{ "content": [ { "type": "text", "text": "# PERL5100DELTA (man)\n\n## NAME\n\nperl5100delta - what is new for perl 5.10.0\n\n## DESCRIPTION\n\nThis document describes the differences between the 5.8.8 release and the 5.10.0 release.\n\n## Sections\n\n- **NAME**\n- **DESCRIPTION**\n- **Core Enhancements** (29 subsections)\n- **Incompatible Changes** (20 subsections)\n- **Modules and Pragmata** (4 subsections)\n- **Utility Changes**\n- **New Documentation**\n- **Performance Enhancements** (11 subsections)\n- **Installation and Configuration Improvements** (4 subsections)\n- **Selected Bug Fixes**\n- **New or Changed Diagnostics**\n- **Changed Internals** (11 subsections)\n- **Known Problems** (1 subsections)\n- **Platform Specific Problems**\n- **Reporting Bugs**\n- **SEE ALSO**\n\nUse structuredContent.sections for detailed options, examples, and full documentation.\n" } ], "structuredContent": { "command": "PERL5100DELTA", "section": "", "mode": "man", "summary": "perl5100delta - what is new for perl 5.10.0", "synopsis": null, "tldr_summary": null, "tldr_examples": [], "tldr_source": null, "flags": [ { "flag": "-e", "long": null, "arg": null, "description": "\":5.10\"\")." } ], "examples": [], "see_also": [], "section_outline": [ { "name": "NAME", "lines": 2, "subsections": [] }, { "name": "DESCRIPTION", "lines": 5, "subsections": [] }, { "name": "Core Enhancements", "lines": 1, "subsections": [ { "name": "The \"feature\" pragma", "lines": 11 }, { "name": "New -E command-line switch", "lines": 1 }, { "name": "-E -e", "lines": 2, "flag": "-e" }, { "name": "Defined-or operator", "lines": 21 }, { "name": "Switch and Smart Match operator", "lines": 18 }, { "name": "Regular expressions", "lines": 113 }, { "name": "say()", "lines": 4 }, { "name": "Lexical $_", "lines": 15 }, { "name": "The \"_\" prototype", "lines": 8 }, { "name": "UNITCHECK blocks", "lines": 9 }, { "name": "New Pragma, \"mro\"", "lines": 11 }, { "name": "readdir() may return a \"short filename\" on Windows", "lines": 18 }, { "name": "readpipe() is now overridable", "lines": 4 }, { "name": "Default argument for readline()", "lines": 2 }, { "name": "state() variables", "lines": 12 }, { "name": "Stacked filetest operators", "lines": 3 }, { "name": "UNIVERSAL::DOES()", "lines": 7 }, { "name": "Formats", "lines": 5 }, { "name": "Byte-order modifiers for pack() and unpack()", "lines": 4 }, { "name": "\"no VERSION\"", "lines": 3 }, { "name": "\"chdir\", \"chmod\" and \"chown\" on filehandles", "lines": 4 }, { "name": "OS groups", "lines": 3 }, { "name": "Recursive sort subs", "lines": 2 }, { "name": "Exceptions in constant folding", "lines": 6 }, { "name": "Source filters in @INC", "lines": 5 }, { "name": "New internal variables", "lines": 15 }, { "name": "Miscellaneous", "lines": 11 }, { "name": "UCD 5.0.0", "lines": 10 }, { "name": "kill() on Windows", "lines": 3 } ] }, { "name": "Incompatible Changes", "lines": 1, "subsections": [ { "name": "Packing and UTF-8 strings", "lines": 30 }, { "name": "Byte/character count feature in unpack()", "lines": 10 }, { "name": "substr() lvalues are no longer fixed-length", "lines": 5 }, { "name": "Parsing of \"-f _\"", "lines": 3 }, { "name": "\":unique\"", "lines": 3 }, { "name": "Effect of pragmas in eval", "lines": 5 }, { "name": "chdir FOO", "lines": 3 }, { "name": "Handling of .pmc files", "lines": 7 }, { "name": "$^V is now a \"version\" object instead of a v-string", "lines": 5 }, { "name": "@- and @+ in patterns", "lines": 3 }, { "name": "$AUTOLOAD can now be tainted", "lines": 3 }, { "name": "Tainting and printf", "lines": 3 }, { "name": "undef and signal handlers", "lines": 3 }, { "name": "strictures and dereferencing in defined()", "lines": 13 }, { "name": "\"(?p{})\" has been removed", "lines": 3 }, { "name": "Pseudo-hashes have been removed", "lines": 3 }, { "name": "Removal of the bytecode compiler and of perlcc", "lines": 9 }, { "name": "Removal of the JPL", "lines": 2 }, { "name": "Recursive inheritance detected earlier", "lines": 6 }, { "name": "warnings::enabled and warnings::warnif changed to favor users of modules", "lines": 17 } ] }, { "name": "Modules and Pragmata", "lines": 1, "subsections": [ { "name": "Upgrading individual core modules", "lines": 5 }, { "name": "Pragmata Changes", "lines": 41 }, { "name": "New modules", "lines": 81 }, { "name": "Selected Changes to Core Modules", "lines": 18 } ] }, { "name": "Utility Changes", "lines": 60, "subsections": [] }, { "name": "New Documentation", "lines": 24, "subsections": [] }, { "name": "Performance Enhancements", "lines": 1, "subsections": [ { "name": "In-place sorting", "lines": 6 }, { "name": "Lexical array access", "lines": 3 }, { "name": "XS-assisted SWASHGET", "lines": 3 }, { "name": "Constant subroutines", "lines": 10 }, { "name": "\"PERL_DONT_CREATE_GVSV\"", "lines": 4 }, { "name": "Weak references are cheaper", "lines": 4 }, { "name": "sort() enhancements", "lines": 3 }, { "name": "Memory optimisations", "lines": 3 }, { "name": "UTF-8 cache optimisation", "lines": 2 }, { "name": "Sloppy stat on Windows", "lines": 5 }, { "name": "Regular expressions optimisations", "lines": 30 } ] }, { "name": "Installation and Configuration Improvements", "lines": 1, "subsections": [ { "name": "Configuration improvements", "lines": 32 }, { "name": "Compilation improvements", "lines": 34 }, { "name": "Installation improvements", "lines": 3 }, { "name": "New Or Improved Platforms", "lines": 17 } ] }, { "name": "Selected Bug Fixes", "lines": 103, "subsections": [] }, { "name": "New or Changed Diagnostics", "lines": 56, "subsections": [] }, { "name": "Changed Internals", "lines": 15, "subsections": [ { "name": "Elimination of SVt_PVBM", "lines": 4 }, { "name": "New type SVt_BIND", "lines": 3 }, { "name": "Removal of CPP symbols", "lines": 5 }, { "name": "Less space is used by ops", "lines": 4 }, { "name": "New parser", "lines": 5 }, { "name": "Use of \"const\"", "lines": 5 }, { "name": "Mathoms", "lines": 4 }, { "name": "\"AvFLAGS\" has been removed", "lines": 5 }, { "name": "$^H and %^H", "lines": 3 }, { "name": "B:: modules inheritance changed", "lines": 3 }, { "name": "Anonymous hash and array constructors", "lines": 4 } ] }, { "name": "Known Problems", "lines": 6, "subsections": [ { "name": "UTF-8 problems", "lines": 4 } ] }, { "name": "Platform Specific Problems", "lines": 4, "subsections": [] }, { "name": "Reporting Bugs", "lines": 9, "subsections": [] }, { "name": "SEE ALSO", "lines": 10, "subsections": [] } ], "sections": { "NAME": { "content": "perl5100delta - what is new for perl 5.10.0\n", "subsections": [] }, "DESCRIPTION": { "content": "This document describes the differences between the 5.8.8 release and the 5.10.0 release.\n\nMany of the bug fixes in 5.10.0 were already seen in the 5.8.X maintenance releases; they are\nnot duplicated here and are documented in the set of man pages named perl58[1-8]?delta.\n", "subsections": [] }, "Core Enhancements": { "content": "", "subsections": [ { "name": "The \"feature\" pragma", "content": "The \"feature\" pragma is used to enable new syntax that would break Perl's backwards-\ncompatibility with older releases of the language. It's a lexical pragma, like \"strict\" or\n\"warnings\".\n\nCurrently the following new features are available: \"switch\" (adds a switch statement), \"say\"\n(adds a \"say\" built-in function), and \"state\" (adds a \"state\" keyword for declaring \"static\"\nvariables). Those features are described in their own sections of this document.\n\nThe \"feature\" pragma is also implicitly loaded when you require a minimal perl version (with\nthe \"use VERSION\" construct) greater than, or equal to, 5.9.5. See feature for details.\n" }, { "name": "New -E command-line switch", "content": "" }, { "name": "-E -e", "content": "\":5.10\"\").