3 perldelta - what is new for perl v5.8.0
7 This document describes differences between the 5.6.0 release and
10 Many of the bug fixes in 5.8.0 were already seen in the 5.6.1
11 maintenance release since the two releases were kept closely
12 coordinated (while 5.8.0 was still called 5.7.something).
14 Changes that were integrated into the 5.6.1 release are marked C<[561]>.
15 Many of these changes have been further developed since 5.6.1 was released,
16 those are marked C<[561+]>.
18 You can see the list of changes in the 5.6.1 release (both from the
19 5.005_03 release and the 5.6.0 release) by reading L<perl561delta>.
21 =head1 Highlights In 5.8.0
27 Better Unicode support
31 New Thread Implementation
39 Better Numeric Accuracy
47 More Extensive Regression Testing
51 =head1 Incompatible Changes
53 =head2 Binary Incompatibility
55 B<Perl 5.8 is not binary compatible with earlier releases of Perl.>
57 B<You have to recompile your XS modules.>
59 (Pure Perl modules should continue to work.)
61 The major reason for the discontinuity is the new IO architecture
62 called PerlIO. PerlIO is the default configuration because without
63 it many new features of Perl 5.8 cannot be used. In other words:
64 you just have to recompile your modules containing XS code, sorry
67 In future releases of Perl, non-PerlIO aware XS modules may become
68 completely unsupported. This shouldn't be too difficult for module
69 authors, however: PerlIO has been designed as a drop-in replacement
70 (at the source code level) for the stdio interface.
72 Depending on your platform, there are also other reasons why
73 we decided to break binary compatibility, please read on.
75 =head2 64-bit platforms and malloc
77 If your pointers are 64 bits wide, the Perl malloc is no longer being
78 used because it does not work well with 8-byte pointers. Also,
79 usually the system mallocs on such platforms are much better optimized
80 for such large memory models than the Perl malloc. Some memory-hungry
81 Perl applications like the PDL don't work well with Perl's malloc.
82 Finally, other applications than Perl (such as mod_perl) tend to prefer
83 the system malloc. Such platforms include Alpha and 64-bit HPPA,
86 =head2 AIX Dynaloading
88 The AIX dynaloading now uses in AIX releases 4.3 and newer the native
89 dlopen interface of AIX instead of the old emulated interface. This
90 change will probably break backward compatibility with compiled
91 modules. The change was made to make Perl more compliant with other
92 applications like mod_perl which are using the AIX native interface.
94 =head2 Attributes for C<my> variables now handled at run-time.
96 The C<my EXPR : ATTRS> syntax now applies variable attributes at
97 run-time. (Subroutine and C<our> variables still get attributes applied
98 at compile-time.) See L<attributes> for additional details. In particular,
99 however, this allows variable attributes to be useful for C<tie> interfaces,
100 which was a deficiency of earlier releases. Note that the new semantics
101 doesn't work with the Attribute::Handlers module (as of version 0.76).
103 =head2 Socket Extension Dynamic in VMS
105 The Socket extension is now dynamically loaded instead of being
106 statically built in. This may or may not be a problem with ancient
107 TCP/IP stacks of VMS: we do not know since we weren't able to test
108 Perl in such configurations.
110 =head2 IEEE-format Floating Point Default on OpenVMS Alpha
112 Perl now uses IEEE format (T_FLOAT) as the default internal floating
113 point format on OpenVMS Alpha, potentially breaking binary compatibility
114 with external libraries or existing data. G_FLOAT is still available as
115 a configuration option. The default on VAX (D_FLOAT) has not changed.
117 =head2 New Unicode Properties
119 Unicode I<scripts> are now supported. Scripts are similar to (and superior
120 to) Unicode I<blocks>. The difference between scripts and blocks is that
121 scripts are the glyphs used by a language or a group of languages, while
122 the blocks are more artificial groupings of (mostly) 256 characters based
123 on the Unicode numbering.
