert for DEBUGGING builds instead of producing the correct\nerror message. The condition asserted on is detected and reported on correctly without\nthe assertions, so the assertions were removed. [GH #15015]\n\n\n• In some cases, failing to parse a here-doc would attempt to use freed memory. This was\ncaused by a pointer not being restored correctly. [GH #15009]\n\n\n• \"@x = sort { *a = 0; $a <=> $b } 0 .. 1\" no longer frees the GP for *a before restoring\nits SV slot. [GH #14595] \n\n• Multiple problems with the new hexadecimal floating point printf format %a were fixed:\n[GH #15032] , [GH #15033]\n, [GH #15074]\n\n\n• Calling mgset() in leavescope() no longer leaks.\n\n• A regression from Perl v5.20 was fixed in which debugging output of regular expression\ncompilation was wrong. (The pattern was correctly compiled, but what got displayed for\nit was wrong.)\n\n• \"\\b{sb}\" works much better. In Perl v5.22.0, this new construct didn't seem to give the\nexpected results, yet passed all the tests in the extensive suite furnished by Unicode.\nIt turns out that it was because these were short input strings, and the failures had to\ndo with longer inputs.\n\n• Certain syntax errors in \"Extended Bracketed Character Classes\" in perlrecharclass caused\npanics instead of the proper error message. This has now been fixed. [perl #126481]\n\n• Perl 5.20 added a message when a quantifier in a regular expression was useless, but then\ncaused the parser to skip it; this caused the surplus quantifier to be silently ignored,\ninstead of throwing an error. This is now fixed. [perl #126253]\n\n• The switch to building non-XS modules last in win32/makefile.mk (introduced by design as\npart of the changes to enable parallel building) caused the build of POSIX to break due\nto problems with the version module. This is now fixed.\n\n• Improved parsing of hex float constants.\n\n• Fixed an issue with \"pack\" where \"pack \"H\"\" (and \"pack \"h\"\") could read past the source\nwhen given a non-utf8 source, and a utf8 target. [perl #126325]\n\n• Fixed several cases where perl would abort due to a segmentation fault, or a C-level\nassert. [perl #126615], [perl #126602], [perl #126193].\n\n• There were places in regular expression patterns where comments (\"(?#...)\") weren't\nallowed, but should have been. This is now fixed. [GH #12755]\n\n\n• Some regressions from Perl 5.20 have been fixed, in which some syntax errors in\n\"(?[...])\" constructs within regular expression patterns could cause a segfault instead\nof a proper error message. [GH #14933] [GH\n#14996] \n\n• Another problem with \"(?[...])\" constructs has been fixed wherein things like \"\\c]\"\ncould cause panics. [GH #14934] \n\n• Some problems with attempting to extend the perl stack to around 2G or 4G entries have\nbeen fixed. This was particularly an issue on 32-bit perls built to use 64-bit integers,\nand was easily noticeable with the list repetition operator, e.g.\n\n@a = (1) x $bignumber\n\nFormerly perl may have crashed, depending on the exact value of $bignumber; now it will\ntypically raise an exception. [GH #14880] \n\n• In a regex conditional expression \"(?(condition)yes-pattern|no-pattern)\", if the\ncondition is \"(?!)\" then perl failed the match outright instead of matching the no-\npattern. This has been fixed. [GH #14947] \n\n• The special backtracking control verbs \"(*VERB:ARG)\" now all allow an optional argument\nand set \"REGERROR\"/\"REGMARK\" appropriately as well. [GH #14937]\n\n\n• Several bugs, including a segmentation fault, have been fixed with the boundary checking\nconstructs (introduced in Perl 5.22) \"\\b{gcb}\", \"\\b{sb}\", \"\\b{wb}\", \"\\B{gcb}\", \"\\B{sb}\",\nand \"\\B{wb}\". All the \"\\B{}\" ones now match an empty string; none of the \"\\b{}\" ones do.\n[GH #14976] \n\n• Duplicating a closed file handle for write no longer creates a filename of the form\nGLOB(0xXXXXXXXX). [perl #125115]\n\n• Warning fatality is now ignored when rewinding the stack. This prevents infinite\nrecursion when the now fatal error also causes rewinding of the stack. [perl #123398]\n\n• In perl v5.22.0, the logic changed when parsing a numeric parameter to the -C option,\nsuch that the successfully parsed number was not saved as the option value if it parsed\nto the end of the argument. [perl #125381]\n\n• The PadlistNAMES macro is an lvalue again.\n\n• Zero -DPERLTRACEOPS memory for sub-threads.