\n", "flag": "-e" }, { "name": "Defined-or operator", "content": "A new operator \"//\" (defined-or) has been implemented. The following expression:\n\n$a // $b\n\nis merely equivalent to\n\ndefined $a ? $a : $b\n\nand the statement\n\n$c //= $d;\n\ncan now be used instead of\n\n$c = $d unless defined $c;\n\nThe \"//\" operator has the same precedence and associativity as \"||\". Special care has been\ntaken to ensure that this operator Do What You Mean while not breaking old code, but some\nedge cases involving the empty regular expression may now parse differently. See perlop for\ndetails.\n" }, { "name": "Switch and Smart Match operator", "content": "Perl 5 now has a switch statement. It's available when use feature 'switch' is in effect.\nThis feature introduces three new keywords, \"given\", \"when\", and \"default\":\n\ngiven ($foo) {\nwhen (/^abc/) { $abc = 1; }\nwhen (/^def/) { $def = 1; }\nwhen (/^xyz/) { $xyz = 1; }\ndefault { $nothing = 1; }\n}\n\nA more complete description of how Perl matches the switch variable against the \"when\"\nconditions is given in \"Switch statements\" in perlsyn.\n\nThis kind of match is called smart match, and it's also possible to use it outside of switch\nstatements, via the new \"~~\" operator. See \"Smart matching in detail\" in perlsyn.\n\nThis feature was contributed by Robin Houston.\n" }, { "name": "Regular expressions", "content": "Recursive Patterns\nIt is now possible to write recursive patterns without using the \"(??{})\" construct. This\nnew way is more efficient, and in many cases easier to read.\n\nEach capturing parenthesis can now be treated as an independent pattern that can be\nentered by using the \"(?PARNO)\" syntax (\"PARNO\" standing for \"parenthesis number\"). For\nexample, the following pattern will match nested balanced angle brackets:\n\n/\n^ # start of line\n( # start capture buffer 1\n< # match an opening angle bracket\n(?: # match one of:\n(?> # don't backtrack over the inside of this group\n[^<>]+ # one or more non angle brackets\n) # end non backtracking group\n| # ... or ...\n(?1) # recurse to bracket 1 and try it again\n)* # 0 or more times.\n> # match a closing angle bracket\n) # end capture buffer one\n$ # end of line\n/x\n\nPCRE users should note that Perl's recursive regex feature allows backtracking into a\nrecursed pattern, whereas in PCRE the recursion is atomic or \"possessive\" in nature. As\nin the example above, you can add (?>) to control this selectively. (Yves Orton)\n\nNamed Capture Buffers\nIt is now possible to name capturing parenthesis in a pattern and refer to the captured\ncontents by name. The naming syntax is \"(?....)\". It's possible to backreference\nto a named buffer with the \"\\k\" syntax. In code, the new magical hashes \"%+\" and\n\"%-\" can be used to access the contents of the capture buffers.\n\nThus, to replace all doubled chars with a single copy, one could write\n\ns/(?.)\\k/$+{letter}/g\n\nOnly buffers with defined contents will be \"visible\" in the \"%+\" hash, so it's possible\nto do something like\n\nforeach my $name (keys %+) {\nprint \"content of buffer '$name' is $+{$name}\\n\";\n}\n\nThe \"%-\" hash is a bit more complete, since it will contain array refs holding values\nfrom all capture buffers similarly named, if there should be many of them.\n\n\"%+\" and \"%-\" are implemented as tied hashes through the new module\n\"Tie::Hash::NamedCapture\".\n\nUsers exposed to the .NET regex engine will find that the perl implementation differs in\nthat the numerical ordering of the buffers is sequential, and not \"unnamed first, then\nnamed\". Thus in the pattern\n\n/(A)(?B)(C)(?D)/\n\n$1 will be 'A', $2 will be 'B', $3 will be 'C' and $4 will be 'D' and not $1 is 'A', $2\nis 'C' and $3 is 'B' and $4 is 'D' that a .NET programmer would expect. This is\nconsidered a feature. :-) (Yves Orton)\n\nPossessive Quantifiers\nPerl now supports the \"possessive quantifier\" syntax of the \"atomic match\" pattern.\nBasically a possessive quantifier matches as much as it can and never gives any back.\nThus it can be used to control backtracking. The syntax is similar to non-greedy\nmatching, except instead of using a '?' as the modifier the '+' is used. Thus \"?+\", \"*+\",\n\"++\", \"{min,max}+\" are now legal quantifiers. (Yves Orton)\n\nBacktracking control verbs\nThe regex engine now supports a number of special-purpose backtrack control verbs:\n(*THEN), (*PRUNE), (*MARK), (*SKIP), (*COMMIT), (*FAIL) and (*ACCEPT). See perlre for\ntheir descriptions. (Yves Orton)\n\nRelative backreferences\nA new syntax \"\\g{N}\" or \"\\gN\" where \"N\" is a decimal integer allows a safer form of back-\nreference notation as well as allowing relative backreferences. This should make it\neasier to generate and embed patterns that contain backreferences. See \"Capture buffers\"\nin perlre. (Yves Orton)\n\n\"\\K\" escape\nThe functionality of Jeff Pinyan's module Regexp::Keep has been added to the core. In\nregular expressions you can now use the special escape \"\\K\" as a way to do something like\nfloating length positive lookbehind. It is also useful in substitutions like:\n\ns/(foo)bar/$1/g\n\nthat can now be converted to\n\ns/foo\\Kbar//g\n\nwhich is much more efficient. (Yves Orton)\n\nVertical and horizontal whitespace, and linebreak\nRegular expressions now recognize the \"\\v\" and \"\\h\" escapes that match vertical and\nhorizontal whitespace, respectively. \"\\V\" and \"\\H\" logically match their complements.\n\n\"\\R\" matches a generic linebreak, that is, vertical whitespace, plus the multi-character\nsequence \"\\x0D\\x0A\".\n\nOptional pre-match and post-match captures with the /p flag\nThere is a new flag \"/p\" for regular expressions. Using this makes the engine preserve a\ncopy of the part of the matched string before the matching substring to the new special\nvariable \"${^PREMATCH}\", the part after the matching substring to \"${^POSTMATCH}\", and\nthe matched substring itself to \"${^MATCH}\".\n\nPerl is still able to store these substrings to the special variables \"$`\", \"$'\", $&, but\nusing these variables anywhere in the program adds a penalty to all regular expression\nmatches, whereas if you use the \"/p\" flag and the new special variables instead, you pay\nonly for the regular expressions where the flag is used.\n\nFor more detail on the new variables, see perlvar; for the use of the regular expression\nflag, see perlop and perlre.\n" }, { "name": "say()", "content": "say() is a new built-in, only available when \"use feature 'say'\" is in effect, that is\nsimilar to print(), but that implicitly appends a newline to the printed string. See \"say\" in\nperlfunc. (Robin Houston)\n" }, { "name": "Lexical $_", "content": "The default variable $ can now be lexicalized, by declaring it like any other lexical\nvariable, with a simple\n\nmy $;\n\nThe operations that default on $ will use the lexically-scoped version of $ when it exists,\ninstead of the global $.\n\nIn a \"map\" or a \"grep\" block, if $ was previously my'ed, then the $ inside the block is\nlexical as well (and scoped to the block).\n\nIn a scope where $ has been lexicalized, you can still have access to the global version of\n$ by using $::, or, more simply, by overriding the lexical declaration with \"our $\".\n(Rafael Garcia-Suarez)\n" }, { "name": "The \"_\" prototype", "content": "A new prototype character has been added. \"\" is equivalent to \"$\" but defaults to $ if the\ncorresponding argument isn't supplied (both \"$\" and \"\" denote a scalar). Due to the optional\nnature of the argument, you can only use it at the end of a prototype, or before a semicolon.\n\nThis has a small incompatible consequence: the prototype() function has been adjusted to\nreturn \"\" for some built-ins in appropriate cases (for example, prototype('CORE::rmdir')).\n(Rafael Garcia-Suarez)\n" }, { "name": "UNITCHECK blocks", "content": "\"UNITCHECK\", a new special code block has been introduced, in addition to \"BEGIN\", \"CHECK\",\n\"INIT\" and \"END\".