{ "mode": "man", "parameter": "perl561delta", "section": "1", "url": "https://www.chedong.com/phpMan.php/man/perl561delta/1/json", "generated": "2026-08-22T16:07:09Z", "sections": { "NAME": { "content": "perl561delta - what's new for perl v5.6.1\n", "subsections": [] }, "DESCRIPTION": { "content": "This document describes differences between the 5.005 release and the 5.6.1 release.\n", "subsections": [] }, "Summary of changes between 5.6.0 and 5.6.1": { "content": "This section contains a summary of the changes between the 5.6.0 release and the 5.6.1\nrelease. More details about the changes mentioned here may be found in the Changes files\nthat accompany the Perl source distribution. See perlhack for pointers to online resources\nwhere you can inspect the individual patches described by these changes.\n", "subsections": [ { "name": "Security Issues", "content": "suidperl will not run /bin/mail anymore, because some platforms have a /bin/mail that is\nvulnerable to buffer overflow attacks.\n\nNote that suidperl is neither built nor installed by default in any recent version of perl.\nUse of suidperl is highly discouraged. If you think you need it, try alternatives such as\nsudo first. See http://www.courtesan.com/sudo/ .\n" }, { "name": "Core bug fixes", "content": "This is not an exhaustive list. It is intended to cover only the significant user-visible\nchanges.\n\nUNIVERSAL::isa()\nA bug in the caching mechanism used by UNIVERSAL::isa() that affected base.pm has been\nfixed. The bug has existed since the 5.005 releases, but wasn't tickled by base.pm in\nthose releases.\n\nMemory leaks\nVarious cases of memory leaks and attempts to access uninitialized memory have been\ncured. See \"Known Problems\" below for further issues.\n\nNumeric conversions\nNumeric conversions did not recognize changes in the string value properly in certain\ncircumstances.\n\nIn other situations, large unsigned numbers (those above 231) could sometimes lose\ntheir unsignedness, causing bogus results in arithmetic operations.\n\nInteger modulus on large unsigned integers sometimes returned incorrect values.\n\nPerl 5.6.0 generated \"not a number\" warnings on certain conversions where previous\nversions didn't.\n\nThese problems have all been rectified.\n\nInfinity is now recognized as a number.\n\nqw(a\\\\b)\nIn Perl 5.6.0, qw(a\\\\b) produced a string with two backslashes instead of one, in a\ndeparture from the behavior in previous versions. The older behavior has been\nreinstated.\n" }, { "name": "caller()", "content": "caller() could cause core dumps in certain situations. Carp was sometimes affected by\nthis problem.\n\nBugs in regular expressions\nPattern matches on overloaded values are now handled correctly.\n\nPerl 5.6.0 parsed m/\\x{ab}/ incorrectly, leading to spurious warnings. This has been\ncorrected.\n\nThe RE engine found in Perl 5.6.0 accidentally pessimised certain kinds of simple pattern\nmatches. These are now handled better.\n\nRegular expression debug output (whether through \"use re 'debug'\" or via \"-Dr\") now looks\nbetter.\n\nMulti-line matches like \"\"a\\nxb\\n\" =~ /(?!\\A)x/m\" were flawed. The bug has been fixed.\n\nUse of $& could trigger a core dump under some situations. This is now avoided.\n\nMatch variables $1 et al., weren't being unset when a pattern match was backtracking, and\nthe anomaly showed up inside \"/...(?{ ... }).../\" etc. These variables are now tracked\ncorrectly.\n\npos() did not return the correct value within s///ge in earlier versions. This is now\nhandled correctly.\n\n\"slurp\" mode\nreadline() on files opened in \"slurp\" mode could return an extra \"\" at the end in certain\nsituations. This has been corrected.\n\nAutovivification of symbolic references to special variables\nAutovivification of symbolic references of special variables described in perlvar (as in\n\"${$num}\") was accidentally disabled. This works again now.\n\nLexical warnings\nLexical warnings now propagate correctly into \"eval \"...\"\".\n\n\"use warnings qw(FATAL all)\" did not work as intended. This has been corrected.\n\nLexical warnings could leak into other scopes in some situations. This is now fixed.\n\nwarnings::enabled() now reports the state of $^W correctly if the caller isn't using\nlexical warnings.\n\nSpurious warnings and errors\nPerl 5.6.0 could emit spurious warnings about redefinition of dlerror() when statically\nbuilding extensions into perl. This has been corrected.\n\n\"our\" variables could result in bogus \"Variable will not stay shared\" warnings. This is\nnow fixed.\n\n\"our\" variables of the same name declared in two sibling blocks resulted in bogus\nwarnings about \"redeclaration\" of the variables. The problem has been corrected.\n" }, { "name": "glob()", "content": "Compatibility of the builtin glob() with old csh-based glob has been improved with the\naddition of GLOBALPHASORT option. See \"File::Glob\".\n\nFile::Glob::glob() has been renamed to File::Glob::bsdglob() because the name clashes\nwith the builtin glob(). The older name is still available for compatibility, but is\ndeprecated.\n\nSpurious syntax errors generated in certain situations, when glob() caused File::Glob to\nbe loaded for the first time, have been fixed.\n\nTainting\nSome cases of inconsistent taint propagation (such as within hash values) have been\nfixed.\n\nThe tainting behavior of sprintf() has been rationalized. It does not taint the result\nof floating point formats anymore, making the behavior consistent with that of string\ninterpolation.\n" }, { "name": "sort()", "content": "Arguments to sort() weren't being provided the right wantarray() context. The comparison\nblock is now run in scalar context, and the arguments to be sorted are always provided\nlist context.\n\nsort() is also fully reentrant, in the sense that the sort function can itself call\nsort(). This did not work reliably in previous releases.\n\n#line directives\n#line directives now work correctly when they appear at the very beginning of \"eval\n\"...\"\".\n\nSubroutine prototypes\nThe (\\&) prototype now works properly.\n" }, { "name": "map()", "content": "map() could get pathologically slow when the result list it generates is larger than the\nsource list. The performance has been improved for common scenarios.\n\nDebugger\nDebugger exit code now reflects the script exit code.\n\nCondition \"0\" in breakpoints is now treated correctly.\n\nThe \"d\" command now checks the line number.\n\n$. is no longer corrupted by the debugger.\n\nAll debugger output now correctly goes to the socket if RemotePort is set.\n\nPERL5OPT\nPERL5OPT can be set to more than one switch group. Previously, it used to be limited to\none group of options only.\n" }, { "name": "chop()", "content": "chop(@list) in list context returned the characters chopped in reverse order. This has\nbeen reversed to be in the right order.\n\nUnicode support\nUnicode support has seen a large number of incremental improvements, but continues to be\nhighly experimental. It is not expected to be fully supported in the 5.6.x maintenance\nreleases.\n\nsubstr(), join(), repeat(), reverse(), quotemeta() and string concatenation were all\nhandling Unicode strings incorrectly in Perl 5.6.0. This has been corrected.\n\nSupport for \"tr///CU\" and \"tr///UC\" etc., have been removed since we realized the\ninterface is broken. For similar functionality, see \"pack\" in perlfunc.\n\nThe Unicode Character Database has been updated to version 3.0.1 with additions made\navailable to the public as of August 30, 2000.\n\nThe Unicode character classes \\p{Blank} and \\p{SpacePerl} have been added. \"Blank\" is\nlike C isblank(), that is, it contains only \"horizontal whitespace\" (the space character\nis, the newline isn't), and the \"SpacePerl\" is the Unicode equivalent of \"\\s\" (\\p{Space}\nisn't, since that includes the vertical tabulator character, whereas \"\\s\" doesn't.)\n\nIf you are experimenting with Unicode support in perl, the development versions of Perl\nmay have more to offer. In particular, I/O layers are now available in the development\ntrack, but not in the maintenance track, primarily to do backward compatibility issues.\nUnicode support is also evolving rapidly on a daily basis in the development track--the\nmaintenance track only reflects the most conservative of these changes.\n\n64-bit support\nSupport for 64-bit platforms has been improved, but continues to be experimental. The\nlevel of support varies greatly among platforms.\n\nCompiler\nThe B Compiler and its various backends have had many incremental improvements, but they\ncontinue to remain highly experimental. Use in production environments is discouraged.\n\nThe perlcc tool has been rewritten so that the user interface is much more like that of a\nC compiler.\n\nThe perlbc tools has been removed. Use \"perlcc -B\" instead.\n\nLvalue subroutines\nThere have been various bugfixes to support lvalue subroutines better. However, the\nfeature still remains experimental.\n\nIO::Socket\nIO::Socket::INET failed to open the specified port if the service name was not known. It\nnow correctly uses the supplied port number as is.\n\nFile::Find\nFile::Find now chdir()s correctly when chasing symbolic links.\n\nxsubpp\nxsubpp now tolerates embedded POD sections.\n\n\"no Module;\"\n\"no Module;\" does not produce an error even if Module does not have an unimport() method.\nThis parallels the behavior of \"use\" vis-a-vis \"import\".\n\nTests\nA large number of tests have been added.\n" }, { "name": "Core features", "content": "untie() will now call an UNTIE() hook if it exists. See perltie for details.\n\nThe \"-DT\" command line switch outputs copious tokenizing information. See perlrun.\n\nArrays are now always interpolated in double-quotish strings. Previously, \"foo@bar.com\" used\nto be a fatal error at compile time, if an array @bar was not used or declared. This\ntransitional behavior was intended to help migrate perl4 code, and is deemed to be no longer\nuseful. See \"Arrays now always interpolate into double-quoted strings\".\n\nkeys(), each(), pop(), push(), shift(), splice() and unshift() can all be overridden now.\n\n\"my PACKAGE $obj\" now does the expected thing.\n" }, { "name": "Configuration issues", "content": "On some systems (IRIX and Solaris among them) the system malloc is demonstrably better.\nWhile the defaults haven't been changed in order to retain binary compatibility with earlier\nreleases, you may be better off building perl with \"Configure -Uusemymalloc ...\" as discussed\nin the INSTALL file.\n\n\"Configure\" has been enhanced in various ways:\n\n• Minimizes use of temporary files.\n\n• By default, does not link perl with libraries not used by it, such as the various dbm\nlibraries. SunOS 4.x hints preserve behavior on that platform.\n\n• Support for pdp11-style memory models has been removed due to obsolescence.\n\n• Building outside the source tree is supported on systems that have symbolic links. This\nis done by running\n\nsh /path/to/source/Configure -Dmksymlinks ...\nmake all test install\n\nin a directory other than the perl source directory. See INSTALL.\n\n• \"Configure -S\" can be run non-interactively.\n" }, { "name": "Documentation", "content": "README.aix, README.solaris and README.macos have been added. README.posix-bc has been\nrenamed to README.bs2000. These are installed as perlaix, perlsolaris, perlmacos, and\nperlbs2000 respectively.\n\nThe following pod documents are brand new:\n\nperlclib Internal replacements for standard C library functions\nperldebtut Perl debugging tutorial\nperlebcdic Considerations for running Perl on EBCDIC platforms\nperlnewmod Perl modules: preparing a new module for distribution\nperlrequick Perl regular expressions quick start\nperlretut Perl regular expressions tutorial\nperlutil utilities packaged with the Perl distribution\n\nThe INSTALL file has been expanded to cover various issues, such as 64-bit support.\n\nA longer list of contributors has been added to the source distribution. See the file\n\"AUTHORS\".\n\nNumerous other changes have been made to the included documentation and FAQs.\n" }, { "name": "Bundled modules", "content": "The following modules have been added.\n\nB::Concise\nWalks Perl syntax tree, printing concise info about ops. See B::Concise.\n\nFile::Temp\nReturns name and handle of a temporary file safely. See File::Temp.\n\nPod::LaTeX\nConverts Pod data to formatted LaTeX. See Pod::LaTeX.\n\nPod::Text::Overstrike\nConverts POD data to formatted overstrike text. See Pod::Text::Overstrike.\n\nThe following modules have been upgraded.\n\nCGI CGI v2.752 is now included.\n\nCPAN\nCPAN v1.5954 is now included.\n\nClass::Struct\nVarious bugfixes have been added.\n\nDBFile\nDBFile v1.75 supports newer Berkeley DB versions, among other improvements.\n\nDevel::Peek\nDevel::Peek has been enhanced to support dumping of memory statistics, when perl is built\nwith the included malloc().\n\nFile::Find\nFile::Find now supports pre and post-processing of the files in order to sort() them,\netc.\n\nGetopt::Long\nGetopt::Long v2.25 is included.\n\nIO::Poll\nVarious bug fixes have been included.\n\nIPC::Open3\nIPC::Open3 allows use of numeric file descriptors.\n\nMath::BigFloat\nThe fmod() function supports modulus operations. Various bug fixes have also been\nincluded.\n\nMath::Complex\nMath::Complex handles inf, NaN etc., better.\n\nNet::Ping\nping() could fail on odd number of data bytes, and when the echo service isn't running.\nThis has been corrected.\n\nOpcode\nA memory leak has been fixed.\n\nPod::Parser\nVersion 1.13 of the Pod::Parser suite is included.\n\nPod::Text\nPod::Text and related modules have been upgraded to the versions in podlators suite\nv2.08.\n\nSDBMFile\nOn dosish platforms, some keys went missing because of lack of support for files with\n\"holes\". A workaround for the problem has been added.\n\nSys::Syslog\nVarious bug fixes have been included.\n\nTie::RefHash\nNow supports Tie::RefHash::Nestable to automagically tie hashref values.\n\nTie::SubstrHash\nVarious bug fixes have been included.\n" }, { "name": "Platform-specific improvements", "content": "The following new ports are now available.\n\nNCR MP-RAS\nNonStop-UX\n\nPerl now builds under Amdahl UTS.\n\nPerl has also been verified to build under Amiga OS.\n\nSupport for EPOC has been much improved. See README.epoc.\n\nBuilding perl with -Duseithreads or -Duse5005threads now works under HP-UX 10.20 (previously\nit only worked under 10.30 or later). You will need a thread library package installed. See\nREADME.hpux.\n\nLong doubles should now work under Linux.\n\nMac OS Classic is now supported in the mainstream source package. See README.macos.\n\nSupport for MPE/iX has been updated. See README.mpeix.\n\nSupport for OS/2 has been improved. See \"os2/Changes\" and README.os2.\n\nDynamic loading on z/OS (formerly OS/390) has been improved. See README.os390.\n\nSupport for VMS has seen many incremental improvements, including better support for\noperators like backticks and system(), and better %ENV handling. See \"README.vms\" and\nperlvms.\n\nSupport for Stratus VOS has been improved. See \"vos/Changes\" and README.vos.\n\nSupport for Windows has been improved.\n\n• fork() emulation has been improved in various ways, but still continues to be\nexperimental. See perlfork for known bugs and caveats.\n\n• %SIG has been enabled under USEITHREADS, but its use is completely unsupported under all\nconfigurations.\n\n• Borland C++ v5.5 is now a supported compiler that can build Perl. However, the generated\nbinaries continue to be incompatible with those generated by the other supported\ncompilers (GCC and Visual C++).