\n\nperlcloneusing() was missing Zero init of PLopexeccnt[]. This caused sub-threads in\nthreaded -DPERLTRACEOPS builds to spew exceedingly large op-counts at destruct. These\ncounts would print %x as \"ABABABAB\", clearly a mem-poison value.\n\n• A leak in the XS typemap caused one scalar to be leaked each time a \"FILE *\" or a \"PerlIO\n*\" was \"OUTPUT:\"ed or imported to Perl, since perl 5.000. These particular typemap\nentries are thought to be extremely rarely used by XS modules. [perl #124181]\n\n• alarm() and sleep() will now warn if the argument is a negative number and return undef.\nPreviously they would pass the negative value to the underlying C function which may have\nset up a timer with a surprising value.\n\n• Perl can again be compiled with any Unicode version. This used to (mostly) work, but was\nlost in v5.18 through v5.20. The property \"NameAlias\" did not exist prior to Unicode\n5.0. Unicode::UCD incorrectly said it did. This has been fixed.\n\n• Very large code-points (beyond Unicode) in regular expressions no longer cause a buffer\noverflow in some cases when converted to UTF-8. [GH #14858]\n\n\n• The integer overflow check for the range operator (...) in list context now correctly\nhandles the case where the size of the range is larger than the address space. This\ncould happen on 32-bits with -Duse64bitint. [GH #14843]\n\n\n• A crash with \"%::=(); J->${\\\"::\"}\" has been fixed. [GH #14790]\n\n\n• \"qr/(?[ () ])/\" no longer segfaults, giving a syntax error message instead. [perl\n#125805]\n\n• Regular expression possessive quantifier v5.20 regression now fixed.\n\"qr/\"PAT\"{\"min,max\"}+\"\"/\" is supposed to behave identically to\n\"qr/(?>\"PAT\"{\"min,max\"})/\". Since v5.20, this didn't work if min and max were equal.\n[perl #125825]\n\n• \"BEGIN <>\" no longer segfaults and properly produces an error message. [perl #125341]\n\n• In \"tr///\" an illegal backwards range like \"tr/\\x{101}-\\x{100}//\" was not always\ndetected, giving incorrect results. This is now fixed.\n",
"subsections": []
},
"Acknowledgements": {
"content": "Perl 5.24.0 represents approximately 11 months of development since Perl 5.24.0 and contains\napproximately 360,000 lines of changes across 1,800 files from 75 authors.\n\nExcluding auto-generated files, documentation and release tools, there were approximately\n250,000 lines of changes to 1,200 .pm, .t, .c and .h files.\n\nPerl continues to flourish into its third decade thanks to a vibrant community of users and\ndevelopers. The following people are known to have contributed the improvements that became\nPerl 5.24.0:\n\nAaron Crane, Aaron Priven, Abigail, Achim Gratz, Alexander D'Archangel, Alex Vandiver,\nAndreas König, Andy Broad, Andy Dougherty, Aristotle Pagaltzis, Chase Whitener, Chas. Owens,\nChris 'BinGOs' Williams, Craig A. Berry, Dagfinn Ilmari Mannsåker, Dan Collins, Daniel\nDragan, David Golden, David Mitchell, Doug Bell, Dr.Ruud, Ed Avis, Ed J, Father Chrysostomos,\nHerbert Breunung, H.Merijn Brand, Hugo van der Sanden, Ivan Pozdeev, James E Keenan, Jan\nDubois, Jarkko Hietaniemi, Jerry D. Hedden, Jim Cromie, John Peacock, John SJ Anderson, Karen\nEtheridge, Karl Williamson, kmx, Leon Timmermans, Ludovic E. R. Tolhurst-Cleaver, Lukas Mai,\nMartijn Lievaart, Matthew Horsfall, Mattia Barbon, Max Maischein, Mohammed El-Afifi, Nicholas\nClark, Nicolas R., Niko Tyni, Peter John Acklam, Peter Martini, Peter Rabbitson, Pip Cet,\nRafael Garcia-Suarez, Reini Urban, Ricardo Signes, Sawyer X, Shlomi Fish, Sisyphus, Stanislaw\nPusep, Steffen Müller, Stevan Little, Steve Hay, Sullivan Beck, Thomas Sibley, Todd Rinaldo,\nTom Hukins, Tony Cook, Unicode Consortium, Victor Adam, Vincent Pit, Vladimir Timofeev, Yves\nOrton, Zachary Storer, Zefram.\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": []
},
"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 https://rt.perl.org/ . There may\nalso 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 to\na publicly archived mailing list, then see \"SECURITY VULNERABILITY CONTACT INFORMATION\" in\nperlsec for details of how to report the issue.\n",
"subsections": []
},
"SEE ALSO": {
"content": "The Changes file for an explanation of how to view exhaustive details on what changed.\n\nThe INSTALL file for how to build Perl.\n\nThe README file for general stuff.\n\nThe Artistic and Copying files for copyright information.\n\nperl v5.38.2 2026-08-18 PERL5240DELTA(1)",
"subsections": []
}
},
"summary": "perl5240delta - what is new for perl v5.24.0",
"flags": [],
"examples": [],
"see_also": [],
"tldr": {
"source": "not_found",
"examples": []
}
}