\n\n\"CHECK\" and \"INIT\" blocks, while useful for some specialized purposes, are always executed at\nthe transition between the compilation and the execution of the main program, and thus are\nuseless whenever code is loaded at runtime. On the other hand, \"UNITCHECK\" blocks are\nexecuted just after the unit which defined them has been compiled. See perlmod for more\ninformation. (Alex Gough)\n" }, { "name": "New Pragma, \"mro\"", "content": "A new pragma, \"mro\" (for Method Resolution Order) has been added. It permits to switch, on a\nper-class basis, the algorithm that perl uses to find inherited methods in case of a multiple\ninheritance hierarchy. The default MRO hasn't changed (DFS, for Depth First Search). Another\nMRO is available: the C3 algorithm. See mro for more information. (Brandon Black)\n\nNote that, due to changes in the implementation of class hierarchy search, code that used to\nundef the *ISA glob will most probably break. Anyway, undef'ing *ISA had the side-effect of\nremoving the magic on the @ISA array and should not have been done in the first place. Also,\nthe cache *::ISA::CACHE:: no longer exists; to force reset the @ISA cache, you now need to\nuse the \"mro\" API, or more simply to assign to @ISA (e.g. with \"@ISA = @ISA\").\n" }, { "name": "readdir() may return a \"short filename\" on Windows", "content": "The readdir() function may return a \"short filename\" when the long filename contains\ncharacters outside the ANSI codepage. Similarly Cwd::cwd() may return a short directory\nname, and glob() may return short names as well. On the NTFS file system these short names\ncan always be represented in the ANSI codepage. This will not be true for all other file\nsystem drivers; e.g. the FAT filesystem stores short filenames in the OEM codepage, so some\nfiles on FAT volumes remain inaccessible through the ANSI APIs.\n\nSimilarly, $^X, @INC, and $ENV{PATH} are preprocessed at startup to make sure all paths are\nvalid in the ANSI codepage (if possible).\n\nThe Win32::GetLongPathName() function now returns the UTF-8 encoded correct long file name\ninstead of using replacement characters to force the name into the ANSI codepage. The new\nWin32::GetANSIPathName() function can be used to turn a long pathname into a short one only\nif the long one cannot be represented in the ANSI codepage.\n\nMany other functions in the \"Win32\" module have been improved to accept UTF-8 encoded\narguments. Please see Win32 for details.\n" }, { "name": "readpipe() is now overridable", "content": "The built-in function readpipe() is now overridable. Overriding it permits also to override\nits operator counterpart, \"qx//\" (a.k.a. ``). Moreover, it now defaults to $ if no argument\nis provided. (Rafael Garcia-Suarez)\n" }, { "name": "Default argument for readline()", "content": "readline() now defaults to *ARGV if no argument is provided. (Rafael Garcia-Suarez)\n" }, { "name": "state() variables", "content": "A new class of variables has been introduced. State variables are similar to \"my\" variables,\nbut are declared with the \"state\" keyword in place of \"my\". They're visible only in their\nlexical scope, but their value is persistent: unlike \"my\" variables, they're not undefined at\nscope entry, but retain their previous value. (Rafael Garcia-Suarez, Nicholas Clark)\n\nTo use state variables, one needs to enable them by using\n\nuse feature 'state';\n\nor by using the \"-E\" command-line switch in one-liners. See \"Persistent Private Variables\"\nin perlsub.\n" }, { "name": "Stacked filetest operators", "content": "As a new form of syntactic sugar, it's now possible to stack up filetest operators. You can\nnow write \"-f -w -x $file\" in a row to mean \"-x $file && -w && -f \". See \"-X\" in perlfunc.\n" }, { "name": "UNIVERSAL::DOES()", "content": "The \"UNIVERSAL\" class has a new method, DOES(). It has been added to solve semantic problems\nwith the isa() method. isa() checks for inheritance, while DOES() has been designed to be\noverridden when module authors use other types of relations between classes (in addition to\ninheritance). (chromatic)\n\nSee \"$obj->DOES( ROLE )\" in UNIVERSAL.\n" }, { "name": "Formats", "content": "Formats were improved in several ways. A new field, \"^*\", can be used for variable-width,\none-line-at-a-time text. Null characters are now handled correctly in picture lines. Using\n\"@#\" and \"~~\" together will now produce a compile-time error, as those format fields are\nincompatible. perlform has been improved, and miscellaneous bugs fixed.\n" }, { "name": "Byte-order modifiers for pack() and unpack()", "content": "There are two new byte-order modifiers, \">\" (big-endian) and \"<\" (little-endian), that can be\nappended to most pack() and unpack() template characters and groups to force a certain byte-\norder for that type or group. See \"pack\" in perlfunc and perlpacktut for details.\n" }, { "name": "\"no VERSION\"", "content": "You can now use \"no\" followed by a version number to specify that you want to use a version\nof perl older than the specified one.\n" }, { "name": "\"chdir\", \"chmod\" and \"chown\" on filehandles", "content": "\"chdir\", \"chmod\" and \"chown\" can now work on filehandles as well as filenames, if the system\nsupports respectively \"fchdir\", \"fchmod\" and \"fchown\", thanks to a patch provided by Gisle\nAas.\n" }, { "name": "OS groups", "content": "$( and $) now return groups in the order where the OS returns them, thanks to Gisle Aas. This\nwasn't previously the case.\n" }, { "name": "Recursive sort subs", "content": "You can now use recursive subroutines with sort(), thanks to Robin Houston.\n" }, { "name": "Exceptions in constant folding", "content": "The constant folding routine is now wrapped in an exception handler, and if folding throws an\nexception (such as attempting to evaluate 0/0), perl now retains the current optree, rather\nthan aborting the whole program. Without this change, programs would not compile if they had\nexpressions that happened to generate exceptions, even though those expressions were in code\nthat could never be reached at runtime. (Nicholas Clark, Dave Mitchell)\n" }, { "name": "Source filters in @INC", "content": "It's possible to enhance the mechanism of subroutine hooks in @INC by adding a source filter\non top of the filehandle opened and returned by the hook. This feature was planned a long\ntime ago, but wasn't quite working until now. See \"require\" in perlfunc for details.\n(Nicholas Clark)\n" }, { "name": "New internal variables", "content": "\"${^REDEBUGFLAGS}\"\nThis variable controls what debug flags are in effect for the regular expression engine\nwhen running under \"use re \"debug\"\". See re for details.\n\n\"${^CHILDERRORNATIVE}\"\nThis variable gives the native status returned by the last pipe close, backtick command,\nsuccessful call to wait() or waitpid(), or from the system() operator. See perlvar for\ndetails. (Contributed by Gisle Aas.)\n\n\"${^RETRIEMAXBUF}\"\nSee \"Trie optimisation of literal string alternations\".\n\n\"${^WIN32SLOPPYSTAT}\"\nSee \"Sloppy stat on Windows\".\n" }, { "name": "Miscellaneous", "content": "unpack() now defaults to unpacking the $ variable.\n\nmkdir() without arguments now defaults to $.\n\nThe internal dump output has been improved, so that non-printable characters such as newline\nand backspace are output in \"\\x\" notation, rather than octal.\n\nThe -C option can no longer be used on the \"#!\" line. It wasn't working there anyway, since\nthe standard streams are already set up at this point in the execution of the perl\ninterpreter. You can use binmode() instead to get the desired behaviour.\n" }, { "name": "UCD 5.0.0", "content": "The copy of the Unicode Character Database included in Perl 5 has been updated to version\n5.0.0.