\n\n• Non-blocking waits for child processes (or pseudo-processes) are supported via\n\"waitpid($pid, &POSIX::WNOHANG)\".\n\n• A memory leak in accept() has been fixed.\n\n• wait(), waitpid() and backticks now return the correct exit status under Windows 9x.\n\n• Trailing new %ENV entries weren't propagated to child processes. This is now fixed.\n\n• Current directory entries in %ENV are now correctly propagated to child processes.\n\n• Duping socket handles with open(F, \">&MYSOCK\") now works under Windows 9x.\n\n• The makefiles now provide a single switch to bulk-enable all the features enabled in\nActiveState ActivePerl (a popular binary distribution).\n\n• Win32::GetCwd() correctly returns C:\\ instead of C: when at the drive root. Other bugs\nin chdir() and Cwd::cwd() have also been fixed.\n\n• fork() correctly returns undef and sets EAGAIN when it runs out of pseudo-process\nhandles.\n\n• ExtUtils::MakeMaker now uses $ENV{LIB} to search for libraries.\n\n• UNC path handling is better when perl is built to support fork().\n\n• A handle leak in socket handling has been fixed.\n\n• send() works from within a pseudo-process.\n\nUnless specifically qualified otherwise, the remainder of this document covers changes\nbetween the 5.005 and 5.6.0 releases.\n" } ] }, "Core Enhancements": { "content": "", "subsections": [ { "name": "Interpreter cloning, threads, and concurrency", "content": "Perl 5.6.0 introduces the beginnings of support for running multiple interpreters\nconcurrently in different threads. In conjunction with the perlclone() API call, which can\nbe used to selectively duplicate the state of any given interpreter, it is possible to\ncompile a piece of code once in an interpreter, clone that interpreter one or more times, and\nrun all the resulting interpreters in distinct threads.\n\nOn the Windows platform, this feature is used to emulate fork() at the interpreter level.\nSee perlfork for details about that.\n\nThis feature is still in evolution. It is eventually meant to be used to selectively clone a\nsubroutine and data reachable from that subroutine in a separate interpreter and run the\ncloned subroutine in a separate thread. Since there is no shared data between the\ninterpreters, little or no locking will be needed (unless parts of the symbol table are\nexplicitly shared). This is obviously intended to be an easy-to-use replacement for the\nexisting threads support.\n\nSupport for cloning interpreters and interpreter concurrency can be enabled using the\n-Dusethreads Configure option (see win32/Makefile for how to enable it on Windows.) The\nresulting perl executable will be functionally identical to one that was built with\n-Dmultiplicity, but the perlclone() API call will only be available in the former.\n\n-Dusethreads enables the cpp macro USEITHREADS by default, which in turn enables Perl source\ncode changes that provide a clear separation between the op tree and the data it operates\nwith. The former is immutable, and can therefore be shared between an interpreter and all of\nits clones, while the latter is considered local to each interpreter, and is therefore copied\nfor each clone.\n\nNote that building Perl with the -Dusemultiplicity Configure option is adequate if you wish\nto run multiple independent interpreters concurrently in different threads. -Dusethreads\nonly provides the additional functionality of the perlclone() API call and other support for\nrunning cloned interpreters concurrently.\n\nNOTE: This is an experimental feature. Implementation details are\nsubject to change.\n" }, { "name": "Lexically scoped warning categories", "content": "You can now control the granularity of warnings emitted by perl at a finer level using the\n\"use warnings\" pragma. warnings and perllexwarn have copious documentation on this feature.\n" }, { "name": "Unicode and UTF-8 support", "content": "Perl now uses UTF-8 as its internal representation for character strings. The \"utf8\" and\n\"bytes\" pragmas are used to control this support in the current lexical scope. See\nperlunicode, utf8 and bytes for more information.\n\nThis feature is expected to evolve quickly to support some form of I/O disciplines that can\nbe used to specify the kind of input and output data (bytes or characters). Until that\nhappens, additional modules from CPAN will be needed to complete the toolkit for dealing with\nUnicode.\n\nNOTE: This should be considered an experimental feature. Implementation\ndetails are subject to change.\n" }, { "name": "Support for interpolating named characters", "content": "The new \"\\N\" escape interpolates named characters within strings. For example, \"Hi! \\N{WHITE\nSMILING FACE}\" evaluates to a string with a Unicode smiley face at the end.\n" }, { "name": "\"our\" declarations", "content": "An \"our\" declaration introduces a value that can be best understood as a lexically scoped\nsymbolic alias to a global variable in the package that was current where the variable was\ndeclared. This is mostly useful as an alternative to the \"vars\" pragma, but also provides\nthe opportunity to introduce typing and other attributes for such variables. See \"our\" in\nperlfunc.\n" }, { "name": "Support for strings represented as a vector of ordinals", "content": "Literals of the form \"v1.2.3.4\" are now parsed as a string composed of characters with the\nspecified ordinals. This is an alternative, more readable way to construct (possibly\nUnicode) strings instead of interpolating characters, as in \"\\x{1}\\x{2}\\x{3}\\x{4}\". The\nleading \"v\" may be omitted if there are more than two ordinals, so 1.2.3 is parsed the same\nas \"v1.2.3\".\n\nStrings written in this form are also useful to represent version \"numbers\". It is easy to\ncompare such version \"numbers\" (which are really just plain strings) using any of the usual\nstring comparison operators \"eq\", \"ne\", \"lt\", \"gt\", etc., or perform bitwise string\noperations on them using \"|\", \"&\", etc.\n\nIn conjunction with the new $^V magic variable (which contains the perl version as a string),\nsuch literals can be used as a readable way to check if you're running a particular version\nof Perl:\n\n# this will parse in older versions of Perl also\nif ($^V and $^V gt v5.6.0) {\n# new features supported\n}\n\n\"require\" and \"use\" also have some special magic to support such literals. They will be\ninterpreted as a version rather than as a module name:\n\nrequire v5.6.0; # croak if $^V lt v5.6.0\nuse v5.6.0; # same, but croaks at compile-time\n\nAlternatively, the \"v\" may be omitted if there is more than one dot:\n\nrequire 5.6.0;\nuse 5.6.0;\n\nAlso, \"sprintf\" and \"printf\" support the Perl-specific format flag %v to print ordinals of\ncharacters in arbitrary strings:\n\nprintf \"v%vd\", $^V; # prints current version, such as \"v5.5.650\"\nprintf \"%*vX\", \":\", $addr; # formats IPv6 address\nprintf \"%*vb\", \" \", $bits; # displays bitstring\n\nSee \"Scalar value constructors\" in perldata for additional information.\n" }, { "name": "Improved Perl version numbering system", "content": "Beginning with Perl version 5.6.0, the version number convention has been changed to a\n\"dotted integer\" scheme that is more commonly found in open source projects.\n\nMaintenance versions of v5.6.0 will be released as v5.6.1, v5.6.2 etc. The next development\nseries following v5.6.0 will be numbered v5.7.x, beginning with v5.7.0, and the next major\nproduction release following v5.6.0 will be v5.8.0.\n\nThe English module now sets $PERLVERSION to $^V (a string value) rather than $] (a numeric\nvalue). (This is a potential incompatibility. Send us a report via perlbug if you are\naffected by this.)\n\nThe v1.2.3 syntax is also now legal in Perl. See \"Support for strings represented as a\nvector of ordinals\" for more on that.\n\nTo cope with the new versioning system's use of at least three significant digits for each\nversion component, the method used for incrementing the subversion number has also changed\nslightly. We assume that versions older than v5.6.0 have been incrementing the subversion\ncomponent in multiples of 10. Versions after v5.6.0 will increment them by 1. Thus, using\nthe new notation, 5.00503 is the \"same\" as v5.5.30, and the first maintenance version\nfollowing v5.6.0 will be v5.6.1 (which should be read as being equivalent to a floating point\nvalue of 5.006001 in the older format, stored in $]).\n" }, { "name": "New syntax for declaring subroutine attributes", "content": "Formerly, if you wanted to mark a subroutine as being a method call or as requiring an\nautomatic lock() when it is entered, you had to declare that with a \"use attrs\" pragma in the\nbody of the subroutine. That can now be accomplished with declaration syntax, like this:\n\nsub mymethod : locked method;\n...\nsub mymethod : locked method {\n...\n}\n\nsub othermethod :locked :method;\n...\nsub othermethod :locked :method {\n...\n}\n\n(Note how only the first \":\" is mandatory, and whitespace surrounding the \":\" is optional.)\n\nAutoSplit.pm and SelfLoader.pm have been updated to keep the attributes with the stubs they\nprovide. See attributes.\n" }, { "name": "File and directory handles can be autovivified", "content": "Similar to how constructs such as \"$x->[0]\" autovivify a reference, handle constructors\n(open(), opendir(), pipe(), socketpair(), sysopen(), socket(), and accept()) now autovivify a\nfile or directory handle if the handle passed to them is an uninitialized scalar variable.\nThis allows the constructs such as \"open(my $fh, ...)\" and \"open(local $fh,...)\" to be used\nto create filehandles that will conveniently be closed automatically when the scope ends,\nprovided there are no other references to them. This largely eliminates the need for\ntypeglobs when opening filehandles that must be passed around, as in the following example:\n\nsub myopen {\nopen my $fh, \"@\"\nor die \"Can't open '@': $!\";\nreturn $fh;\n}\n\n{\nmy $f = myopen(\";\n# $f implicitly closed here\n}\n" }, { "name": "open() with more than two arguments", "content": "If open() is passed three arguments instead of two, the second argument is used as the mode\nand the third argument is taken to be the file name. This is primarily useful for protecting\nagainst unintended magic behavior of the traditional two-argument form. See \"open\" in\nperlfunc.\n" }, { "name": "64-bit support", "content": "Any platform that has 64-bit integers either\n\n(1) natively as longs or ints\n(2) via special compiler flags\n(3) using long long or int64t\n\nis able to use \"quads\" (64-bit integers) as follows:\n\n• constants (decimal, hexadecimal, octal, binary) in the code\n\n• arguments to oct() and hex()\n\n• arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)\n\n• printed as such\n\n• pack() and unpack() \"q\" and \"Q\" formats\n\n• in basic arithmetics: + - * / % (NOTE: operating close to the limits of the integer\nvalues may produce surprising results)\n\n• in bit arithmetics: & | ^ ~ << >> (NOTE: these used to be forced to be 32 bits wide but\nnow operate on the full native width.)\n\n• vec()\n\nNote that unless you have the case (a) you will have to configure and compile Perl using the\n-Duse64bitint Configure flag.\n\nNOTE: The Configure flags -Duselonglong and -Duse64bits have been\ndeprecated. Use -Duse64bitint instead.\n\nThere are actually two modes of 64-bitness: the first one is achieved using Configure\n-Duse64bitint and the second one using Configure -Duse64bitall. The difference is that the\nfirst one is minimal and the second one maximal. The first works in more places than the\nsecond.\n\nThe \"use64bitint\" does only as much as is required to get 64-bit integers into Perl (this may\nmean, for example, using \"long longs\") while your memory may still be limited to 2 gigabytes\n(because your pointers could still be 32-bit). Note that the name \"64bitint\" does not imply\nthat your C compiler will be using 64-bit \"int\"s (it might, but it doesn't have to): the\n\"use64bitint\" means that you will be able to have 64 bits wide scalar values.\n\nThe \"use64bitall\" goes all the way by attempting to switch also integers (if it can), longs\n(and pointers) to being 64-bit. This may create an even more binary incompatible Perl than\n-Duse64bitint: the resulting executable may not run at all in a 32-bit box, or you may have\nto reboot/reconfigure/rebuild your operating system to be 64-bit aware.\n\nNatively 64-bit systems like Alpha and Cray need neither -Duse64bitint nor -Duse64bitall.\n\nLast but not least: note that due to Perl's habit of always using floating point numbers, the\nquads are still not true integers. When quads overflow their limits\n(0...18446744073709551615 unsigned,\n-9223372036854775808...9223372036854775807 signed), they are silently promoted to\nfloating point numbers, after which they will start losing precision (in their lower digits).\n\nNOTE: 64-bit support is still experimental on most platforms.\nExisting support only covers the LP64 data model. In particular, the\nLLP64 data model is not yet supported. 64-bit libraries and system\nAPIs on many platforms have not stabilized--your mileage may vary.\n" }, { "name": "Large file support", "content": "If you have filesystems that support \"large files\" (files larger than 2 gigabytes), you may\nnow also be able to create and access them from Perl.\n\nNOTE: The default action is to enable large file support, if\navailable on the platform.\n\nIf the large file support is on, and you have a Fcntl constant OLARGEFILE, the OLARGEFILE\nis automatically added to the flags of sysopen().\n\nBeware that unless your filesystem also supports \"sparse files\" seeking to umpteen petabytes\nmay be inadvisable.\n\nNote that in addition to requiring a proper file system to do large files you may also need\nto adjust your per-process (or your per-system, or per-process-group, or per-user-group)\nmaximum filesize limits before running Perl scripts that try to handle large files,\nespecially if you intend to write such files.\n\nFinally, in addition to your process/process group maximum filesize limits, you may have\nquota limits on your filesystems that stop you (your user id or your user group id) from\nusing large files.\n\nAdjusting your process/user/group/file system/operating system limits is outside the scope of\nPerl core language. For process limits, you may try increasing the limits using your shell's\nlimits/limit/ulimit command before running Perl. The BSD::Resource extension (not included\nwith the standard Perl distribution) may also be of use, it offers the getrlimit/setrlimit\ninterface that can be used to adjust process resource usage limits, including the maximum\nfilesize limit.\n" }, { "name": "Long doubles", "content": "In some systems you may be able to use long doubles to enhance the range and precision of\nyour double precision floating point numbers (that is, Perl's numbers). Use Configure\n-Duselongdouble to enable this support (if it is available).\n" }, { "name": "\"more bits\"", "content": "You can \"Configure -Dusemorebits\" to turn on both the 64-bit support and the long double\nsupport.\n" }, { "name": "Enhanced support for sort() subroutines", "content": "Perl subroutines with a prototype of \"($$)\", and XSUBs in general, can now be used as sort\nsubroutines. In either case, the two elements to be compared are passed as normal parameters\nin @. See \"sort\" in perlfunc.\n\nFor unprototyped sort subroutines, the historical behavior of passing the elements to be\ncompared as the global variables $a and $b remains unchanged.\n" }, { "name": "\"sort $coderef @foo\" allowed", "content": "sort() did not accept a subroutine reference as the comparison function in earlier versions.