125 In general, scripts are more inclusive, but not universally so. For
126 example, while the script C<Latin> includes all the Latin characters and
127 their various diacritic-adorned versions, it does not include the various
128 punctuation or digits (since they are not solely C<Latin>).
130 A number of other properties are now supported, including C<\p{L&}>,
131 C<\p{Any}> C<\p{Assigned}>, C<\p{Unassigned}>, C<\p{Blank}> [561] and
132 C<\p{SpacePerl}> [561] (along with their C<\P{...}> versions, of course).
133 See L<perlunicode> for details, and more additions.
135 The C<In> or C<Is> prefix to names used with the C<\p{...}> and C<\P{...}>
136 are now almost always optional. The only exception is that a C<In> prefix
137 is required to signify a Unicode block when a block name conflicts with a
138 script name. For example, C<\p{Tibetan}> refers to the script, while
139 C<\p{InTibetan}> refers to the block. When there is no name conflict, you
140 can omit the C<In> from the block name (e.g. C<\p{BraillePatterns}>), but
141 to be safe, it's probably best to always use the C<In>).
143 =head2 REF(...) Instead Of SCALAR(...)
145 A reference to a reference now stringifies as "REF(0x81485ec)" instead
146 of "SCALAR(0x81485ec)" in order to be more consistent with the return
149 =head2 pack/unpack D/F recycled
151 The undocumented pack/unpack template letters D/F have been recycled
152 for better use: now they stand for long double (if supported by the
153 platform) and NV (Perl internal floating point type). (They used
154 to be aliases for d/f, but you never knew that.)
162 The semantics of bless(REF, REF) were unclear and until someone proves
163 it to make some sense, it is forbidden.
167 The obsolete chat2 library that should never have been allowed
168 to escape the laboratory has been decommissioned.
172 The builtin dump() function has probably outlived most of its
173 usefulness. The core-dumping functionality will remain in future
174 available as an explicit call to C<CORE::dump()>, but in future
175 releases the behaviour of an unqualified C<dump()> call may change.
179 The very dusty examples in the eg/ directory have been removed.
180 Suggestions for new shiny examples welcome but the main issue is that
181 the examples need to be documented, tested and (most importantly)
186 The (bogus) escape sequences \8 and \9 now give an optional warning
187 ("Unrecognized escape passed through"). There is no need to \-escape
192 The list of filenames from glob() (or <...>) is now by default sorted
193 alphabetically to be csh-compliant (which is what happened before
194 in most UNIX platforms). (bsd_glob() does still sort platform
195 natively, ASCII or EBCDIC, unless GLOB_ALPHASORT is specified.) [561]
199 Spurious syntax errors generated in certain situations, when glob()
200 caused File::Glob to be loaded for the first time, have been fixed. [561]
204 Although "you shouldn't do that", it was possible to write code that
205 depends on Perl's hashed key order (Data::Dumper does this). The new
206 algorithm "One-at-a-Time" produces a different hashed key order.
207 More details are in L</"Performance Enhancements">.
211 lstat(FILEHANDLE) now gives a warning because the operation makes no sense.
212 In future releases this may become a fatal error.
216 The C<package;> syntax (C<package> without an argument) has been
217 deprecated. Its semantics were never that clear and its
218 implementation even less so. If you have used that feature to
219 disallow all but fully qualified variables, C<use strict;> instead.
223 The unimplemented POSIX regex features [[.cc.]] and [[=c=]] are still
224 recognised but now cause fatal errors. The previous behaviour of
225 ignoring them by default and warning if requested was unacceptable
226 since it, in a way, falsely promised that the features could be used.
230 In future releases, non-PerlIO aware XS modules may become completely
231 unsupported. Since PerlIO is a drop-in replacement for stdio at the
232 source code level, this shouldn't be that drastic a change.