\n\nMAD\nMAD, which stands for Miscellaneous Attribute Decoration, is a still-in-development work\nleading to a Perl 5 to Perl 6 converter. To enable it, it's necessary to pass the argument\n\"-Dmad\" to Configure. The obtained perl isn't binary compatible with a regular perl 5.10, and\nhas space and speed penalties; moreover not all regression tests still pass with it. (Larry\nWall, Nicholas Clark)\n" }, { "name": "kill() on Windows", "content": "On Windows platforms, \"kill(-9, $pid)\" now kills a process tree. (On Unix, this delivers the\nsignal to all processes in the same process group.)\n" } ] }, "Incompatible Changes": { "content": "", "subsections": [ { "name": "Packing and UTF-8 strings", "content": "The semantics of pack() and unpack() regarding UTF-8-encoded data has been changed.\nProcessing is now by default character per character instead of byte per byte on the\nunderlying encoding. Notably, code that used things like \"pack(\"a*\", $string)\" to see through\nthe encoding of string will now simply get back the original $string. Packed strings can also\nget upgraded during processing when you store upgraded characters. You can get the old\nbehaviour by using \"use bytes\".\n\nTo be consistent with pack(), the \"C0\" in unpack() templates indicates that the data is to be\nprocessed in character mode, i.e. character by character; on the contrary, \"U0\" in unpack()\nindicates UTF-8 mode, where the packed string is processed in its UTF-8-encoded Unicode form\non a byte by byte basis. This is reversed with regard to perl 5.8.X, but now consistent\nbetween pack() and unpack().\n\nMoreover, \"C0\" and \"U0\" can also be used in pack() templates to specify respectively\ncharacter and byte modes.\n\n\"C0\" and \"U0\" in the middle of a pack or unpack format now switch to the specified encoding\nmode, honoring parens grouping. Previously, parens were ignored.\n\nAlso, there is a new pack() character format, \"W\", which is intended to replace the old \"C\".\n\"C\" is kept for unsigned chars coded as bytes in the strings internal representation. \"W\"\nrepresents unsigned (logical) character values, which can be greater than 255. It is\ntherefore more robust when dealing with potentially UTF-8-encoded data (as \"C\" will wrap\nvalues outside the range 0..255, and not respect the string encoding).\n\nIn practice, that means that pack formats are now encoding-neutral, except \"C\".\n\nFor consistency, \"A\" in unpack() format now trims all Unicode whitespace from the end of the\nstring. Before perl 5.9.2, it used to strip only the classical ASCII space characters.\n" }, { "name": "Byte/character count feature in unpack()", "content": "A new unpack() template character, \".\", returns the number of bytes or characters (depending\non the selected encoding mode, see above) read so far.\n\nThe $* and $# variables have been removed\n$*, which was deprecated in favor of the \"/s\" and \"/m\" regexp modifiers, has been removed.\n\nThe deprecated $# variable (output format for numbers) has been removed.\n\nTwo new severe warnings, \"$#/$* is no longer supported\", have been added.\n" }, { "name": "substr() lvalues are no longer fixed-length", "content": "The lvalues returned by the three argument form of substr() used to be a \"fixed length\nwindow\" on the original string. In some cases this could cause surprising action at distance\nor other undefined behaviour. Now the length of the window adjusts itself to the length of\nthe string assigned to it.\n" }, { "name": "Parsing of \"-f _\"", "content": "The identifier \"\" is now forced to be a bareword after a filetest operator. This solves a\nnumber of misparsing issues when a global \"\" subroutine is defined.\n" }, { "name": "\":unique\"", "content": "The \":unique\" attribute has been made a no-op, since its current implementation was\nfundamentally flawed and not threadsafe.\n" }, { "name": "Effect of pragmas in eval", "content": "The compile-time value of the \"%^H\" hint variable can now propagate into eval(\"\")uated code.\nThis makes it more useful to implement lexical pragmas.\n\nAs a side-effect of this, the overloaded-ness of constants now propagates into eval(\"\").\n" }, { "name": "chdir FOO", "content": "A bareword argument to chdir() is now recognized as a file handle. Earlier releases\ninterpreted the bareword as a directory name. (Gisle Aas)\n" }, { "name": "Handling of .pmc files", "content": "An old feature of perl was that before \"require\" or \"use\" look for a file with a .pm\nextension, they will first look for a similar filename with a .pmc extension. If this file is\nfound, it will be loaded in place of any potentially existing file ending in a .pm extension.\n\nPreviously, .pmc files were loaded only if more recent than the matching .pm file. Starting\nwith 5.9.4, they'll be always loaded if they exist.\n" }, { "name": "$^V is now a \"version\" object instead of a v-string", "content": "$^V can still be used with the %vd format in printf, but any character-level operations will\nnow access the string representation of the \"version\" object and not the ordinals of a\nv-string. Expressions like \"substr($^V, 0, 2)\" or \"split //, $^V\" no longer work and must be\nrewritten.\n" }, { "name": "@- and @+ in patterns", "content": "The special arrays \"@-\" and \"@+\" are no longer interpolated in regular expressions. (Sadahiro\nTomoyuki)\n" }, { "name": "$AUTOLOAD can now be tainted", "content": "If you call a subroutine by a tainted name, and if it defers to an AUTOLOAD function, then\n$AUTOLOAD will be (correctly) tainted. (Rick Delaney)\n" }, { "name": "Tainting and printf", "content": "When perl is run under taint mode, printf() and sprintf() will now reject any tainted format\nargument. (Rafael Garcia-Suarez)\n" }, { "name": "undef and signal handlers", "content": "Undefining or deleting a signal handler via \"undef $SIG{FOO}\" is now equivalent to setting it\nto 'DEFAULT'. (Rafael Garcia-Suarez)\n" }, { "name": "strictures and dereferencing in defined()", "content": "\"use strict 'refs'\" was ignoring taking a hard reference in an argument to defined(), as in :\n\nuse strict 'refs';\nmy $x = 'foo';\nif (defined $$x) {...}\n\nThis now correctly produces the run-time error \"Can't use string as a SCALAR ref while\n\"strict refs\" in use\".\n\n\"defined @$foo\" and \"defined %$bar\" are now also subject to strict 'refs' (that is, $foo and\n$bar shall be proper references there.) (defined(@foo) and defined(%bar) are discouraged\nconstructs anyway.) (Nicholas Clark)\n" }, { "name": "\"(?p{})\" has been removed", "content": "The regular expression construct \"(?p{})\", which was deprecated in perl 5.8, has been\nremoved. Use \"(??{})\" instead. (Rafael Garcia-Suarez)\n" }, { "name": "Pseudo-hashes have been removed", "content": "Support for pseudo-hashes has been removed from Perl 5.9. (The \"fields\" pragma remains here,\nbut uses an alternate implementation.)\n" }, { "name": "Removal of the bytecode compiler and of perlcc", "content": "\"perlcc\", the byteloader and the supporting modules (B::C, B::CC, B::Bytecode, etc.) are no\nlonger distributed with the perl sources. Those experimental tools have never worked\nreliably, and, due to the lack of volunteers to keep them in line with the perl interpreter\ndevelopments, it was decided to remove them instead of shipping a broken version of those.\nThe last version of those modules can be found with perl 5.9.4.\n\nHowever the B compiler framework stays supported in the perl core, as with the more useful\nmodules it has permitted (among others, B::Deparse and B::Concise).\n" }, { "name": "Removal of the JPL", "content": "The JPL (Java-Perl Lingo) has been removed from the perl sources tarball.\n" }, { "name": "Recursive inheritance detected earlier", "content": "Perl will now immediately throw an exception if you modify any package's @ISA in such a way\nthat it would cause recursive inheritance.\n\nPreviously, the exception would not occur until Perl attempted to make use of the recursive\ninheritance while resolving a method or doing a \"$foo->isa($bar)\" lookup.