\nThis is now permitted.\n" }, { "name": "File globbing implemented internally", "content": "Perl now uses the File::Glob implementation of the glob() operator automatically. This\navoids using an external csh process and the problems associated with it.\n\nNOTE: This is currently an experimental feature. Interfaces and\nimplementation are subject to change.\n" }, { "name": "Support for CHECK blocks", "content": "In addition to \"BEGIN\", \"INIT\", \"END\", \"DESTROY\" and \"AUTOLOAD\", subroutines named \"CHECK\"\nare now special. These are queued up during compilation and behave similar to END blocks,\nexcept they are called at the end of compilation rather than at the end of execution. They\ncannot be called directly.\n" }, { "name": "POSIX character class syntax [: :] supported", "content": "For example to match alphabetic characters use /[[:alpha:]]/. See perlre for details.\n" }, { "name": "Better pseudo-random number generator", "content": "In 5.0050x and earlier, perl's rand() function used the C library rand(3) function. As of\n5.00552, Configure tests for drand48(), random(), and rand() (in that order) and picks the\nfirst one it finds.\n\nThese changes should result in better random numbers from rand().\n" }, { "name": "Improved \"qw//\" operator", "content": "The \"qw//\" operator is now evaluated at compile time into a true list instead of being\nreplaced with a run time call to split(). This removes the confusing misbehaviour of \"qw//\"\nin scalar context, which had inherited that behaviour from split().\n\nThus:\n\n$foo = ($bar) = qw(a b c); print \"$foo|$bar\\n\";\n\nnow correctly prints \"3|a\", instead of \"2|a\".\n" }, { "name": "Better worst-case behavior of hashes", "content": "Small changes in the hashing algorithm have been implemented in order to improve the\ndistribution of lower order bits in the hashed value. This is expected to yield better\nperformance on keys that are repeated sequences.\n" }, { "name": "pack() format 'Z' supported", "content": "The new format type 'Z' is useful for packing and unpacking null-terminated strings. See\n\"pack\" in perlfunc.\n" }, { "name": "pack() format modifier '!' supported", "content": "The new format type modifier '!' is useful for packing and unpacking native shorts, ints, and\nlongs. See \"pack\" in perlfunc.\n" }, { "name": "pack() and unpack() support counted strings", "content": "The template character '/' can be used to specify a counted string type to be packed or\nunpacked. See \"pack\" in perlfunc.\n" }, { "name": "Comments in pack() templates", "content": "The '#' character in a template introduces a comment up to end of the line. This facilitates\ndocumentation of pack() templates.\n" }, { "name": "Weak references", "content": "In previous versions of Perl, you couldn't cache objects so as to allow them to be deleted if\nthe last reference from outside the cache is deleted. The reference in the cache would hold\na reference count on the object and the objects would never be destroyed.\n\nAnother familiar problem is with circular references. When an object references itself, its\nreference count would never go down to zero, and it would not get destroyed until the program\nis about to exit.\n\nWeak references solve this by allowing you to \"weaken\" any reference, that is, make it not\ncount towards the reference count. When the last non-weak reference to an object is deleted,\nthe object is destroyed and all the weak references to the object are automatically undef-ed.\n\nTo use this feature, you need the Devel::WeakRef package from CPAN, which contains additional\ndocumentation.\n\nNOTE: This is an experimental feature. Details are subject to change.\n" }, { "name": "Binary numbers supported", "content": "Binary numbers are now supported as literals, in s?printf formats, and oct():\n\n$answer = 0b101010;\nprintf \"The answer is: %b\\n\", oct(\"0b101010\");\n" }, { "name": "Lvalue subroutines", "content": "Subroutines can now return modifiable lvalues. See \"Lvalue subroutines\" in perlsub.\n\nNOTE: This is an experimental feature. Details are subject to change.\n" }, { "name": "Some arrows may be omitted in calls through references", "content": "Perl now allows the arrow to be omitted in many constructs involving subroutine calls through\nreferences. For example, \"$foo[10]->('foo')\" may now be written \"$foo[10]('foo')\". This is\nrather similar to how the arrow may be omitted from \"$foo[10]->{'foo'}\". Note however, that\nthe arrow is still required for \"foo(10)->('bar')\".\n" }, { "name": "Boolean assignment operators are legal lvalues", "content": "Constructs such as \"($a ||= 2) += 1\" are now allowed.\n" }, { "name": "exists() is supported on subroutine names", "content": "The exists() builtin now works on subroutine names. A subroutine is considered to exist if\nit has been declared (even if implicitly). See \"exists\" in perlfunc for examples.\n" }, { "name": "exists() and delete() are supported on array elements", "content": "The exists() and delete() builtins now work on simple arrays as well. The behavior is\nsimilar to that on hash elements.\n\nexists() can be used to check whether an array element has been initialized. This avoids\nautovivifying array elements that don't exist. If the array is tied, the EXISTS() method in\nthe corresponding tied package will be invoked.\n\ndelete() may be used to remove an element from the array and return it. The array element at\nthat position returns to its uninitialized state, so that testing for the same element with\nexists() will return false. If the element happens to be the one at the end, the size of the\narray also shrinks up to the highest element that tests true for exists(), or 0 if none such\nis found. If the array is tied, the DELETE() method in the corresponding tied package will\nbe invoked.\n\nSee \"exists\" in perlfunc and \"delete\" in perlfunc for examples.\n" }, { "name": "Pseudo-hashes work better", "content": "Dereferencing some types of reference values in a pseudo-hash, such as \"$ph->{foo}[1]\", was\naccidentally disallowed. This has been corrected.\n\nWhen applied to a pseudo-hash element, exists() now reports whether the specified value\nexists, not merely if the key is valid.\n\ndelete() now works on pseudo-hashes. When given a pseudo-hash element or slice it deletes\nthe values corresponding to the keys (but not the keys themselves). See \"Pseudo-hashes:\nUsing an array as a hash\" in perlref.\n\nPseudo-hash slices with constant keys are now optimized to array lookups at compile-time.\n\nList assignments to pseudo-hash slices are now supported.\n\nThe \"fields\" pragma now provides ways to create pseudo-hashes, via fields::new() and\nfields::phash(). See fields.\n\nNOTE: The pseudo-hash data type continues to be experimental.\nLimiting oneself to the interface elements provided by the\nfields pragma will provide protection from any future changes.\n" }, { "name": "Automatic flushing of output buffers", "content": "fork(), exec(), system(), qx//, and pipe open()s now flush buffers of all files opened for\noutput when the operation was attempted. This mostly eliminates confusing buffering mishaps\nsuffered by users unaware of how Perl internally handles I/O.\n\nThis is not supported on some platforms like Solaris where a suitably correct implementation\nof fflush(NULL) isn't available.\n" }, { "name": "Better diagnostics on meaningless filehandle operations", "content": "Constructs such as open() and close() are compile time errors. Attempting to read\nfrom filehandles that were opened only for writing will now produce warnings (just as writing\nto read-only filehandles does).\n" }, { "name": "Where possible, buffered data discarded from duped input filehandle", "content": "\"open(NEW, \"<&OLD\")\" now attempts to discard any data that was previously read and buffered\nin \"OLD\" before duping the handle. On platforms where doing this is allowed, the next read\noperation on \"NEW\" will return the same data as the corresponding operation on \"OLD\".\nFormerly, it would have returned the data from the start of the following disk block instead.\n" }, { "name": "eof() has the same old magic as <>", "content": "eof() would return true if no attempt to read from \"<>\" had yet been made. eof() has been\nchanged to have a little magic of its own, it now opens the \"<>\" files.\n" }, { "name": "binmode() can be used to set :crlf and :raw modes", "content": "binmode() now accepts a second argument that specifies a discipline for the handle in\nquestion. The two pseudo-disciplines \":raw\" and \":crlf\" are currently supported on DOS-\nderivative platforms. See \"binmode\" in perlfunc and open.\n" }, { "name": "\"-T\" filetest recognizes UTF-8 encoded files as \"text\"", "content": "The algorithm used for the \"-T\" filetest has been enhanced to correctly identify UTF-8\ncontent as \"text\".\n" }, { "name": "system(), backticks and pipe open now reflect exec() failure", "content": "On Unix and similar platforms, system(), qx() and open(FOO, \"cmd |\") etc., are implemented\nvia fork() and exec(). When the underlying exec() fails, earlier versions did not report the\nerror properly, since the exec() happened to be in a different process.\n\nThe child process now communicates with the parent about the error in launching the external\ncommand, which allows these constructs to return with their usual error value and set $!.\n" }, { "name": "Improved diagnostics", "content": "Line numbers are no longer suppressed (under most likely circumstances) during the global\ndestruction phase.\n\nDiagnostics emitted from code running in threads other than the main thread are now\naccompanied by the thread ID.\n\nEmbedded null characters in diagnostics now actually show up. They used to truncate the\nmessage in prior versions.\n\n$foo::a and $foo::b are now exempt from \"possible typo\" warnings only if sort() is\nencountered in package \"foo\".\n\nUnrecognized alphabetic escapes encountered when parsing quote constructs now generate a\nwarning, since they may take on new semantics in later versions of Perl.\n\nMany diagnostics now report the internal operation in which the warning was provoked, like\nso:\n\nUse of uninitialized value in concatenation (.) at (eval 1) line 1.\nUse of uninitialized value in print at (eval 1) line 1.\n\nDiagnostics that occur within eval may also report the file and line number where the eval\nis located, in addition to the eval sequence number and the line number within the evaluated\ntext itself. For example:\n\nNot enough arguments for scalar at (eval 4)[newlib/perl5db.pl:1411] line 2, at EOF\n" }, { "name": "Diagnostics follow STDERR", "content": "Diagnostic output now goes to whichever file the \"STDERR\" handle is pointing at, instead of\nalways going to the underlying C runtime library's \"stderr\".\n" }, { "name": "More consistent close-on-exec behavior", "content": "On systems that support a close-on-exec flag on filehandles, the flag is now set for any\nhandles created by pipe(), socketpair(), socket(), and accept(), if that is warranted by the\nvalue of $^F that may be in effect. Earlier versions neglected to set the flag for handles\ncreated with these operators. See \"pipe\" in perlfunc, \"socketpair\" in perlfunc, \"socket\" in\nperlfunc, \"accept\" in perlfunc, and \"$^F\" in perlvar.\n" }, { "name": "syswrite() ease-of-use", "content": "The length argument of syswrite() has become optional.\n" }, { "name": "Better syntax checks on parenthesized unary operators", "content": "Expressions such as:\n\nprint defined(&foo,&bar,&baz);\nprint uc(\"foo\",\"bar\",\"baz\");\nundef($foo,&bar);\n\nused to be accidentally allowed in earlier versions, and produced unpredictable behaviour.\nSome produced ancillary warnings when used in this way; others silently did the wrong thing.\n\nThe parenthesized forms of most unary operators that expect a single argument now ensure that\nthey are not called with more than one argument, making the cases shown above syntax errors.\nThe usual behaviour of:\n\nprint defined &foo, &bar, &baz;\nprint uc \"foo\", \"bar\", \"baz\";\nundef $foo, &bar;\n\nremains unchanged. See perlop.\n" }, { "name": "Bit operators support full native integer width", "content": "The bit operators (& | ^ ~ << >>) now operate on the full native integral width (the exact\nsize of which is available in $Config{ivsize}). For example, if your platform is either\nnatively 64-bit or if Perl has been configured to use 64-bit integers, these operations apply\nto 8 bytes (as opposed to 4 bytes on 32-bit platforms). For portability, be sure to mask off\nthe excess bits in the result of unary \"~\", e.g., \"~$x & 0xffffffff\".\n" }, { "name": "Improved security features", "content": "More potentially unsafe operations taint their results for improved security.\n\nThe \"passwd\" and \"shell\" fields returned by the getpwent(), getpwnam(), and getpwuid() are\nnow tainted, because the user can affect their own encrypted password and login shell.\n\nThe variable modified by shmread(), and messages returned by msgrcv() (and its object-\noriented interface IPC::SysV::Msg::rcv) are also tainted, because other untrusted processes\ncan modify messages and shared memory segments for their own nefarious purposes.\n\nMore functional bareword prototype (*)\nBareword prototypes have been rationalized to enable them to be used to override builtins\nthat accept barewords and interpret them in a special way, such as \"require\" or \"do\".\n\nArguments prototyped as \"*\" will now be visible within the subroutine as either a simple\nscalar or as a reference to a typeglob. See \"Prototypes\" in perlsub.\n" }, { "name": "\"require\" and \"do\" may be overridden", "content": "\"require\" and \"do 'file'\" operations may be overridden locally by importing subroutines of\nthe same name into the current package (or globally by importing them into the CORE::GLOBAL::\nnamespace). Overriding \"require\" will also affect \"use\", provided the override is visible at\ncompile-time. See \"Overriding Built-in Functions\" in perlsub.\n" }, { "name": "$^X variables may now have names longer than one character", "content": "Formerly, $^X was synonymous with ${\"\\cX\"}, but $^XY was a syntax error. Now variable names\nthat begin with a control character may be arbitrarily long. However, for compatibility\nreasons, these variables must be written with explicit braces, as \"${^XY}\" for example.\n\"${^XYZ}\" is synonymous with ${\"\\cXYZ\"}. Variable names with more than one control\ncharacter, such as \"${^XY^Z}\", are illegal.\n\nThe old syntax has not changed. As before, `^X' may be either a literal control-X character\nor the two-character sequence `caret' plus `X'. When braces are omitted, the variable name\nstops after the control character. Thus \"$^XYZ\" continues to be synonymous with \"$^X . \"YZ\"\"\nas before.\n\nAs before, lexical variables may not have names beginning with control characters. As\nbefore, variables whose names begin with a control character are always forced to be in\npackage `main'. All such variables are reserved for future extensions, except those that\nbegin with \"^\", which may be used by user programs and are guaranteed not to acquire special\nmeaning in any future version of Perl.\n" }, { "name": "New variable $^C reflects \"-c\" switch", "content": "$^C has a boolean value that reflects whether perl is being run in compile-only mode (i.e.\nvia the \"-c\" switch). Since BEGIN blocks are executed under such conditions, this variable\nenables perl code to determine whether actions that make sense only during normal running are\nwarranted. See perlvar.\n" }, { "name": "New variable $^V contains Perl version as a string", "content": "$^V contains the Perl version number as a string composed of characters whose ordinals match\nthe version numbers, i.e. v5.6.0. This may be used in string comparisons.\n\nSee \"Support for strings represented as a vector of ordinals\" for an example.\n" }, { "name": "Optional Y2K warnings", "content": "If Perl is built with the cpp macro \"PERLY2KWARN\" defined, it emits optional warnings when\nconcatenating the number 19 with another number.\n\nThis behavior must be specifically enabled when running Configure. See INSTALL and\nREADME.Y2K.\n" }, { "name": "Arrays now always interpolate into double-quoted strings", "content": "In double-quoted strings, arrays now interpolate, no matter what. The behavior in earlier\nversions of perl 5 was that arrays would interpolate into strings if the array had been\nmentioned before the string was compiled, and otherwise Perl would raise a fatal compile-time\nerror. In versions 5.000 through 5.003, the error was\n\nLiteral @example now requires backslash\n\nIn versions 5.00401 through 5.6.0, the error was\n\nIn string, @example now must be written as \\@example\n\nThe idea here was to get people into the habit of writing \"fred\\@example.com\" when they\nwanted a literal \"@\" sign, just as they have always written \"Give me back my \\$5\" when they\nwanted a literal \"$\" sign.\n\nStarting with 5.6.1, when Perl now sees an \"@\" sign in a double-quoted string, it always\nattempts to interpolate an array, regardless of whether or not the array has been used or\ndeclared already. The fatal error has been downgraded to an optional warning:\n\nPossible unintended interpolation of @example in string\n\nThis warns you that \"fred@example.com\" is going to turn into \"fred.com\" if you don't\nbackslash the \"@\". See http://perl.plover.com/at-error.html for more details about the\nhistory here.\n" }, { "name": "@- and @+ provide starting/ending offsets of regex submatches", "content": "The new magic variables @- and @+ provide the starting and ending offsets, respectively, of\n$&, $1, $2, etc. See perlvar for details.\n" } ] }, "Modules and Pragmata": { "content": "", "subsections": [ { "name": "Modules", "content": "attributes\nWhile used internally by Perl as a pragma, this module also provides a way to fetch\nsubroutine and variable attributes. See attributes.\n\nB The Perl Compiler suite has been extensively reworked for this release. More of the\nstandard Perl test suite passes when run under the Compiler, but there is still a\nsignificant way to go to achieve production quality compiled executables.\n\nNOTE: The Compiler suite remains highly experimental. The\ngenerated code may not be correct, even when it manages to execute\nwithout errors.\n\nBenchmark\nOverall, Benchmark results exhibit lower average error and better timing accuracy.\n\nYou can now run tests for n seconds instead of guessing the right number of tests to run:\ne.g., timethese(-5, ...) will run each code for at least 5 CPU seconds. Zero as the\n\"number of repetitions\" means \"for at least 3 CPU seconds\". The output format has also\nchanged. For example:\n\nuse Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x2}})\n\nwill now output something like this:\n\nBenchmark: running a, b, each for at least 5 CPU seconds...\na: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)\nb: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)\n\nNew features: \"each for at least N CPU seconds...\", \"wallclock secs\", and the \"@\noperations/CPU second (n=operations)\".\n\ntimethese() now returns a reference to a hash of Benchmark objects containing the test\nresults, keyed on the names of the tests.\n\ntimethis() now returns the iterations field in the Benchmark result object instead of 0.\n\ntimethese(), timethis(), and the new cmpthese() (see below) can also take a format\nspecifier of 'none' to suppress output.\n\nA new function countit() is just like timeit() except that it takes a TIME instead of a\nCOUNT.\n\nA new function cmpthese() prints a chart comparing the results of each test returned from\na timethese() call. For each possible pair of tests, the percentage speed difference\n(iters/sec or seconds/iter) is shown.\n\nFor other details, see Benchmark.\n\nByteLoader\nThe ByteLoader is a dedicated extension to generate and run Perl bytecode. See\nByteLoader.\n\nconstant\nReferences can now be used.\n\nThe new version also allows a leading underscore in constant names, but disallows a\ndouble leading underscore (as in \"LINE\"). Some other names are disallowed or warned\nagainst, including BEGIN, END, etc. Some names which were forced into main:: used to\nfail silently in some cases; now they're fatal (outside of main::) and an optional\nwarning (inside of main::). The ability to detect whether a constant had been set with a\ngiven name has been added.\n\nSee constant.\n\ncharnames\nThis pragma implements the \"\\N\" string escape. See charnames.\n\nData::Dumper\nA \"Maxdepth\" setting can be specified to avoid venturing too deeply into deep data\nstructures. See Data::Dumper.\n\nThe XSUB implementation of Dump() is now automatically called if the \"Useqq\" setting is\nnot in use.\n\nDumping \"qr//\" objects works correctly.\n\nDB \"DB\" is an experimental module that exposes a clean abstraction to Perl's debugging API.\n\nDBFile\nDBFile can now be built with Berkeley DB versions 1, 2 or 3. See \"ext/DBFile/Changes\".\n\nDevel::DProf\nDevel::DProf, a Perl source code profiler has been added. See Devel::DProf and dprofpp.\n\nDevel::Peek\nThe Devel::Peek module provides access to the internal representation of Perl variables\nand data. It is a data debugging tool for the XS programmer.\n\nDumpvalue\nThe Dumpvalue module provides screen dumps of Perl data.\n\nDynaLoader\nDynaLoader now supports a dlunloadfile() function on platforms that support unloading\nshared objects using dlclose().\n\nPerl can also optionally arrange to unload all extension shared objects loaded by Perl.\nTo enable this, build Perl with the Configure option \"-Accflags=-DDLUNLOADALLATEXIT\".\n(This maybe useful if you are using Apache with modperl.)\n\nEnglish\n$PERLVERSION now stands for $^V (a string value) rather than for $] (a numeric value).\n\nEnv Env now supports accessing environment variables like PATH as array variables.\n\nFcntl\nMore Fcntl constants added: FSETLK64, FSETLKW64, OLARGEFILE for large file (more than\n4GB) access (NOTE: the OLARGEFILE is automatically added to sysopen() flags if large\nfile support has been configured, as is the default), Free/Net/OpenBSD locking behaviour\nflags FFLOCK, FPOSIX, Linux FSHLCK, and OACCMODE: the combined mask of ORDONLY,\nOWRONLY, and ORDWR. The seek()/sysseek() constants SEEKSET, SEEKCUR, and SEEKEND\nare available via the \":seek\" tag. The chmod()/stat() SIF* constants and SIS*\nfunctions are available via the \":mode\" tag.\n\nFile::Compare\nA comparetext() function has been added, which allows custom comparison functions. See\nFile::Compare.\n\nFile::Find\nFile::Find now works correctly when the wanted() function is either autoloaded or is a\nsymbolic reference.\n\nA bug that caused File::Find to lose track of the working directory when pruning top-\nlevel directories has been fixed.\n\nFile::Find now also supports several other options to control its behavior. It can\nfollow symbolic links if the \"follow\" option is specified. Enabling the \"nochdir\"\noption will make File::Find skip changing the current directory when walking directories.\nThe \"untaint\" flag can be useful when running with taint checks enabled.\n\nSee File::Find.\n\nFile::Glob\nThis extension implements BSD-style file globbing. By default, it will also be used for\nthe internal implementation of the glob() operator. See File::Glob.\n\nFile::Spec\nNew methods have been added to the File::Spec module: devnull() returns the name of the\nnull device (/dev/null on Unix) and tmpdir() the name of the temp directory (normally\n/tmp on Unix). There are now also methods to convert between absolute and relative\nfilenames: abs2rel() and rel2abs(). For compatibility with operating systems that\nspecify volume names in file paths, the splitpath(), splitdir(), and catdir() methods\nhave been added.\n\nFile::Spec::Functions\nThe new File::Spec::Functions modules provides a function interface to the File::Spec\nmodule. Allows shorthand\n\n$fullname = catfile($dir1, $dir2, $file);\n\ninstead of\n\n$fullname = File::Spec->catfile($dir1, $dir2, $file);\n\nGetopt::Long\nGetopt::Long licensing has changed to allow the Perl Artistic License as well as the GPL.\nIt used to be GPL only, which got in the way of non-GPL applications that wanted to use\nGetopt::Long.\n\nGetopt::Long encourages the use of Pod::Usage to produce help messages. For example:\n\nuse Getopt::Long;\nuse Pod::Usage;\nmy $man = 0;\nmy $help = 0;\nGetOptions('help|?' => \\$help, man => \\$man) or pod2usage(2);\npod2usage(1) if $help;\npod2usage(-exitstatus => 0, -verbose => 2) if $man;\n\nEND\n\n=head1 NAME\n\nsample - Using Getopt::Long and Pod::Usage\n\n=head1 SYNOPSIS\n\nsample [options] [file ...]\n\nOptions:\n-help brief help message\n-man full documentation\n\n=head1 OPTIONS\n\n=over 8\n\n=item B<-help>\n\nPrint a brief help message and exits.\n\n=item B<-man>\n\nPrints the manual page and exits.\n\n=back\n\n=head1 DESCRIPTION\n\nB will read the given input file(s) and do something\nuseful with the contents thereof.\n\n=cut\n\nSee Pod::Usage for details.\n\nA bug that prevented the non-option call-back <> from being specified as the first\nargument has been fixed.\n\nTo specify the characters < and > as option starters, use ><. Note, however, that\nchanging option starters is strongly deprecated.\n\nIO write() and syswrite() will now accept a single-argument form of the call, for\nconsistency with Perl's syswrite().\n\nYou can now create a TCP-based IO::Socket::INET without forcing a connect attempt. This\nallows you to configure its options (like making it non-blocking) and then call connect()\nmanually.\n\nA bug that prevented the IO::Socket::protocol() accessor from ever returning the correct\nvalue has been corrected.\n\nIO::Socket::connect now uses non-blocking IO instead of alarm() to do connect timeouts.\n\nIO::Socket::accept now uses select() instead of alarm() for doing timeouts.\n\nIO::Socket::INET->new now sets $! correctly on failure. $@ is still set for backwards\ncompatibility.\n\nJPL Java Perl Lingo is now distributed with Perl. See jpl/README for more information.\n\nlib \"use lib\" now weeds out any trailing duplicate entries. \"no lib\" removes all named\nentries.\n\nMath::BigInt\nThe bitwise operations \"<<\", \">>\", \"&\", \"|\", and \"~\" are now supported on bigints.\n\nMath::Complex\nThe accessor methods Re, Im, arg, abs, rho, and theta can now also act as mutators\n(accessor $z->Re(), mutator $z->Re(3)).\n\nThe class method \"displayformat\" and the corresponding object method \"displayformat\",\nin addition to accepting just one argument, now can also accept a parameter hash.\nRecognized keys of a parameter hash are \"style\", which corresponds to the old one\nparameter case, and two new parameters: \"format\", which is a printf()-style format string\n(defaults usually to \"%.15g\", you can revert to the default by setting the format string\nto \"undef\") used for both parts of a complex number, and \"polarprettyprint\" (defaults\nto true), which controls whether an attempt is made to try to recognize small multiples\nand rationals of pi (2pi, pi/2) at the argument (angle) of a polar complex number.\n\nThe potentially disruptive change is that in list context both methods now return the\nparameter hash, instead of only the value of the \"style\" parameter.\n\nMath::Trig\nA little bit of radial trigonometry (cylindrical and spherical), radial coordinate\nconversions, and the great circle distance were added.\n\nPod::Parser, Pod::InputObjects\nPod::Parser is a base class for parsing and selecting sections of pod documentation from\nan input stream. This module takes care of identifying pod paragraphs and commands in\nthe input and hands off the parsed paragraphs and commands to user-defined methods which\nare free to interpret or translate them as they see fit.\n\nPod::InputObjects defines some input objects needed by Pod::Parser, and for advanced\nusers of Pod::Parser that need more about a command besides its name and text.\n\nAs of release 5.6.0 of Perl, Pod::Parser is now the officially sanctioned \"base parser\ncode\" recommended for use by all pod2xxx translators. Pod::Text (pod2text) and Pod::Man\n(pod2man) have already been converted to use Pod::Parser and efforts to convert Pod::HTML\n(pod2html) are already underway. For any questions or comments about pod parsing and\ntranslating issues and utilities, please use the pod-people@perl.org mailing list.\n\nFor further information, please see Pod::Parser and Pod::InputObjects.\n\nPod::Checker, podchecker\nThis utility checks pod files for correct syntax, according to perlpod. Obvious errors\nare flagged as such, while warnings are printed for mistakes that can be handled\ngracefully. The checklist is not complete yet. See Pod::Checker.\n\nPod::ParseUtils, Pod::Find\nThese modules provide a set of gizmos that are useful mainly for pod translators.\nPod::Find traverses directory structures and returns found pod files, along with their\ncanonical names (like \"File::Spec::Unix\"). Pod::ParseUtils contains Pod::List (useful\nfor storing pod list information), Pod::Hyperlink (for parsing the contents of \"L<>\"\nsequences) and Pod::Cache (for caching information about pod files, e.g., link nodes).\n\nPod::Select, podselect\nPod::Select is a subclass of Pod::Parser which provides a function named \"podselect()\" to\nfilter out user-specified sections of raw pod documentation from an input stream.\npodselect is a script that provides access to Pod::Select from other scripts to be used\nas a filter. See Pod::Select.\n\nPod::Usage, pod2usage\nPod::Usage provides the function \"pod2usage()\" to print usage messages for a Perl script\nbased on its embedded pod documentation. The pod2usage() function is generally useful to\nall script authors since it lets them write and maintain a single source (the pods) for\ndocumentation, thus removing the need to create and maintain redundant usage message text\nconsisting of information already in the pods.\n\nThere is also a pod2usage script which can be used from other kinds of scripts to print\nusage messages from pods (even for non-Perl scripts with pods embedded in comments).\n\nFor details and examples, please see Pod::Usage.\n\nPod::Text and Pod::Man\nPod::Text has been rewritten to use Pod::Parser. While pod2text() is still available for\nbackwards compatibility, the module now has a new preferred interface. See Pod::Text for\nthe details. The new Pod::Text module is easily subclassed for tweaks to the output, and\ntwo such subclasses (Pod::Text::Termcap for man-page-style bold and underlining using\ntermcap information, and Pod::Text::Color for markup with ANSI color sequences) are now\nstandard.\n\npod2man has been turned into a module, Pod::Man, which also uses Pod::Parser. In the\nprocess, several outstanding bugs related to quotes in section headers, quoting of code\nescapes, and nested lists have been fixed. pod2man is now a wrapper script around this\nmodule.\n\nSDBMFile\nAn EXISTS method has been added to this module (and sdbmexists() has been added to the\nunderlying sdbm library), so one can now call exists on an SDBMFile tied hash and get\nthe correct result, rather than a runtime error.