\n" }, { "name": "warnings::enabled and warnings::warnif changed to favor users of modules", "content": "The behaviour in 5.10.x favors the person using the module; The behaviour in 5.8.x favors the\nmodule writer;\n\nAssume the following code:\n\nmain calls Foo::Bar::baz()\nFoo::Bar inherits from Foo::Base\nFoo::Bar::baz() calls Foo::Base::bazbaz()\nFoo::Base::bazbaz() calls: warnings::warnif('substr', 'some warning\nmessage');\n\nOn 5.8.x, the code warns when Foo::Bar contains \"use warnings;\" It does not matter if\nFoo::Base or main have warnings enabled to disable the warning one has to modify Foo::Bar.\n\nOn 5.10.0 and newer, the code warns when main contains \"use warnings;\" It does not matter if\nFoo::Base or Foo::Bar have warnings enabled to disable the warning one has to modify main.\n" } ] }, "Modules and Pragmata": { "content": "", "subsections": [ { "name": "Upgrading individual core modules", "content": "Even more core modules are now also available separately through the CPAN. If you wish to\nupdate one of these modules, you don't need to wait for a new perl release. From within the\ncpan shell, running the 'r' command will report on modules with upgrades available. See\n\"perldoc CPAN\" for more information.\n" }, { "name": "Pragmata Changes", "content": "\"feature\"\nThe new pragma \"feature\" is used to enable new features that might break old code. See\n\"The \"feature\" pragma\" above.\n\n\"mro\"\nThis new pragma enables to change the algorithm used to resolve inherited methods. See\n\"New Pragma, \"mro\"\" above.\n\nScoping of the \"sort\" pragma\nThe \"sort\" pragma is now lexically scoped. Its effect used to be global.\n\nScoping of \"bignum\", \"bigint\", \"bigrat\"\nThe three numeric pragmas \"bignum\", \"bigint\" and \"bigrat\" are now lexically scoped.\n(Tels)\n\n\"base\"\nThe \"base\" pragma now warns if a class tries to inherit from itself. (Curtis \"Ovid\" Poe)\n\n\"strict\" and \"warnings\"\n\"strict\" and \"warnings\" will now complain loudly if they are loaded via incorrect casing\n(as in \"use Strict;\"). (Johan Vromans)\n\n\"version\"\nThe \"version\" module provides support for version objects.\n\n\"warnings\"\nThe \"warnings\" pragma doesn't load \"Carp\" anymore. That means that code that used \"Carp\"\nroutines without having loaded it at compile time might need to be adjusted; typically,\nthe following (faulty) code won't work anymore, and will require parentheses to be added\nafter the function name:\n\nuse warnings;\nrequire Carp;\nCarp::confess 'argh';\n\n\"less\"\n\"less\" now does something useful (or at least it tries to). In fact, it has been turned\ninto a lexical pragma. So, in your modules, you can now test whether your users have\nrequested to use less CPU, or less memory, less magic, or maybe even less fat. See less\nfor more. (Joshua ben Jore)\n" }, { "name": "New modules", "content": "• \"encoding::warnings\", by Audrey Tang, is a module to emit warnings whenever an ASCII\ncharacter string containing high-bit bytes is implicitly converted into UTF-8. It's a\nlexical pragma since Perl 5.9.4; on older perls, its effect is global.\n\n• \"Module::CoreList\", by Richard Clamp, is a small handy module that tells you what\nversions of core modules ship with any versions of Perl 5. It comes with a command-line\nfrontend, \"corelist\".\n\n• \"Math::BigInt::FastCalc\" is an XS-enabled, and thus faster, version of\n\"Math::BigInt::Calc\".\n\n• \"Compress::Zlib\" is an interface to the zlib compression library. It comes with a bundled\nversion of zlib, so having a working zlib is not a prerequisite to install it. It's used\nby \"Archive::Tar\" (see below).\n\n• \"IO::Zlib\" is an \"IO::\"-style interface to \"Compress::Zlib\".\n\n• \"Archive::Tar\" is a module to manipulate \"tar\" archives.\n\n• \"Digest::SHA\" is a module used to calculate many types of SHA digests, has been included\nfor SHA support in the CPAN module.\n\n• \"ExtUtils::CBuilder\" and \"ExtUtils::ParseXS\" have been added.\n\n• \"Hash::Util::FieldHash\", by Anno Siegel, has been added. This module provides support for\nfield hashes: hashes that maintain an association of a reference with a value, in a\nthread-safe garbage-collected way. Such hashes are useful to implement inside-out\nobjects.\n\n• \"Module::Build\", by Ken Williams, has been added. It's an alternative to\n\"ExtUtils::MakeMaker\" to build and install perl modules.\n\n• \"Module::Load\", by Jos Boumans, has been added. It provides a single interface to load\nPerl modules and .pl files.\n\n• \"Module::Loaded\", by Jos Boumans, has been added. It's used to mark modules as loaded or\nunloaded.\n\n• \"Package::Constants\", by Jos Boumans, has been added. It's a simple helper to list all\nconstants declared in a given package.\n\n• \"Win32API::File\", by Tye McQueen, has been added (for Windows builds). This module\nprovides low-level access to Win32 system API calls for files/dirs.\n\n• \"Locale::Maketext::Simple\", needed by CPANPLUS, is a simple wrapper around\n\"Locale::Maketext::Lexicon\". Note that \"Locale::Maketext::Lexicon\" isn't included in the\nperl core; the behaviour of \"Locale::Maketext::Simple\" gracefully degrades when the later\nisn't present.\n\n• \"Params::Check\" implements a generic input parsing/checking mechanism. It is used by\nCPANPLUS.\n\n• \"Term::UI\" simplifies the task to ask questions at a terminal prompt.\n\n• \"Object::Accessor\" provides an interface to create per-object accessors.\n\n• \"Module::Pluggable\" is a simple framework to create modules that accept pluggable sub-\nmodules.\n\n• \"Module::Load::Conditional\" provides simple ways to query and possibly load installed\nmodules.\n\n• \"Time::Piece\" provides an object oriented interface to time functions, overriding the\nbuilt-ins localtime() and gmtime().\n\n• \"IPC::Cmd\" helps to find and run external commands, possibly interactively.\n\n• \"File::Fetch\" provide a simple generic file fetching mechanism.\n\n• \"Log::Message\" and \"Log::Message::Simple\" are used by the log facility of \"CPANPLUS\".\n\n• \"Archive::Extract\" is a generic archive extraction mechanism for .tar (plain, gzipped or\nbzipped) or .zip files.\n\n• \"CPANPLUS\" provides an API and a command-line tool to access the CPAN mirrors.\n\n• \"Pod::Escapes\" provides utilities that are useful in decoding Pod E<...> sequences.\n\n• \"Pod::Simple\" is now the backend for several of the Pod-related modules included with\nPerl.\n" }, { "name": "Selected Changes to Core Modules", "content": "\"Attribute::Handlers\"\n\"Attribute::Handlers\" can now report the caller's file and line number. (David Feldman)\n\nAll interpreted attributes are now passed as array references. (Damian Conway)\n\n\"B::Lint\"\n\"B::Lint\" is now based on \"Module::Pluggable\", and so can be extended with plugins.\n(Joshua ben Jore)\n\n\"B\" It's now possible to access the lexical pragma hints (\"%^H\") by using the method\nB::COP::hintshash(). It returns a \"B::RHE\" object, which in turn can be used to get a\nhash reference via the method B::RHE::HASH(). (Joshua ben Jore)\n\n\"Thread\"\nAs the old 5005thread threading model has been removed, in favor of the ithreads scheme,\nthe \"Thread\" module is now a compatibility wrapper, to be used in old code only. It has\nbeen removed from the default list of dynamic extensions.\n" } ] }, "Utility Changes": { "content": "perl -d\nThe Perl debugger can now save all debugger commands for sourcing later; notably, it can\nnow emulate stepping backwards, by restarting and rerunning all bar the last command from\na saved command history.\n\nIt can also display the parent inheritance tree of a given class, with the \"i\" command.\n\nptar\n\"ptar\" is a pure perl implementation of \"tar\" that comes with \"Archive::Tar\".\n\nptardiff\n\"ptardiff\" is a small utility used to generate a diff between the contents of a tar\narchive and a directory tree. Like \"ptar\", it comes with \"Archive::Tar\".