\n\nA bug that may have caused data loss when more than one disk block happens to be read\nfrom the database in a single FETCH() has been fixed.\n\nSys::Syslog\nSys::Syslog now uses XSUBs to access facilities from syslog.h so it no longer requires\nsyslog.ph to exist.\n\nSys::Hostname\nSys::Hostname now uses XSUBs to call the C library's gethostname() or uname() if they\nexist.\n\nTerm::ANSIColor\nTerm::ANSIColor is a very simple module to provide easy and readable access to the ANSI\ncolor and highlighting escape sequences, supported by most ANSI terminal emulators. It\nis now included standard.\n\nTime::Local\nThe timelocal() and timegm() functions used to silently return bogus results when the\ndate fell outside the machine's integer range. They now consistently croak() if the date\nfalls in an unsupported range.\n\nWin32\nThe error return value in list context has been changed for all functions that return a\nlist of values. Previously these functions returned a list with a single element \"undef\"\nif an error occurred. Now these functions return the empty list in these situations.\nThis applies to the following functions:\n\nWin32::FsType\nWin32::GetOSVersion\n\nThe remaining functions are unchanged and continue to return \"undef\" on error even in\nlist context.\n\nThe Win32::SetLastError(ERROR) function has been added as a complement to the\nWin32::GetLastError() function.\n\nThe new Win32::GetFullPathName(FILENAME) returns the full absolute pathname for FILENAME\nin scalar context. In list context it returns a two-element list containing the fully\nqualified directory name and the filename. See Win32.\n\nXSLoader\nThe XSLoader extension is a simpler alternative to DynaLoader. See XSLoader.\n\nDBM Filters\nA new feature called \"DBM Filters\" has been added to all the DBM modules--DBFile,\nGDBMFile, NDBMFile, ODBMFile, and SDBMFile. DBM Filters add four new methods to each\nDBM module:\n\nfilterstorekey\nfilterstorevalue\nfilterfetchkey\nfilterfetchvalue\n\nThese can be used to filter key-value pairs before the pairs are written to the database\nor just after they are read from the database. See perldbmfilter for further\ninformation.\n" }, { "name": "Pragmata", "content": "\"use attrs\" is now obsolete, and is only provided for backward-compatibility. It's been\nreplaced by the \"sub : attributes\" syntax. See \"Subroutine Attributes\" in perlsub and\nattributes.\n\nLexical warnings pragma, \"use warnings;\", to control optional warnings. See perllexwarn.\n\n\"use filetest\" to control the behaviour of filetests (\"-r\" \"-w\" ...). Currently only one\nsubpragma implemented, \"use filetest 'access';\", that uses access(2) or equivalent to check\npermissions instead of using stat(2) as usual. This matters in filesystems where there are\nACLs (access control lists): the stat(2) might lie, but access(2) knows better.\n\nThe \"open\" pragma can be used to specify default disciplines for handle constructors (e.g.\nopen()) and for qx//. The two pseudo-disciplines \":raw\" and \":crlf\" are currently supported\non DOS-derivative platforms (i.e. where binmode is not a no-op). See also \"binmode() can be\nused to set :crlf and :raw modes\".\n" } ] }, "Utility Changes": { "content": "", "subsections": [ { "name": "dprofpp", "content": "\"dprofpp\" is used to display profile data generated using \"Devel::DProf\". See dprofpp.\n" }, { "name": "find2perl", "content": "The \"find2perl\" utility now uses the enhanced features of the File::Find module. The -depth\nand -follow options are supported. Pod documentation is also included in the script.\n" }, { "name": "h2xs", "content": "The \"h2xs\" tool can now work in conjunction with \"C::Scan\" (available from CPAN) to\nautomatically parse real-life header files. The \"-M\", \"-a\", \"-k\", and \"-o\" options are new.\n" }, { "name": "perlcc", "content": "\"perlcc\" now supports the C and Bytecode backends. By default, it generates output from the\nsimple C backend rather than the optimized C backend.\n\nSupport for non-Unix platforms has been improved.\n" }, { "name": "perldoc", "content": "\"perldoc\" has been reworked to avoid possible security holes. It will not by default let\nitself be run as the superuser, but you may still use the -U switch to try to make it drop\nprivileges first.\n" }, { "name": "The Perl Debugger", "content": "Many bug fixes and enhancements were added to perl5db.pl, the Perl debugger. The help\ndocumentation was rearranged. New commands include \"< ?\", \"> ?\", and \"{ ?\" to list out\ncurrent actions, \"man docpage\" to run your doc viewer on some perl docset, and support for\nquoted options. The help information was rearranged, and should be viewable once again if\nyou're using less as your pager. A serious security hole was plugged--you should immediately\nremove all older versions of the Perl debugger as installed in previous releases, all the way\nback to perl3, from your system to avoid being bitten by this.\n" } ] }, "Improved Documentation": { "content": "Many of the platform-specific README files are now part of the perl installation. See perl\nfor the complete list.\n\nperlapi.pod\nThe official list of public Perl API functions.\n\nperlboot.pod\nA tutorial for beginners on object-oriented Perl.\n\nperlcompile.pod\nAn introduction to using the Perl Compiler suite.\n\nperldbmfilter.pod\nA howto document on using the DBM filter facility.\n\nperldebug.pod\nAll material unrelated to running the Perl debugger, plus all low-level guts-like details\nthat risked crushing the casual user of the debugger, have been relocated from the old\nmanpage to the next entry below.\n\nperldebguts.pod\nThis new manpage contains excessively low-level material not related to the Perl\ndebugger, but slightly related to debugging Perl itself. It also contains some arcane\ninternal details of how the debugging process works that may only be of interest to\ndevelopers of Perl debuggers.\n\nperlfork.pod\nNotes on the fork() emulation currently available for the Windows platform.\n\nperlfilter.pod\nAn introduction to writing Perl source filters.\n\nperlhack.pod\nSome guidelines for hacking the Perl source code.\n\nperlintern.pod\nA list of internal functions in the Perl source code. (List is currently empty.)\n\nperllexwarn.pod\nIntroduction and reference information about lexically scoped warning categories.\n\nperlnumber.pod\nDetailed information about numbers as they are represented in Perl.\n\nperlopentut.pod\nA tutorial on using open() effectively.\n\nperlreftut.pod\nA tutorial that introduces the essentials of references.\n\nperltootc.pod\nA tutorial on managing class data for object modules.\n\nperltodo.pod\nDiscussion of the most often wanted features that may someday be supported in Perl.\n\nperlunicode.pod\nAn introduction to Unicode support features in Perl.\n", "subsections": [] }, "Performance enhancements": { "content": "", "subsections": [ { "name": "Simple sort() using { $a <=> $b } and the like are optimized", "content": "Many common sort() operations using a simple inlined block are now optimized for faster\nperformance.\n" }, { "name": "Optimized assignments to lexical variables", "content": "Certain operations in the RHS of assignment statements have been optimized to directly set\nthe lexical variable on the LHS, eliminating redundant copying overheads.\n" }, { "name": "Faster subroutine calls", "content": "Minor changes in how subroutine calls are handled internally provide marginal improvements in\nperformance.\n" }, { "name": "delete(), each(), values() and hash iteration are faster", "content": "The hash values returned by delete(), each(), values() and hashes in a list context are the\nactual values in the hash, instead of copies. This results in significantly better\nperformance, because it eliminates needless copying in most situations.\n" } ] }, "Installation and Configuration Improvements": { "content": "", "subsections": [ { "name": "-Dusethreads means something different", "content": "The -Dusethreads flag now enables the experimental interpreter-based thread support by\ndefault. To get the flavor of experimental threads that was in 5.005 instead, you need to\nrun Configure with \"-Dusethreads -Duse5005threads\".\n\nAs of v5.6.0, interpreter-threads support is still lacking a way to create new threads from\nPerl (i.e., \"use Thread;\" will not work with interpreter threads). \"use Thread;\" continues\nto be available when you specify the -Duse5005threads option to Configure, bugs and all.\n\nNOTE: Support for threads continues to be an experimental feature.\nInterfaces and implementation are subject to sudden and drastic changes.\n" }, { "name": "New Configure flags", "content": "The following new flags may be enabled on the Configure command line by running Configure\nwith \"-Dflag\".\n\nusemultiplicity\nusethreads useithreads (new interpreter threads: no Perl API yet)\nusethreads use5005threads (threads as they were in 5.005)\n\nuse64bitint (equal to now deprecated 'use64bits')\nuse64bitall\n\nuselongdouble\nusemorebits\nuselargefiles\nusesocks (only SOCKS v5 supported)\n" }, { "name": "Threadedness and 64-bitness now more daring", "content": "The Configure options enabling the use of threads and the use of 64-bitness are now more\ndaring in the sense that they no more have an explicit list of operating systems of known\nthreads/64-bit capabilities. In other words: if your operating system has the necessary APIs\nand datatypes, you should be able just to go ahead and use them, for threads by Configure\n-Dusethreads, and for 64 bits either explicitly by Configure -Duse64bitint or implicitly if\nyour system has 64-bit wide datatypes. See also \"64-bit support\".\n" }, { "name": "Long Doubles", "content": "Some platforms have \"long doubles\", floating point numbers of even larger range than ordinary\n\"doubles\". To enable using long doubles for Perl's scalars, use -Duselongdouble.\n" }, { "name": "-Dusemorebits", "content": "You can enable both -Duse64bitint and -Duselongdouble with -Dusemorebits. See also \"64-bit\nsupport\".\n" }, { "name": "-Duselargefiles", "content": "Some platforms support system APIs that are capable of handling large files (typically, files\nlarger than two gigabytes). Perl will try to use these APIs if you ask for -Duselargefiles.\n\nSee \"Large file support\" for more information.\n" }, { "name": "installusrbinperl", "content": "You can use \"Configure -Uinstallusrbinperl\" which causes installperl to skip installing perl\nalso as /usr/bin/perl. This is useful if you prefer not to modify /usr/bin for some reason\nor another but harmful because many scripts assume to find Perl in /usr/bin/perl.\n" }, { "name": "SOCKS support", "content": "You can use \"Configure -Dusesocks\" which causes Perl to probe for the SOCKS proxy protocol\nlibrary (v5, not v4). For more information on SOCKS, see:\n\nhttp://www.socks.nec.com/\n" }, { "name": "\"-A\" flag", "content": "You can \"post-edit\" the Configure variables using the Configure \"-A\" switch. The editing\nhappens immediately after the platform specific hints files have been processed but before\nthe actual configuration process starts. Run \"Configure -h\" to find out the full \"-A\"\nsyntax.\n" }, { "name": "Enhanced Installation Directories", "content": "The installation structure has been enriched to improve the support for maintaining multiple\nversions of perl, to provide locations for vendor-supplied modules, scripts, and manpages,\nand to ease maintenance of locally-added modules, scripts, and manpages. See the section on\nInstallation Directories in the INSTALL file for complete details. For most users building\nand installing from source, the defaults should be fine.\n\nIf you previously used \"Configure -Dsitelib\" or \"-Dsitearch\" to set special values for\nlibrary directories, you might wish to consider using the new \"-Dsiteprefix\" setting instead.\nAlso, if you wish to re-use a config.sh file from an earlier version of perl, you should be\nsure to check that Configure makes sensible choices for the new directories. See INSTALL for\ncomplete details.\n" }, { "name": "gcc automatically tried if 'cc' does not seem to be working", "content": "In many platforms the vendor-supplied 'cc' is too stripped-down to build Perl (basically, the\n'cc' doesn't do ANSI C). If this seems to be the case and the 'cc' does not seem to be the\nGNU C compiler 'gcc', an automatic attempt is made to find and use 'gcc' instead.\n" } ] }, "Platform specific changes": { "content": "", "subsections": [ { "name": "Supported platforms", "content": "• The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread extension.\n\n• GNU/Hurd is now supported.\n\n• Rhapsody/Darwin is now supported.\n\n• EPOC is now supported (on Psion 5).\n\n• The cygwin port (formerly cygwin32) has been greatly improved.\n\nDOS\n• Perl now works with djgpp 2.02 (and 2.03 alpha).\n\n• Environment variable names are not converted to uppercase any more.\n\n• Incorrect exit codes from backticks have been fixed.\n\n• This port continues to use its own builtin globbing (not File::Glob).\n" }, { "name": "OS390 (OpenEdition MVS)", "content": "Support for this EBCDIC platform has not been renewed in this release. There are\ndifficulties in reconciling Perl's standardization on UTF-8 as its internal representation\nfor characters with the EBCDIC character set, because the two are incompatible.\n\nIt is unclear whether future versions will renew support for this platform, but the\npossibility exists.\n\nVMS\nNumerous revisions and extensions to configuration, build, testing, and installation process\nto accommodate core changes and VMS-specific options.\n\nExpand %ENV-handling code to allow runtime mapping to logical names, CLI symbols, and CRTL\nenviron array.\n\nExtension of subprocess invocation code to accept filespecs as command \"verbs\".\n\nAdd to Perl command line processing the ability to use default file types and to recognize\nUnix-style \"2>&1\".\n\nExpansion of File::Spec::VMS routines, and integration into ExtUtils::MMVMS.\n\nExtension of ExtUtils::MMVMS to handle complex extensions more flexibly.\n\nBarewords at start of Unix-syntax paths may be treated as text rather than only as logical\nnames.\n\nOptional secure translation of several logical names used internally by Perl.\n\nMiscellaneous bugfixing and porting of new core code to VMS.\n\nThanks are gladly extended to the many people who have contributed VMS patches, testing, and\nideas.\n" }, { "name": "Win32", "content": "Perl can now emulate fork() internally, using multiple interpreters running in different\nconcurrent threads. This support must be enabled at build time. See perlfork for detailed\ninformation.\n\nWhen given a pathname that consists only of a drivename, such as \"A:\", opendir() and stat()\nnow use the current working directory for the drive rather than the drive root.\n\nThe builtin XSUB functions in the Win32:: namespace are documented. See Win32.\n\n$^X now contains the full path name of the running executable.\n\nA Win32::GetLongPathName() function is provided to complement Win32::GetFullPathName() and\nWin32::GetShortPathName(). See Win32.\n\nPOSIX::uname() is supported.\n\nsystem(1,...) now returns true process IDs rather than process handles. kill() accepts any\nreal process id, rather than strictly return values from system(1,...).\n\nFor better compatibility with Unix, \"kill(0, $pid)\" can now be used to test whether a process\nexists.\n\nThe \"Shell\" module is supported.\n\nBetter support for building Perl under command.com in Windows 95 has been added.\n\nScripts are read in binary mode by default to allow ByteLoader (and the filter mechanism in\ngeneral) to work properly. For compatibility, the DATA filehandle will be set to text mode\nif a carriage return is detected at the end of the line containing the END or DATA\ntoken; if not, the DATA filehandle will be left open in binary mode. Earlier versions always\nopened the DATA filehandle in text mode.\n\nThe glob() operator is implemented via the \"File::Glob\" extension, which supports glob syntax\nof the C shell. This increases the flexibility of the glob() operator, but there may be\ncompatibility issues for programs that relied on the older globbing syntax. If you want to\npreserve compatibility with the older syntax, you might want to run perl with\n\"-MFile::DosGlob\". For details and compatibility information, see File::Glob.\n" } ] }, "Significant bug fixes": { "content": "", "subsections": [ { "name": " on empty files", "content": "With $/ set to \"undef\", \"slurping\" an empty file returns a string of zero length (instead of\n\"undef\", as it used to) the first time the HANDLE is read after $/ is set to \"undef\".\nFurther reads yield \"undef\".\n\nThis means that the following will append \"foo\" to an empty file (it used to do nothing):\n\nperl -0777 -pi -e 's/^/foo/' emptyfile\n\nThe behaviour of:\n\nperl -pi -e 's/^/foo/' emptyfile\n\nis unchanged (it continues to leave the file empty).\n" }, { "name": "\"eval '...'\" improvements", "content": "Line numbers (as reflected by caller() and most diagnostics) within \"eval '...'\" were often\nincorrect where here documents were involved. This has been corrected.\n\nLexical lookups for variables appearing in \"eval '...'\" within functions that were themselves\ncalled within an \"eval '...'\" were searching the wrong place for lexicals. The lexical\nsearch now correctly ends at the subroutine's block boundary.\n\nThe use of \"return\" within \"eval {...}\" caused $@ not to be reset correctly when no exception\noccurred within the eval. This has been fixed.\n\nParsing of here documents used to be flawed when they appeared as the replacement expression\nin \"eval 's/.../.../e'\". This has been fixed.\n" }, { "name": "All compilation errors are true errors", "content": "Some \"errors\" encountered at compile time were by necessity generated as warnings followed by\neventual termination of the program. This enabled more such errors to be reported in a\nsingle run, rather than causing a hard stop at the first error that was encountered.\n\nThe mechanism for reporting such errors has been reimplemented to queue compile-time errors\nand report them at the end of the compilation as true errors rather than as warnings. This\nfixes cases where error messages leaked through in the form of warnings when code was\ncompiled at run time using \"eval STRING\", and also allows such errors to be reliably trapped\nusing \"eval \"...\"\".\n" }, { "name": "Implicitly closed filehandles are safer", "content": "Sometimes implicitly closed filehandles (as when they are localized, and Perl automatically\ncloses them on exiting the scope) could inadvertently set $? or $!. This has been corrected.\n" }, { "name": "Behavior of list slices is more consistent", "content": "When taking a slice of a literal list (as opposed to a slice of an array or hash), Perl used\nto return an empty list if the result happened to be composed of all undef values.\n\nThe new behavior is to produce an empty list if (and only if) the original list was empty.\nConsider the following example:\n\n@a = (1,undef,undef,2)[2,1,2];\n\nThe old behavior would have resulted in @a having no elements. The new behavior ensures it\nhas three undefined elements.\n\nNote in particular that the behavior of slices of the following cases remains unchanged:\n\n@a = ()[1,2];\n@a = (getpwent)[7,0];\n@a = (anythingreturningemptylist())[2,1,2];\n@a = @b[2,1,2];\n@a = @c{'a','b','c'};\n\nSee perldata.\n" }, { "name": "\"(\\$)\" prototype and $foo{a}", "content": "A scalar reference prototype now correctly allows a hash or array element in that slot.\n" }, { "name": "\"goto &sub\" and AUTOLOAD", "content": "The \"goto &sub\" construct works correctly when &sub happens to be autoloaded.\n" }, { "name": "\"-bareword\" allowed under \"use integer\"", "content": "The autoquoting of barewords preceded by \"-\" did not work in prior versions when the\n\"integer\" pragma was enabled. This has been fixed.\n" }, { "name": "Failures in DESTROY()", "content": "When code in a destructor threw an exception, it went unnoticed in earlier versions of Perl,\nunless someone happened to be looking in $@ just after the point the destructor happened to\nrun. Such failures are now visible as warnings when warnings are enabled.\n" }, { "name": "Locale bugs fixed", "content": "printf() and sprintf() previously reset the numeric locale back to the default \"C\" locale.\nThis has been fixed.\n\nNumbers formatted according to the local numeric locale (such as using a decimal comma\ninstead of a decimal dot) caused \"isn't numeric\" warnings, even while the operations\naccessing those numbers produced correct results. These warnings have been discontinued.\n" }, { "name": "Memory leaks", "content": "The \"eval 'return sub {...}'\" construct could sometimes leak memory. This has been fixed.\n\nOperations that aren't filehandle constructors used to leak memory when used on invalid\nfilehandles. This has been fixed.\n\nConstructs that modified @ could fail to deallocate values in @ and thus leak memory. This\nhas been corrected.\n" }, { "name": "Spurious subroutine stubs after failed subroutine calls", "content": "Perl could sometimes create empty subroutine stubs when a subroutine was not found in the\npackage. Such cases stopped later method lookups from progressing into base packages. This\nhas been corrected.\n" }, { "name": "Taint failures under \"-U\"", "content": "When running in unsafe mode, taint violations could sometimes cause silent failures. This\nhas been fixed.\n" }, { "name": "END blocks and the \"-c\" switch", "content": "Prior versions used to run BEGIN and END blocks when Perl was run in compile-only mode.\nSince this is typically not the expected behavior, END blocks are not executed anymore when\nthe \"-c\" switch is used, or if compilation fails.\n\nSee \"Support for CHECK blocks\" for how to run things when the compile phase ends.\n" }, { "name": "Potential to leak DATA filehandles", "content": "Using the \"DATA\" token creates an implicit filehandle to the file that contains the\ntoken. It is the program's responsibility to close it when it is done reading from it.\n\nThis caveat is now better explained in the documentation. See perldata.\n" } ] }, "New or Changed Diagnostics": { "content": "\"%s\" variable %s masks earlier declaration in same %s\n(W misc) A \"my\" or \"our\" variable has been redeclared in the current scope or statement,\neffectively eliminating all access to the previous instance. This is almost always a\ntypographical error. Note that the earlier variable will still exist until the end of\nthe scope or until all closure referents to it are destroyed.\n\n\"my sub\" not yet implemented\n(F) Lexically scoped subroutines are not yet implemented. Don't try that yet.\n\n\"our\" variable %s redeclared\n(W misc) You seem to have already declared the same global once before in the current\nlexical scope.\n\n'!' allowed only after types %s\n(F) The '!' is allowed in pack() and unpack() only after certain types. See \"pack\" in\nperlfunc.\n\n/ cannot take a count\n(F) You had an unpack template indicating a counted-length string, but you have also\nspecified an explicit size for the string. See \"pack\" in perlfunc.\n\n/ must be followed by a, A or Z\n(F) You had an unpack template indicating a counted-length string, which must be followed\nby one of the letters a, A or Z to indicate what sort of string is to be unpacked. See\n\"pack\" in perlfunc.\n\n/ must be followed by a*, A* or Z*\n(F) You had a pack template indicating a counted-length string, Currently the only things\nthat can have their length counted are a*, A* or Z*. See \"pack\" in perlfunc.\n\n/ must follow a numeric type\n(F) You had an unpack template that contained a '#', but this did not follow some numeric\nunpack specification. See \"pack\" in perlfunc.\n\n/%s/: Unrecognized escape \\\\%c passed through\n(W regexp) You used a backslash-character combination which is not recognized by Perl.\nThis combination appears in an interpolated variable or a \"'\"-delimited regular\nexpression. The character was understood literally.\n\n/%s/: Unrecognized escape \\\\%c in character class passed through\n(W regexp) You used a backslash-character combination which is not recognized by Perl\ninside character classes. The character was understood literally.\n\n/%s/ should probably be written as \"%s\"\n(W syntax) You have used a pattern where Perl expected to find a string, as in the first\nargument to \"join\". Perl will treat the true or false result of matching the pattern\nagainst $ as the string, which is probably not what you had in mind.\n\n%s() called too early to check prototype\n(W prototype) You've called a function that has a prototype before the parser saw a\ndefinition or declaration for it, and Perl could not check that the call conforms to the\nprototype. You need to either add an early prototype declaration for the subroutine in\nquestion, or move the subroutine definition ahead of the call to get proper prototype\nchecking. Alternatively, if you are certain that you're calling the function correctly,\nyou may put an ampersand before the name to avoid the warning. See perlsub.\n\n%s argument is not a HASH or ARRAY element\n(F) The argument to exists() must be a hash or array element, such as:\n\n$foo{$bar}\n$ref->{\"susie\"}[12]\n\n%s argument is not a HASH or ARRAY element or slice\n(F) The argument to delete() must be either a hash or array element, such as:\n\n$foo{$bar}\n$ref->{\"susie\"}[12]\n\nor a hash or array slice, such as:\n\n@foo[$bar, $baz, $xyzzy]\n@{$ref->[12]}{\"susie\", \"queue\"}\n\n%s argument is not a subroutine name\n(F) The argument to exists() for \"exists &sub\" must be a subroutine name, and not a\nsubroutine call. \"exists &sub()\" will generate this error.\n\n%s package attribute may clash with future reserved word: %s\n(W reserved) A lowercase attribute name was used that had a package-specific handler.\nThat name might have a meaning to Perl itself some day, even though it doesn't yet.\nPerhaps you should use a mixed-case attribute name, instead. See attributes.\n\n(in cleanup) %s\n(W misc) This prefix usually indicates that a DESTROY() method raised the indicated\nexception. Since destructors are usually called by the system at arbitrary points during\nexecution, and often a vast number of times, the warning is issued only once for any\nnumber of failures that would otherwise result in the same message being repeated.\n\nFailure of user callbacks dispatched using the \"GKEEPERR\" flag could also result in this\nwarning. See \"GKEEPERR\" in perlcall.\n\n<> should be quotes\n(F) You wrote \"require \" when you should have written \"require 'file'\".\n\nAttempt to join self\n(F) You tried to join a thread from within itself, which is an impossible task. You may\nbe joining the wrong thread, or you may need to move the join() to some other thread.\n\nBad evalled substitution pattern\n(F) You've used the /e switch to evaluate the replacement for a substitution, but perl\nfound a syntax error in the code to evaluate, most likely an unexpected right brace '}'.\n\nBad realloc() ignored\n(S) An internal routine called realloc() on something that had never been malloc()ed in\nthe first place. Mandatory, but can be disabled by setting environment variable\n\"PERLBADFREE\" to 1.\n\nBareword found in conditional\n(W bareword) The compiler found a bareword where it expected a conditional, which often\nindicates that an || or && was parsed as part of the last argument of the previous\nconstruct, for example:\n\nopen FOO || die;\n\nIt may also indicate a misspelled constant that has been interpreted as a bareword:\n\nuse constant TYPO => 1;\nif (TYOP) { print \"foo\" }\n\nThe \"strict\" pragma is useful in avoiding such errors.\n\nBinary number > 0b11111111111111111111111111111111 non-portable\n(W portable) The binary number you specified is larger than 232-1 (4294967295) and\ntherefore non-portable between systems. See perlport for more on portability concerns.\n\nBit vector size > 32 non-portable\n(W portable) Using bit vector sizes larger than 32 is non-portable.\n\nBuffer overflow in primeenviter: %s\n(W internal) A warning peculiar to VMS. While Perl was preparing to iterate over %ENV,\nit encountered a logical name or symbol definition which was too long, so it was\ntruncated to the string shown.\n\nCan't check filesystem of script \"%s\"\n(P) For some reason you can't check the filesystem of the script for nosuid.\n\nCan't declare class for non-scalar %s in \"%s\"\n(S) Currently, only scalar variables can declared with a specific class qualifier in a\n\"my\" or \"our\" declaration. The semantics may be extended for other types of variables in\nfuture.\n\nCan't declare %s in \"%s\"\n(F) Only scalar, array, and hash variables may be declared as \"my\" or \"our\" variables.\nThey must have ordinary identifiers as names.\n\nCan't ignore signal CHLD, forcing to default\n(W signal) Perl has detected that it is being run with the SIGCHLD signal (sometimes\nknown as SIGCLD) disabled. Since disabling this signal will interfere with proper\ndetermination of exit status of child processes, Perl has reset the signal to its default\nvalue. This situation typically indicates that the parent program under which Perl may\nbe running (e.g., cron) is being very careless.\n\nCan't modify non-lvalue subroutine call\n(F) Subroutines meant to be used in lvalue context should be declared as such, see\n\"Lvalue subroutines\" in perlsub.\n\nCan't read CRTL environ\n(S) A warning peculiar to VMS. Perl tried to read an element of %ENV from the CRTL's\ninternal environment array and discovered the array was missing. You need to figure out\nwhere your CRTL misplaced its environ or define PERLENVTABLES (see perlvms) so that\nenviron is not searched.\n\nCan't remove %s: %s, skipping file\n(S) You requested an inplace edit without creating a backup file. Perl was unable to\nremove the original file to replace it with the modified file. The file was left\nunmodified.\n\nCan't return %s from lvalue subroutine\n(F) Perl detected an attempt to return illegal lvalues (such as temporary or readonly\nvalues) from a subroutine used as an lvalue. This is not allowed.\n\nCan't weaken a nonreference\n(F) You attempted to weaken something that was not a reference. Only references can be\nweakened.\n\nCharacter class [:%s:] unknown\n(F) The class in the character class [: :] syntax is unknown. See perlre.\n\nCharacter class syntax [%s] belongs inside character classes\n(W unsafe) The character class constructs [: :], [= =], and [. .] go inside character\nclasses, the [] are part of the construct, for example: /[012[:alpha:]345]/. Note that\n[= =] and [. .] are not currently implemented; they are simply placeholders for future\nextensions.\n\nConstant is not %s reference\n(F) A constant value (perhaps declared using the \"use constant\" pragma) is being\ndereferenced, but it amounts to the wrong type of reference. The message indicates the\ntype of reference that was expected. This usually indicates a syntax error in\ndereferencing the constant value. See \"Constant Functions\" in perlsub and constant.\n\nconstant(%s): %s\n(F) The parser found inconsistencies either while attempting to define an overloaded\nconstant, or when trying to find the character name specified in the \"\\N{...