\n\nshasum\n\"shasum\" is a command-line utility, used to print or to check SHA digests. It comes with\nthe new \"Digest::SHA\" module.\n\ncorelist\nThe \"corelist\" utility is now installed with perl (see \"New modules\" above).\n\nh2ph and h2xs\n\"h2ph\" and \"h2xs\" have been made more robust with regard to \"modern\" C code.\n\n\"h2xs\" implements a new option \"--use-xsloader\" to force use of \"XSLoader\" even in\nbackwards compatible modules.\n\nThe handling of authors' names that had apostrophes has been fixed.\n\nAny enums with negative values are now skipped.\n\nperlivp\n\"perlivp\" no longer checks for *.ph files by default. Use the new \"-a\" option to run all\ntests.\n\nfind2perl\n\"find2perl\" now assumes \"-print\" as a default action. Previously, it needed to be\nspecified explicitly.\n\nSeveral bugs have been fixed in \"find2perl\", regarding \"-exec\" and \"-eval\". Also the\noptions \"-path\", \"-ipath\" and \"-iname\" have been added.\n\nconfigdata\n\"configdata\" is a new utility that comes with \"Module::Build\". It provides a command-\nline interface to the configuration of Perl modules that use Module::Build's framework of\nconfigurability (that is, *::ConfigData modules that contain local configuration\ninformation for their parent modules.)\n\ncpanp\n\"cpanp\", the CPANPLUS shell, has been added. (\"cpanp-run-perl\", a helper for CPANPLUS\noperation, has been added too, but isn't intended for direct use).\n\ncpan2dist\n\"cpan2dist\" is a new utility that comes with CPANPLUS. It's a tool to create\ndistributions (or packages) from CPAN modules.\n\npod2html\nThe output of \"pod2html\" has been enhanced to be more customizable via CSS. Some\nformatting problems were also corrected. (Jari Aalto)\n", "subsections": [] }, "New Documentation": { "content": "The perlpragma manpage documents how to write one's own lexical pragmas in pure Perl\n(something that is possible starting with 5.9.4).\n\nThe new perlglossary manpage is a glossary of terms used in the Perl documentation, technical\nand otherwise, kindly provided by O'Reilly Media, Inc.\n\nThe perlreguts manpage, courtesy of Yves Orton, describes internals of the Perl regular\nexpression engine.\n\nThe perlreapi manpage describes the interface to the perl interpreter used to write pluggable\nregular expression engines (by Ævar Arnfjörð Bjarmason).\n\nThe perlunitut manpage is a tutorial for programming with Unicode and string encodings in\nPerl, courtesy of Juerd Waalboer.\n\nA new manual page, perlunifaq (the Perl Unicode FAQ), has been added (Juerd Waalboer).\n\nThe perlcommunity manpage gives a description of the Perl community on the Internet and in\nreal life. (Edgar \"Trizor\" Bering)\n\nThe CORE manual page documents the \"CORE::\" namespace. (Tels)\n\nThe long-existing feature of \"/(?{...})/\" regexps setting $ and pos() is now documented.\n", "subsections": [] }, "Performance Enhancements": { "content": "", "subsections": [ { "name": "In-place sorting", "content": "Sorting arrays in place (\"@a = sort @a\") is now optimized to avoid making a temporary copy of\nthe array.\n\nLikewise, \"reverse sort ...\" is now optimized to sort in reverse, avoiding the generation of\na temporary intermediate list.\n" }, { "name": "Lexical array access", "content": "Access to elements of lexical arrays via a numeric constant between 0 and 255 is now faster.\n(This used to be only the case for global arrays.)\n" }, { "name": "XS-assisted SWASHGET", "content": "Some pure-perl code that perl was using to retrieve Unicode properties and transliteration\nmappings has been reimplemented in XS.\n" }, { "name": "Constant subroutines", "content": "The interpreter internals now support a far more memory efficient form of inlineable\nconstants. Storing a reference to a constant value in a symbol table is equivalent to a full\ntypeglob referencing a constant subroutine, but using about 400 bytes less memory. This proxy\nconstant subroutine is automatically upgraded to a real typeglob with subroutine if\nnecessary. The approach taken is analogous to the existing space optimisation for subroutine\nstub declarations, which are stored as plain scalars in place of the full typeglob.\n\nSeveral of the core modules have been converted to use this feature for their system\ndependent constants - as a result \"use POSIX;\" now takes about 200K less memory.\n" }, { "name": "\"PERL_DONT_CREATE_GVSV\"", "content": "The new compilation flag \"PERLDONTCREATEGVSV\", introduced as an option in perl 5.8.8, is\nturned on by default in perl 5.9.3. It prevents perl from creating an empty scalar with every\nnew typeglob. See perl589delta for details.\n" }, { "name": "Weak references are cheaper", "content": "Weak reference creation is now O(1) rather than O(n), courtesy of Nicholas Clark. Weak\nreference deletion remains O(n), but if deletion only happens at program exit, it may be\nskipped completely.\n" }, { "name": "sort() enhancements", "content": "Salvador Fandiño provided improvements to reduce the memory usage of \"sort\" and to speed up\nsome cases.\n" }, { "name": "Memory optimisations", "content": "Several internal data structures (typeglobs, GVs, CVs, formats) have been restructured to use\nless memory. (Nicholas Clark)\n" }, { "name": "UTF-8 cache optimisation", "content": "The UTF-8 caching code is now more efficient, and used more often. (Nicholas Clark)\n" }, { "name": "Sloppy stat on Windows", "content": "On Windows, perl's stat() function normally opens the file to determine the link count and\nupdate attributes that may have been changed through hard links. Setting\n${^WIN32SLOPPYSTAT} to a true value speeds up stat() by not performing this operation. (Jan\nDubois)\n" }, { "name": "Regular expressions optimisations", "content": "Engine de-recursivised\nThe regular expression engine is no longer recursive, meaning that patterns that used to\noverflow the stack will either die with useful explanations, or run to completion, which,\nsince they were able to blow the stack before, will likely take a very long time to\nhappen. If you were experiencing the occasional stack overflow (or segfault) and upgrade\nto discover that now perl apparently hangs instead, look for a degenerate regex. (Dave\nMitchell)\n\nSingle char char-classes treated as literals\nClasses of a single character are now treated the same as if the character had been used\nas a literal, meaning that code that uses char-classes as an escaping mechanism will see\na speedup. (Yves Orton)\n\nTrie optimisation of literal string alternations\nAlternations, where possible, are optimised into more efficient matching structures.\nString literal alternations are merged into a trie and are matched simultaneously. This\nmeans that instead of O(N) time for matching N alternations at a given point, the new\ncode performs in O(1) time. A new special variable, ${^RETRIEMAXBUF}, has been added\nto fine-tune this optimization. (Yves Orton)\n\nNote: Much code exists that works around perl's historic poor performance on\nalternations. Often the tricks used to do so will disable the new optimisations.\nHopefully the utility modules used for this purpose will be educated about these new\noptimisations.\n\nAho-Corasick start-point optimisation\nWhen a pattern starts with a trie-able alternation and there aren't better optimisations\navailable, the regex engine will use Aho-Corasick matching to find the start point. (Yves\nOrton)\n" } ] }, "Installation and Configuration Improvements": { "content": "", "subsections": [ { "name": "Configuration improvements", "content": "\"-Dusesitecustomize\"\nRun-time customization of @INC can be enabled by passing the \"-Dusesitecustomize\" flag to\nConfigure. When enabled, this will make perl run $sitelibexp/sitecustomize.pl before\nanything else. This script can then be set up to add additional entries to @INC.