}\" escape.\nPerhaps you forgot to load the corresponding \"overload\" or \"charnames\" pragma? See\ncharnames and overload.\n\nCORE::%s is not a keyword\n(F) The CORE:: namespace is reserved for Perl keywords.\n\ndefined(@array) is deprecated\n(D) defined() is not usually useful on arrays because it checks for an undefined scalar\nvalue. If you want to see if the array is empty, just use \"if (@array) { # not empty }\"\nfor example.\n\ndefined(%hash) is deprecated\n(D) defined() is not usually useful on hashes because it checks for an undefined scalar\nvalue. If you want to see if the hash is empty, just use \"if (%hash) { # not empty }\"\nfor example.\n\nDid not produce a valid header\nSee Server error.\n\n(Did you mean \"local\" instead of \"our\"?)\n(W misc) Remember that \"our\" does not localize the declared global variable. You have\ndeclared it again in the same lexical scope, which seems superfluous.\n\nDocument contains no data\nSee Server error.\n\nentering effective %s failed\n(F) While under the \"use filetest\" pragma, switching the real and effective uids or gids\nfailed.\n\nfalse [] range \"%s\" in regexp\n(W regexp) A character class range must start and end at a literal character, not another\ncharacter class like \"\\d\" or \"[:alpha:]\". The \"-\" in your false range is interpreted as\na literal \"-\". Consider quoting the \"-\", \"\\-\". See perlre.\n\nFilehandle %s opened only for output\n(W io) You tried to read from a filehandle opened only for writing. If you intended it\nto be a read/write filehandle, you needed to open it with \"+<\" or \"+>\" or \"+>>\" instead\nof with \"<\" or nothing. If you intended only to read from the file, use \"<\". See \"open\"\nin perlfunc.\n\nflock() on closed filehandle %s\n(W closed) The filehandle you're attempting to flock() got itself closed some time before\nnow. Check your logic flow. flock() operates on filehandles. Are you attempting to\ncall flock() on a dirhandle by the same name?\n\nGlobal symbol \"%s\" requires explicit package name\n(F) You've said \"use strict vars\", which indicates that all variables must either be\nlexically scoped (using \"my\"), declared beforehand using \"our\", or explicitly qualified\nto say which package the global variable is in (using \"::\").\n\nHexadecimal number > 0xffffffff non-portable\n(W portable) The hexadecimal number you specified is larger than 232-1 (4294967295) and\ntherefore non-portable between systems. See perlport for more on portability concerns.\n\nIll-formed CRTL environ value \"%s\"\n(W internal) A warning peculiar to VMS. Perl tried to read the CRTL's internal environ\narray, and encountered an element without the \"=\" delimiter used to separate keys from\nvalues. The element is ignored.\n\nIll-formed message in primeenviter: |%s|\n(W internal) A warning peculiar to VMS. Perl tried to read a logical name or CLI symbol\ndefinition when preparing to iterate over %ENV, and didn't see the expected delimiter\nbetween key and value, so the line was ignored.\n\nIllegal binary digit %s\n(F) You used a digit other than 0 or 1 in a binary number.\n\nIllegal binary digit %s ignored\n(W digit) You may have tried to use a digit other than 0 or 1 in a binary number.\nInterpretation of the binary number stopped before the offending digit.\n\nIllegal number of bits in vec\n(F) The number of bits in vec() (the third argument) must be a power of two from 1 to 32\n(or 64, if your platform supports that).\n\nInteger overflow in %s number\n(W overflow) The hexadecimal, octal or binary number you have specified either as a\nliteral or as an argument to hex() or oct() is too big for your architecture, and has\nbeen converted to a floating point number. On a 32-bit architecture the largest\nhexadecimal, octal or binary number representable without overflow is 0xFFFFFFFF,\n037777777777, or 0b11111111111111111111111111111111 respectively. Note that Perl\ntransparently promotes all numbers to a floating point representation internally--subject\nto loss of precision errors in subsequent operations.\n\nInvalid %s attribute: %s\nThe indicated attribute for a subroutine or variable was not recognized by Perl or by a\nuser-supplied handler. See attributes.\n\nInvalid %s attributes: %s\nThe indicated attributes for a subroutine or variable were not recognized by Perl or by a\nuser-supplied handler. See attributes.\n\ninvalid [] range \"%s\" in regexp\nThe offending range is now explicitly displayed.\n\nInvalid separator character %s in attribute list\n(F) Something other than a colon or whitespace was seen between the elements of an\nattribute list. If the previous attribute had a parenthesised parameter list, perhaps\nthat list was terminated too soon. See attributes.\n\nInvalid separator character %s in subroutine attribute list\n(F) Something other than a colon or whitespace was seen between the elements of a\nsubroutine attribute list. If the previous attribute had a parenthesised parameter list,\nperhaps that list was terminated too soon.\n\nleaving effective %s failed\n(F) While under the \"use filetest\" pragma, switching the real and effective uids or gids\nfailed.\n\nLvalue subs returning %s not implemented yet\n(F) Due to limitations in the current implementation, array and hash values cannot be\nreturned in subroutines used in lvalue context. See \"Lvalue subroutines\" in perlsub.\n\nMethod %s not permitted\nSee Server error.\n\nMissing %sbrace%s on \\N{}\n(F) Wrong syntax of character name literal \"\\N{charname}\" within double-quotish context.\n\nMissing command in piped open\n(W pipe) You used the \"open(FH, \"| command\")\" or \"open(FH, \"command |\")\" construction,\nbut the command was missing or blank.\n\nMissing name in \"my sub\"\n(F) The reserved syntax for lexically scoped subroutines requires that they have a name\nwith which they can be found.\n\nNo %s specified for -%c\n(F) The indicated command line switch needs a mandatory argument, but you haven't\nspecified one.\n\nNo package name allowed for variable %s in \"our\"\n(F) Fully qualified variable names are not allowed in \"our\" declarations, because that\ndoesn't make much sense under existing semantics. Such syntax is reserved for future\nextensions.\n\nNo space allowed after -%c\n(F) The argument to the indicated command line switch must follow immediately after the\nswitch, without intervening spaces.\n\nno UTC offset information; assuming local time is UTC\n(S) A warning peculiar to VMS. Perl was unable to find the local timezone offset, so\nit's assuming that local system time is equivalent to UTC. If it's not, define the\nlogical name SYS$TIMEZONEDIFFERENTIAL to translate to the number of seconds which need\nto be added to UTC to get local time.\n\nOctal number > 037777777777 non-portable\n(W portable) The octal number you specified is larger than 232-1 (4294967295) and\ntherefore non-portable between systems. See perlport for more on portability concerns.\n\nSee also perlport for writing portable code.\n\npanic: delbackref\n(P) Failed an internal consistency check while trying to reset a weak reference.\n\npanic: kid popen errno read\n(F) forked child returned an incomprehensible message about its errno.\n\npanic: magickillbackrefs\n(P) Failed an internal consistency check while trying to reset all weak references to an\nobject.\n\nParentheses missing around \"%s\" list\n(W parenthesis) You said something like\n\nmy $foo, $bar = @;\n\nwhen you meant\n\nmy ($foo, $bar) = @;\n\nRemember that \"my\", \"our\", and \"local\" bind tighter than comma.\n\nPossible unintended interpolation of %s in string\n(W ambiguous) It used to be that Perl would try to guess whether you wanted an array\ninterpolated or a literal @. It no longer does this; arrays are now always interpolated\ninto strings. This means that if you try something like:\n\nprint \"fred@example.com\";\n\nand the array @example doesn't exist, Perl is going to print \"fred.com\", which is\nprobably not what you wanted. To get a literal \"@\" sign in a string, put a backslash\nbefore it, just as you would to get a literal \"$\" sign.\n\nPossible Y2K bug: %s\n(W y2k) You are concatenating the number 19 with another number, which could be a\npotential Year 2000 problem.\n\npragma \"attrs\" is deprecated, use \"sub NAME : ATTRS\" instead\n(W deprecated) You have written something like this:\n\nsub doit\n{\nuse attrs qw(locked);\n}\n\nYou should use the new declaration syntax instead.\n\nsub doit : locked\n{\n...\n\nThe \"use attrs\" pragma is now obsolete, and is only provided for backward-compatibility.\nSee \"Subroutine Attributes\" in perlsub.\n\nPremature end of script headers\nSee Server error.\n\nRepeat count in pack overflows\n(F) You can't specify a repeat count so large that it overflows your signed integers.\nSee \"pack\" in perlfunc.\n\nRepeat count in unpack overflows\n(F) You can't specify a repeat count so large that it overflows your signed integers.\nSee \"unpack\" in perlfunc.\n\nrealloc() of freed memory ignored\n(S) An internal routine called realloc() on something that had already been freed.\n\nReference is already weak\n(W misc) You have attempted to weaken a reference that is already weak. Doing so has no\neffect.\n\nsetpgrp can't take arguments\n(F) Your system has the setpgrp() from BSD 4.2, which takes no arguments, unlike POSIX\nsetpgid(), which takes a process ID and process group ID.\n\nStrange *+?{} on zero-length expression\n(W regexp) You applied a regular expression quantifier in a place where it makes no\nsense, such as on a zero-width assertion. Try putting the quantifier inside the\nassertion instead. For example, the way to match \"abc\" provided that it is followed by\nthree repetitions of \"xyz\" is \"/abc(?=(?:xyz){3})/\", not \"/abc(?=xyz){3}/\".\n\nswitching effective %s is not implemented\n(F) While under the \"use filetest\" pragma, we cannot switch the real and effective uids\nor gids.\n\nThis Perl can't reset CRTL environ elements (%s)\nThis Perl can't set CRTL environ elements (%s=%s)\n(W internal) Warnings peculiar to VMS. You tried to change or delete an element of the\nCRTL's internal environ array, but your copy of Perl wasn't built with a CRTL that\ncontained the setenv() function. You'll need to rebuild Perl with a CRTL that does, or\nredefine PERLENVTABLES (see perlvms) so that the environ array isn't the target of the\nchange to %ENV which produced the warning.\n\nToo late to run %s block\n(W void) A CHECK or INIT block is being defined during run time proper, when the\nopportunity to run them has already passed. Perhaps you are loading a file with\n\"require\" or \"do\" when you should be using \"use\" instead. Or perhaps you should put the\n\"require\" or \"do\" inside a BEGIN block.\n\nUnknown open() mode '%s'\n(F) The second argument of 3-argument open() is not among the list of valid modes: \"<\",\n\">\", \">>\", \"+<\", \"+>\", \"+>>\", \"-|\", \"|-\".\n\nUnknown process %x sent message to primeenviter: %s\n(P) An error peculiar to VMS. Perl was reading values for %ENV before iterating over it,\nand someone else stuck a message in the stream of data Perl expected. Someone's very\nconfused, or perhaps trying to subvert Perl's population of %ENV for nefarious purposes.\n\nUnrecognized escape \\\\%c passed through\n(W misc) You used a backslash-character combination which is not recognized by Perl. The\ncharacter was understood literally.\n\nUnterminated attribute parameter in attribute list\n(F) The lexer saw an opening (left) parenthesis character while parsing an attribute\nlist, but the matching closing (right) parenthesis character was not found. You may need\nto add (or remove) a backslash character to get your parentheses to balance. See\nattributes.\n\nUnterminated attribute list\n(F) The lexer found something other than a simple identifier at the start of an\nattribute, and it wasn't a semicolon or the start of a block. Perhaps you terminated the\nparameter list of the previous attribute too soon. See attributes.\n\nUnterminated attribute parameter in subroutine attribute list\n(F) The lexer saw an opening (left) parenthesis character while parsing a subroutine\nattribute list, but the matching closing (right) parenthesis character was not found.\nYou may need to add (or remove) a backslash character to get your parentheses to balance.\n\nUnterminated subroutine attribute list\n(F) The lexer found something other than a simple identifier at the start of a subroutine\nattribute, and it wasn't a semicolon or the start of a block. Perhaps you terminated the\nparameter list of the previous attribute too soon.\n\nValue of CLI symbol \"%s\" too long\n(W misc) A warning peculiar to VMS. Perl tried to read the value of an %ENV element from\na CLI symbol table, and found a resultant string longer than 1024 characters. The return\nvalue has been truncated to 1024 characters.\n\nVersion number must be a constant number\n(P) The attempt to translate a \"use Module n.n LIST\" statement into its equivalent\n\"BEGIN\" block found an internal inconsistency with the version number.\n\nNew tests\nlib/attrs\nCompatibility tests for \"sub : attrs\" vs the older \"use attrs\".\n\nlib/env\nTests for new environment scalar capability (e.g., \"use Env qw($BAR);\").\n\nlib/env-array\nTests for new environment array capability (e.g., \"use Env qw(@PATH);\").\n\nlib/ioconst\nIO constants (SEEK*, IO*).\n\nlib/iodir\nDirectory-related IO methods (new, read, close, rewind, tied delete).\n\nlib/iomultihomed\nINET sockets with multi-homed hosts.\n\nlib/iopoll\nIO poll().\n\nlib/iounix\nUNIX sockets.\n\nop/attrs\nRegression tests for \"my ($x,@y,%z) : attrs\" and .\n\nop/filetest\nFile test operators.\n\nop/lexassign\nVerify operations that access pad objects (lexicals and temporaries).\n\nop/existssub\nVerify \"exists &sub\" operations.\n", "subsections": [] }, "Incompatible Changes": { "content": "", "subsections": [ { "name": "Perl Source Incompatibilities", "content": "Beware that any new warnings that have been added or old ones that have been enhanced are not\nconsidered incompatible changes.\n\nSince all new warnings must be explicitly requested via the \"-w\" switch or the \"warnings\"\npragma, it is ultimately the programmer's responsibility to ensure that warnings are enabled\njudiciously.\n\nCHECK is a new keyword\nAll subroutine definitions named CHECK are now special. See \"/\"Support for CHECK\nblocks\"\" for more information.\n\nTreatment of list slices of undef has changed\nThere is a potential incompatibility in the behavior of list slices that are comprised\nentirely of undefined values. See \"Behavior of list slices is more consistent\".\n\nFormat of $English::PERLVERSION is different\nThe English module now sets $PERLVERSION to $^V (a string value) rather than $] (a\nnumeric value). This is a potential incompatibility. Send us a report via perlbug if\nyou are affected by this.\n\nSee \"Improved Perl version numbering system\" for the reasons for this change.\n\nLiterals of the form 1.2.3 parse differently\nPreviously, numeric literals with more than one dot in them were interpreted as a\nfloating point number concatenated with one or more numbers. Such \"numbers\" are now\nparsed as strings composed of the specified ordinals.\n\nFor example, \"print 97.98.99\" used to output 97.9899 in earlier versions, but now prints\n\"abc\".\n\nSee \"Support for strings represented as a vector of ordinals\".\n\nPossibly changed pseudo-random number generator\nPerl programs that depend on reproducing a specific set of pseudo-random numbers may now\nproduce different output due to improvements made to the rand() builtin. You can use \"sh\nConfigure -Drandfunc=rand\" to obtain the old behavior.\n\nSee \"Better pseudo-random number generator\".\n\nHashing function for hash keys has changed\nEven though Perl hashes are not order preserving, the apparently random order encountered\nwhen iterating on the contents of a hash is actually determined by the hashing algorithm\nused. Improvements in the algorithm may yield a random order that is different from that\nof previous versions, especially when iterating on hashes.\n\nSee \"Better worst-case behavior of hashes\" for additional information.\n\n\"undef\" fails on read only values\nUsing the \"undef\" operator on a readonly value (such as $1) has the same effect as\nassigning \"undef\" to the readonly value--it throws an exception.\n\nClose-on-exec bit may be set on pipe and socket handles\nPipe and socket handles are also now subject to the close-on-exec behavior determined by\nthe special variable $^F.\n\nSee \"More consistent close-on-exec behavior\".\n\nWriting \"$$1\" to mean \"${$}1\" is unsupported\nPerl 5.004 deprecated the interpretation of $$1 and similar within interpolated strings\nto mean \"$$ . \"1\"\", but still allowed it.\n\nIn Perl 5.6.0 and later, \"$$1\" always means \"${$1}\".\n\ndelete(), each(), values() and \"\\(%h)\"\noperate on aliases to values, not copies\n\ndelete(), each(), values() and hashes (e.g. \"\\(%h)\") in a list context return the actual\nvalues in the hash, instead of copies (as they used to in earlier versions). Typical\nidioms for using these constructs copy the returned values, but this can make a\nsignificant difference when creating references to the returned values. Keys in the hash\nare still returned as copies when iterating on a hash.\n\nSee also \"delete(), each(), values() and hash iteration are faster\".\n\nvec(EXPR,OFFSET,BITS) enforces powers-of-two BITS\nvec() generates a run-time error if the BITS argument is not a valid power-of-two\ninteger.\n\nText of some diagnostic output has changed\nMost references to internal Perl operations in diagnostics have been changed to be more\ndescriptive. This may be an issue for programs that may incorrectly rely on the exact\ntext of diagnostics for proper functioning.\n\n\"%@\" has been removed\nThe undocumented special variable \"%@\" that used to accumulate \"background\" errors (such\nas those that happen in DESTROY()) has been removed, because it could potentially result\nin memory leaks.\n\nParenthesized not() behaves like a list operator\nThe \"not\" operator now falls under the \"if it looks like a function, it behaves like a\nfunction\" rule.\n\nAs a result, the parenthesized form can be used with \"grep\" and \"map\". The following\nconstruct used to be a syntax error before, but it works as expected now:\n\ngrep not($), @things;\n\nOn the other hand, using \"not\" with a literal list slice may not work. The following\npreviously allowed construct:\n\nprint not (1,2,3)[0];\n\nneeds to be written with additional parentheses now:\n\nprint not((1,2,3)[0]);\n\nThe behavior remains unaffected when \"not\" is not followed by parentheses.\n\nSemantics of bareword prototype \"(*)\" have changed\nThe semantics of the bareword prototype \"*\" have changed. Perl 5.005 always coerced\nsimple scalar arguments to a typeglob, which wasn't useful in situations where the\nsubroutine must distinguish between a simple scalar and a typeglob. The new behavior is\nto not coerce bareword arguments to a typeglob. The value will always be visible as\neither a simple scalar or as a reference to a typeglob.\n\nSee \"More functional bareword prototype (*)\".\n\nSemantics of bit operators may have changed on 64-bit platforms\nIf your platform is either natively 64-bit or if Perl has been configured to used 64-bit\nintegers, i.e., $Config{ivsize} is 8, there may be a potential incompatibility in the\nbehavior of bitwise numeric operators (& | ^ ~ << >>). These operators used to strictly\noperate on the lower 32 bits of integers in previous versions, but now operate over the\nentire native integral width. In particular, note that unary \"~\" will produce different\nresults on platforms that have different $Config{ivsize}. For portability, be sure to\nmask off the excess bits in the result of unary \"~\", e.g., \"~$x & 0xffffffff\".\n\nSee \"Bit operators support full native integer width\".\n\nMore builtins taint their results\nAs described in \"Improved security features\", there may be more sources of taint in a\nPerl program.\n\nTo avoid these new tainting behaviors, you can build Perl with the Configure option\n\"-Accflags=-DINCOMPLETETAINTS\". Beware that the ensuing perl binary may be insecure.\n" }, { "name": "C Source Incompatibilities", "content": "\"PERLPOLLUTE\"\nRelease 5.005 grandfathered old global symbol names by providing preprocessor macros for\nextension source compatibility. As of release 5.6.0, these preprocessor definitions are\nnot available by default. You need to explicitly compile perl with \"-DPERLPOLLUTE\" to\nget these definitions. For extensions still using the old symbols, this option can be\nspecified via MakeMaker:\n\nperl Makefile.PL POLLUTE=1\n\n\"PERLIMPLICITCONTEXT\"\nThis new build option provides a set of macros for all API functions such that an\nimplicit interpreter/thread context argument is passed to every API function. As a\nresult of this, something like \"svsetsv(foo,bar)\" amounts to a macro invocation that\nactually translates to something like \"Perlsvsetsv(myperl,foo,bar)\". While this is\ngenerally expected to not have any significant source compatibility issues, the\ndifference between a macro and a real function call will need to be considered.\n\nThis means that there is a source compatibility issue as a result of this if your\nextensions attempt to use pointers to any of the Perl API functions.\n\nNote that the above issue is not relevant to the default build of Perl, whose interfaces\ncontinue to match those of prior versions (but subject to the other options described\nhere).\n\nSee \"Background and PERLIMPLICITCONTEXT\" in perlguts for detailed information on the\nramifications of building Perl with this option.\n\nNOTE: PERLIMPLICITCONTEXT is automatically enabled whenever Perl is built\nwith one of -Dusethreads, -Dusemultiplicity, or both. It is not\nintended to be enabled by users at this time.\n\n\"PERLPOLLUTEMALLOC\"\nEnabling Perl's malloc in release 5.005 and earlier caused the namespace of the system's\nmalloc family of functions to be usurped by the Perl versions, since by default they used\nthe same names. Besides causing problems on platforms that do not allow these functions\nto be cleanly replaced, this also meant that the system versions could not be called in\nprograms that used Perl's malloc. Previous versions of Perl have allowed this behaviour\nto be suppressed with the HIDEMYMALLOC and EMBEDMYMALLOC preprocessor definitions.\n\nAs of release 5.6.0, Perl's malloc family of functions have default names distinct from\nthe system versions. You need to explicitly compile perl with \"-DPERLPOLLUTEMALLOC\" to\nget the older behaviour. HIDEMYMALLOC and EMBEDMYMALLOC have no effect, since the\nbehaviour they enabled is now the default.\n\nNote that these functions do not constitute Perl's memory allocation API. See \"Memory\nAllocation\" in perlguts for further information about that.\n" }, { "name": "Compatible C Source API Changes", "content": "\"PATCHLEVEL\" is now \"PERLVERSION\"\nThe cpp macros \"PERLREVISION\", \"PERLVERSION\", and \"PERLSUBVERSION\" are now available\nby default from perl.h, and reflect the base revision, patchlevel, and subversion\nrespectively. \"PERLREVISION\" had no prior equivalent, while \"PERLVERSION\" and\n\"PERLSUBVERSION\" were previously available as \"PATCHLEVEL\" and \"SUBVERSION\".\n\nThe new names cause less pollution of the cpp namespace and reflect what the numbers have\ncome to stand for in common practice. For compatibility, the old names are still\nsupported when patchlevel.h is explicitly included (as required before), so there is no\nsource incompatibility from the change.\n" }, { "name": "Binary Incompatibilities", "content": "In general, the default build of this release is expected to be binary compatible for\nextensions built with the 5.005 release or its maintenance versions. However, specific\nplatforms may have broken binary compatibility due to changes in the defaults used in hints\nfiles. Therefore, please be sure to always check the platform-specific README files for any\nnotes to the contrary.\n\nThe usethreads or usemultiplicity builds are not binary compatible with the corresponding\nbuilds in 5.005.\n\nOn platforms that require an explicit list of exports (AIX, OS/2 and Windows, among others),\npurely internal symbols such as parser functions and the run time opcodes are not exported by\ndefault. Perl 5.005 used to export all functions irrespective of whether they were\nconsidered part of the public API or not.\n\nFor the full list of public API functions, see perlapi.\n" } ] }, "Known Problems": { "content": "", "subsections": [ { "name": "Localizing a tied hash element may leak memory", "content": "As of the 5.6.1 release, there is a known leak when code such as this is executed:\n\nuse Tie::Hash;\ntie my %tiehash => 'Tie::StdHash';\n\n...\n\nlocal($tiehash{Foo}) = 1; # leaks\n" }, { "name": "Known test failures", "content": "• 64-bit builds\n\nSubtest #15 of lib/b.t may fail under 64-bit builds on platforms such as HP-UX PA64 and\nLinux IA64. The issue is still being investigated.\n\nThe lib/iomultihomed test may hang in HP-UX if Perl has been configured to be 64-bit.\nBecause other 64-bit platforms do not hang in this test, HP-UX is suspect. All other\ntests pass in 64-bit HP-UX. The test attempts to create and connect to \"multihomed\"\nsockets (sockets which have multiple IP addresses).\n\nNote that 64-bit support is still experimental.\n\n• Failure of Thread tests\n\nThe subtests 19 and 20 of lib/thr5005.t test are known to fail due to fundamental\nproblems in the 5.005 threading implementation. These are not new failures--Perl\n5.0050x has the same bugs, but didn't have these tests. (Note that support for\n5.005-style threading remains experimental.)\n\n• NEXTSTEP 3.3 POSIX test failure\n\nIn NEXTSTEP 3.3p2 the implementation of the strftime(3) in the operating system libraries\nis buggy: the %j format numbers the days of a month starting from zero, which, while\nbeing logical to programmers, will cause the subtests 19 to 27 of the lib/posix test may\nfail.\n\n• Tru64 (aka Digital UNIX, aka DEC OSF/1) lib/sdbm test failure with gcc\n\nIf compiled with gcc 2.95 the lib/sdbm test will fail (dump core). The cure is to use\nthe vendor cc, it comes with the operating system and produces good code.\n" }, { "name": "EBCDIC platforms not fully supported", "content": "In earlier releases of Perl, EBCDIC environments like OS390 (also known as Open Edition MVS)\nand VM-ESA were supported. Due to changes required by the UTF-8 (Unicode) support, the\nEBCDIC platforms are not supported in Perl 5.6.0.\n\nThe 5.6.1 release improves support for EBCDIC platforms, but they are not fully supported\nyet.\n" }, { "name": "UNICOS/mk CC failures during Configure run", "content": "In UNICOS/mk the following errors may appear during the Configure run:\n\nGuessing which symbols your C compiler and preprocessor define...\nCC-20 cc: ERROR File = try.c, Line = 3\n...\nbad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K\n...\n4 errors detected in the compilation of \"try.c\".\n\nThe culprit is the broken awk of UNICOS/mk. The effect is fortunately rather mild: Perl\nitself is not adversely affected by the error, only the h2ph utility coming with Perl, and\nthat is rather rarely needed these days.\n" }, { "name": "Arrow operator and arrays", "content": "When the left argument to the arrow operator \"->\" is an array, or the \"scalar\" operator\noperating on an array, the result of the operation must be considered erroneous. For example:\n\n@x->[2]\nscalar(@x)->[2]\n\nThese expressions will get run-time errors in some future release of Perl.\n" }, { "name": "Experimental features", "content": "As discussed above, many features are still experimental. Interfaces and implementation of\nthese features are subject to change, and in extreme cases, even subject to removal in some\nfuture release of Perl. These features include the following:\n\nThreads\nUnicode\n64-bit support\nLvalue subroutines\nWeak references\nThe pseudo-hash data type\nThe Compiler suite\nInternal implementation of file globbing\nThe DB module\nThe regular expression code constructs:\n\"(?{ code })\" and \"(??{ code })\"\n" } ] }, "Obsolete Diagnostics": { "content": "Character class syntax [: :] is reserved for future extensions\n(W) Within regular expression character classes ([]) the syntax beginning with \"[:\" and\nending with \":]\" is reserved for future extensions. If you need to represent those\ncharacter sequences inside a regular expression character class, just quote the square\nbrackets with the backslash: \"\\[:\" and \":\\]\".\n\nIll-formed logical name |%s| in primeenviter\n(W) A warning peculiar to VMS. A logical name was encountered when preparing to iterate\nover %ENV which violates the syntactic rules governing logical names. Because it cannot\nbe translated normally, it is skipped, and will not appear in %ENV. This may be a benign\noccurrence, as some software packages might directly modify logical name tables and\nintroduce nonstandard names, or it may indicate that a logical name table has been\ncorrupted.\n\nIn string, @%s now must be written as \\@%s\nThe description of this error used to say:\n\n(Someday it will simply assume that an unbackslashed @\ninterpolates an array.)\n\nThat day has come, and this fatal error has been removed. It has been replaced by a non-\nfatal warning instead. See \"Arrays now always interpolate into double-quoted strings\"\nfor details.\n\nProbable precedence problem on %s\n(W) The compiler found a bareword where it expected a conditional, which often indicates\nthat an || or && was parsed as part of the last argument of the previous construct, for\nexample:\n\nopen FOO || die;\n\nregexp too big\n(F) The current implementation of regular expressions uses shorts as address offsets\nwithin a string. Unfortunately this means that if the regular expression compiles to\nlonger than 32767, it'll blow up. Usually when you want a regular expression this big,\nthere is a better way to do it with multiple statements. See perlre.\n\nUse of \"$$\" to mean \"${$}\" is deprecated\n(D) Perl versions before 5.004 misinterpreted any type marker followed by \"$\" and a\ndigit. For example, \"$$0\" was incorrectly taken to mean \"${$}0\" instead of \"${$0}\".\nThis bug is (mostly) fixed in Perl 5.004.\n\nHowever, the developers of Perl 5.004 could not fix this bug completely, because at least\ntwo widely-used modules depend on the old meaning of \"$$0\" in a string. So Perl 5.004\nstill interprets \"$$\" in the old (broken) way inside strings; but it generates\nthis message as a warning. And in Perl 5.005, this special treatment will cease.\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. There may also be information at http://www.perl.com/ , the\nPerl 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", "subsections": [] }, "SEE ALSO": { "content": "The Changes file for 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", "subsections": [] }, "HISTORY": { "content": "Written by Gurusamy Sarathy , with many contributions from The Perl\nPorters.\n\nSend omissions or corrections to .\n\nperl v5.38.2 2026-06-12 PERL561DELTA(1)", "subsections": [] } }, "summary": "perl561delta - what's new for perl v5.6.1", "flags": [], "examples": [], "see_also": [] }