\n\nRelocatable installations\nThere is now Configure support for creating a relocatable perl tree. If you Configure\nwith \"-Duserelocatableinc\", then the paths in @INC (and everything else in %Config) can\nbe optionally located via the path of the perl executable.\n\nThat means that, if the string \".../\" is found at the start of any path, it's substituted\nwith the directory of $^X. So, the relocation can be configured on a per-directory basis,\nalthough the default with \"-Duserelocatableinc\" is that everything is relocated. The\ninitial install is done to the original configured prefix.\n\nstrlcat() and strlcpy()\nThe configuration process now detects whether strlcat() and strlcpy() are available.\nWhen they are not available, perl's own version is used (from Russ Allbery's public\ndomain implementation). Various places in the perl interpreter now use them. (Steve\nPeters)\n\n\"dpseudofork\" and \"dprintfformatnull\"\nA new configuration variable, available as $Config{dpseudofork} in the Config module,\nhas been added, to distinguish real fork() support from fake pseudofork used on Windows\nplatforms.\n\nA new configuration variable, \"dprintfformatnull\", has been added, to see if printf-\nlike formats are allowed to be NULL.\n\nConfigure help\n\"Configure -h\" has been extended with the most commonly used options.\n" }, { "name": "Compilation improvements", "content": "Parallel build\nParallel makes should work properly now, although there may still be problems if \"make\ntest\" is instructed to run in parallel.\n\nBorland's compilers support\nBuilding with Borland's compilers on Win32 should work more smoothly. In particular Steve\nHay has worked to side step many warnings emitted by their compilers and at least one C\ncompiler internal error.\n\nStatic build on Windows\nPerl extensions on Windows now can be statically built into the Perl DLL.\n\nAlso, it's now possible to build a \"perl-static.exe\" that doesn't depend on the Perl DLL\non Win32. See the Win32 makefiles for details. (Vadim Konovalov)\n\nppport.h files\nAll ppport.h files in the XS modules bundled with perl are now autogenerated at build\ntime. (Marcus Holland-Moritz)\n\nC++ compatibility\nEfforts have been made to make perl and the core XS modules compilable with various C++\ncompilers (although the situation is not perfect with some of the compilers on some of\nthe platforms tested.)\n\nSupport for Microsoft 64-bit compiler\nSupport for building perl with Microsoft's 64-bit compiler has been improved.\n(ActiveState)\n\nVisual C++\nPerl can now be compiled with Microsoft Visual C++ 2005 (and 2008 Beta 2).\n\nWin32 builds\nAll win32 builds (MS-Win, WinCE) have been merged and cleaned up.\n" }, { "name": "Installation improvements", "content": "Module auxiliary files\nREADME files and changelogs for CPAN modules bundled with perl are no longer installed.\n" }, { "name": "New Or Improved Platforms", "content": "Perl has been reported to work on Symbian OS. See perlsymbian for more information.\n\nMany improvements have been made towards making Perl work correctly on z/OS.\n\nPerl has been reported to work on DragonFlyBSD and MidnightBSD.\n\nPerl has also been reported to work on NexentaOS ( http://www.gnusolaris.org/ ).\n\nThe VMS port has been improved. See perlvms.\n\nSupport for Cray XT4 Catamount/Qk has been added. See hints/catamount.sh in the source code\ndistribution for more information.\n\nVendor patches have been merged for RedHat and Gentoo.\n\nDynaLoader::dlunloadfile() now works on Windows.\n" } ] }, "Selected Bug Fixes": { "content": "strictures in regexp-eval blocks\n\"strict\" wasn't in effect in regexp-eval blocks (\"/(?{...})/\").\n\nCalling CORE::require()\nCORE::require() and CORE::do() were always parsed as require() and do() when they were\noverridden. This is now fixed.\n\nSubscripts of slices\nYou can now use a non-arrowed form for chained subscripts after a list slice, like in:\n\n({foo => \"bar\"})[0]{foo}\n\nThis used to be a syntax error; a \"->\" was required.\n\n\"no warnings 'category'\" works correctly with -w\nPreviously when running with warnings enabled globally via \"-w\", selective disabling of\nspecific warning categories would actually turn off all warnings. This is now fixed; now\n\"no warnings 'io';\" will only turn off warnings in the \"io\" class. Previously it would\nerroneously turn off all warnings.\n\nthreads improvements\nSeveral memory leaks in ithreads were closed. Also, ithreads were made less memory-\nintensive.\n\n\"threads\" is now a dual-life module, also available on CPAN. It has been expanded in many\nways. A kill() method is available for thread signalling. One can get thread status, or\nthe list of running or joinable threads.\n\nA new \"threads->exit()\" method is used to exit from the application (this is the default\nfor the main thread) or from the current thread only (this is the default for all other\nthreads). On the other hand, the exit() built-in now always causes the whole application\nto terminate. (Jerry D. Hedden)\n\nchr() and negative values\nchr() on a negative value now gives \"\\x{FFFD}\", the Unicode replacement character, unless\nwhen the \"bytes\" pragma is in effect, where the low eight bits of the value are used.\n\nPERL5SHELL and tainting\nOn Windows, the PERL5SHELL environment variable is now checked for taintedness. (Rafael\nGarcia-Suarez)\n\nUsing *FILE{IO}\nstat() and \"-X\" filetests now treat *FILE{IO} filehandles like *FILE filehandles. (Steve\nPeters)\n\nOverloading and reblessing\nOverloading now works when references are reblessed into another class. Internally, this\nhas been implemented by moving the flag for \"overloading\" from the reference to the\nreferent, which logically is where it should always have been. (Nicholas Clark)\n\nOverloading and UTF-8\nA few bugs related to UTF-8 handling with objects that have stringification overloaded\nhave been fixed. (Nicholas Clark)\n\neval memory leaks fixed\nTraditionally, \"eval 'syntax error'\" has leaked badly. Many (but not all) of these leaks\nhave now been eliminated or reduced. (Dave Mitchell)\n\nRandom device on Windows\nIn previous versions, perl would read the file /dev/urandom if it existed when seeding\nits random number generator. That file is unlikely to exist on Windows, and if it did\nwould probably not contain appropriate data, so perl no longer tries to read it on\nWindows. (Alex Davies)\n\nPERLIODEBUG\nThe \"PERLIODEBUG\" environment variable no longer has any effect for setuid scripts and\nfor scripts run with -T.\n\nMoreover, with a thread-enabled perl, using \"PERLIODEBUG\" could lead to an internal\nbuffer overflow. This has been fixed.\n\nPerlIO::scalar and read-only scalars\nPerlIO::scalar will now prevent writing to read-only scalars. Moreover, seek() is now\nsupported with PerlIO::scalar-based filehandles, the underlying string being zero-filled\nas needed. (Rafael, Jarkko Hietaniemi)\n\nstudy() and UTF-8\nstudy() never worked for UTF-8 strings, but could lead to false results. It's now a no-\nop on UTF-8 data. (Yves Orton)\n\nCritical signals\nThe signals SIGILL, SIGBUS and SIGSEGV are now always delivered in an \"unsafe\" manner\n(contrary to other signals, that are deferred until the perl interpreter reaches a\nreasonably stable state; see \"Deferred Signals (Safe Signals)\" in perlipc). (Rafael)\n\n@INC-hook fix\nWhen a module or a file is loaded through an @INC-hook, and when this hook has set a\nfilename entry in %INC, FILE is now set for this module accordingly to the contents\nof that %INC entry. (Rafael)\n\n\"-t\" switch fix\nThe \"-w\" and \"-t\" switches can now be used together without messing up which categories\nof warnings are activated. (Rafael)\n\nDuping UTF-8 filehandles\nDuping a filehandle which has the \":utf8\" PerlIO layer set will now properly carry that\nlayer on the duped filehandle. (Rafael)\n\nLocalisation of hash elements\nLocalizing a hash element whose key was given as a variable didn't work correctly if the\nvariable was changed while the local() was in effect (as in \"local $h{$x}; ++$x\"). (Bo\nLindbergh)\n", "subsections": [] }, "New or Changed Diagnostics": { "content": "Use of uninitialized value\nPerl will now try to tell you the name of the variable (if any) that was undefined.\n\nDeprecated use of my() in false conditional\nA new deprecation warning, Deprecated use of my() in false conditional, has been added,\nto warn against the use of the dubious and deprecated construct\n\nmy $x if 0;\n\nSee perldiag. Use \"state\" variables instead.\n\n!=~ should be !~\nA new warning, \"!=~ should be !~\", is emitted to prevent this misspelling of the non-\nmatching operator.\n\nNewline in left-justified string\nThe warning Newline in left-justified string has been removed.\n\nToo late for \"-T\" option\nThe error Too late for \"-T\" option has been reformulated to be more descriptive.\n\n\"%s\" variable %s masks earlier declaration\nThis warning is now emitted in more consistent cases; in short, when one of the\ndeclarations involved is a \"my\" variable:\n\nmy $x; my $x; # warns\nmy $x; our $x; # warns\nour $x; my $x; # warns\n\nOn the other hand, the following:\n\nour $x; our $x;\n\nnow gives a \"\"our\" variable %s redeclared\" warning.\n\nreaddir()/closedir()/etc. attempted on invalid dirhandle\nThese new warnings are now emitted when a dirhandle is used but is either closed or not\nreally a dirhandle.\n\nOpening dirhandle/filehandle %s also as a file/directory\nTwo deprecation warnings have been added: (Rafael)\n\nOpening dirhandle %s also as a file\nOpening filehandle %s also as a directory\n\nUse of -P is deprecated\nPerl's command-line switch \"-P\" is now deprecated.\n\nv-string in use/require is non-portable\nPerl will warn you against potential backwards compatibility problems with the \"use\nVERSION\" syntax.\n\nperl -V\n\"perl -V\" has several improvements, making it more useable from shell scripts to get the\nvalue of configuration variables. See perlrun for details.\n", "subsections": [] }, "Changed Internals": { "content": "In general, the source code of perl has been refactored, tidied up, and optimized in many\nplaces. Also, memory management and allocation has been improved in several points.\n\nWhen compiling the perl core with gcc, as many gcc warning flags are turned on as is possible\non the platform. (This quest for cleanliness doesn't extend to XS code because we cannot\nguarantee the tidiness of code we didn't write.) Similar strictness flags have been added or\ntightened for various other C compilers.\n\nReordering of SVt* constants\nThe relative ordering of constants that define the various types of \"SV\" have changed; in\nparticular, \"SVtPVGV\" has been moved before \"SVtPVLV\", \"SVtPVAV\", \"SVtPVHV\" and\n\"SVtPVCV\". This is unlikely to make any difference unless you have code that explicitly\nmakes assumptions about that ordering. (The inheritance hierarchy of \"B::*\" objects has been\nchanged to reflect this.)\n", "subsections": [ { "name": "Elimination of SVt_PVBM", "content": "Related to this, the internal type \"SVtPVBM\" has been removed. This dedicated type of \"SV\"\nwas used by the \"index\" operator and parts of the regexp engine to facilitate fast Boyer-\nMoore matches. Its use internally has been replaced by \"SV\"s of type \"SVtPVGV\".\n" }, { "name": "New type SVt_BIND", "content": "A new type \"SVtBIND\" has been added, in readiness for the project to implement Perl 6 on 5.\nThere deliberately is no implementation yet, and they cannot yet be created or destroyed.\n" }, { "name": "Removal of CPP symbols", "content": "The C preprocessor symbols \"PERLPMAPIVERSION\" and \"PERLXSAPIVERSION\", which were supposed\nto give the version number of the oldest perl binary-compatible (resp. source-compatible)\nwith the present one, were not used, and sometimes had misleading values. They have been\nremoved.\n" }, { "name": "Less space is used by ops", "content": "The \"BASEOP\" structure now uses less space. The \"opseq\" field has been removed and replaced\nby a single bit bit-field \"opopt\". \"optype\" is now 9 bits long. (Consequently, the \"B::OP\"\nclass doesn't provide an \"seq\" method anymore.)\n" }, { "name": "New parser", "content": "perl's parser is now generated by bison (it used to be generated by byacc.) As a result, it\nseems to be a bit more robust.\n\nAlso, Dave Mitchell improved the lexer debugging output under \"-DT\".\n" }, { "name": "Use of \"const\"", "content": "Andy Lester supplied many improvements to determine which function parameters and local\nvariables could actually be declared \"const\" to the C compiler. Steve Peters provided new\n*set macros and reworked the core to use these rather than assigning to macros in LVALUE\ncontext.\n" }, { "name": "Mathoms", "content": "A new file, mathoms.c, has been added. It contains functions that are no longer used in the\nperl core, but that remain available for binary or source compatibility reasons. However,\nthose functions will not be compiled in if you add \"-DNOMATHOMS\" in the compiler flags.\n" }, { "name": "\"AvFLAGS\" has been removed", "content": "The \"AvFLAGS\" macro has been removed.\n\n\"av*\" changes\nThe \"av*()\" functions, used to manipulate arrays, no longer accept null \"AV*\" parameters.\n" }, { "name": "$^H and %^H", "content": "The implementation of the special variables $^H and %^H has changed, to allow implementing\nlexical pragmas in pure Perl.\n" }, { "name": "B:: modules inheritance changed", "content": "The inheritance hierarchy of \"B::\" modules has changed; \"B::NV\" now inherits from \"B::SV\" (it\nused to inherit from \"B::IV\").\n" }, { "name": "Anonymous hash and array constructors", "content": "The anonymous hash and array constructors now take 1 op in the optree instead of 3, now that\nppanonhash and ppanonlist return a reference to a hash/array when the op is flagged with\nOPfSPECIAL. (Nicholas Clark)\n" } ] }, "Known Problems": { "content": "There's still a remaining problem in the implementation of the lexical $: it doesn't work\ninside \"/(?{...})/\" blocks. (See the TODO test in t/op/mydef.t.)\n\nStacked filetest operators won't work when the \"filetest\" pragma is in effect, because they\nrely on the stat() buffer \"\" being populated, and filetest bypasses stat().\n", "subsections": [ { "name": "UTF-8 problems", "content": "The handling of Unicode still is unclean in several places, where it's dependent on whether a\nstring is internally flagged as UTF-8. This will be made more consistent in perl 5.12, but\nthat won't be possible without a certain amount of backwards incompatibility.\n" } ] }, "Platform Specific Problems": { "content": "When compiled with g++ and thread support on Linux, it's reported that the $! stops working\ncorrectly. This is related to the fact that the glibc provides two strerrorr(3)\nimplementation, and perl selects the wrong one.\n", "subsections": [] }, "Reporting Bugs": { "content": "If you find what you think is a bug, you might check the articles recently posted to the\ncomp.lang.perl.misc newsgroup and the perl bug database at http://rt.perl.org/rt3/ . There\nmay also be information at http://www.perl.org/ , the Perl Home Page.\n\nIf you believe you have an unreported bug, please run the perlbug program included with your\nrelease. Be sure to trim your bug down to a tiny but sufficient test case. Your bug report,\nalong with the output of \"perl -V\", will be sent off to perlbug@perl.org to be analysed by\nthe Perl porting team.\n", "subsections": [] }, "SEE ALSO": { "content": "The Changes file and the perl590delta to perl595delta man pages for exhaustive details on\nwhat changed.\n\nThe INSTALL file for how to build Perl.\n\nThe README file for general stuff.\n\nThe Artistic and Copying files for copyright information.\n\nperl v5.38.2 2026-06-12 PERL5100DELTA(1)", "subsections": [] } } } }