236 The current user-visible implementation of pseudo-hashes (the weird
237 use of the first array element) is deprecated starting from Perl 5.8.0
238 and will be removed in Perl 5.10.0, and the feature will be
239 implemented differently. Not only is the current interface rather
240 ugly, but the current implementation slows down normal array and hash
241 use quite noticeably. The C<fields> pragma interface will remain
242 available. The I<restricted hashes> interface is expected to
243 be the replacement interface (see L<Hash::Util>).
247 The syntaxes C<< @a->[...] >> and C<< %h->{...} >> have now been deprecated.
251 After years of trying, suidperl is considered to be too complex to
252 ever be considered truly secure. The suidperl functionality is likely
253 to be removed in a future release.
257 The 5.005 threads model (module C<Thread>) is deprecated and expected
258 to be removed in Perl 5.10. Multithreaded code should be migrated to
259 the new ithreads model (see L<threads>, L<threads::shared> and
264 The long deprecated uppercase aliases for the string comparison
265 operators (EQ, NE, LT, LE, GE, GT) have now been removed.
269 The tr///C and tr///U features have been removed and will not return;
270 the interface was a mistake. Sorry about that. For similar
271 functionality, see pack('U0', ...) and pack('C0', ...). [561]
275 Earlier Perls treated "sub foo (@bar)" as equivalent to "sub foo (@)".
276 The prototypes are now checked better at compile-time for invalid
277 syntax. An optional warning is generated ("Illegal character in
278 prototype...") but this may be upgraded to a fatal error in a future
283 The C<exec LIST> and C<system LIST> operations will produce fatal
284 errors on tainted data in some future release.
288 The existing behaviour when localising tied arrays and hashes is wrong,
289 and will be changed in a future release, so do not rely on the existing
290 behaviour. See L<"Localising Tied Arrays and Hashes Is Broken">.
294 =head1 Core Enhancements
296 =head2 PerlIO is Now The Default
302 IO is now by default done via PerlIO rather than system's "stdio".
303 PerlIO allows "layers" to be "pushed" onto a file handle to alter the
304 handle's behaviour. Layers can be specified at open time via 3-arg
307 open($fh,'>:crlf :utf8', $path) || ...
309 or on already opened handles via extended C<binmode>:
311 binmode($fh,':encoding(iso-8859-7)');
313 The built-in layers are: unix (low level read/write), stdio (as in
314 previous Perls), perlio (re-implementation of stdio buffering in a
315 portable manner), crlf (does CRLF <=> "\n" translation as on Win32,
316 but available on any platform). A mmap layer may be available if
317 platform supports it (mostly UNIXes).
319 Layers to be applied by default may be specified via the 'open' pragma.
321 See L</"Installation and Configuration Improvements"> for the effects
322 of PerlIO on your architecture name.
326 File handles can be marked as accepting Perl's internal encoding of Unicode
327 (UTF-8 or UTF-EBCDIC depending on platform) by a pseudo layer ":utf8" :
329 open($fh,">:utf8","Uni.txt");
331 Note for EBCDIC users: the pseudo layer ":utf8" is erroneously named
332 for you since it's not UTF-8 what you will be getting but instead
333 UTF-EBCDIC. See L<perlunicode>, L<utf8>, and
334 http://www.unicode.org/unicode/reports/tr16/ for more information.
335 In future releases this naming may change. See L<perluniintro>
336 for more information about UTF-8.
340 If your environment variables (LC_ALL, LC_CTYPE, LANG, LANGUAGE) look
341 like you want to use UTF-8 (any of the the variables match C</utf-?8/i>),
342 your STDIN, STDOUT, STDERR handles and the default open discipline
343 (see L<open>) are marked as UTF-8. (This feature, like other new
344 features that combine Unicode and I/O, work only if you are using
345 PerlIO, but that's is the default.)
347 Note that after this Perl really does assume that everything is UTF-8:
348 for example if some input handle is not, Perl will probably very soon
349 complain about the input data like this "Malformed UTF-8 ..." since
350 any old eight-bit data is not legal UTF-8.
352 Note for code authors: if you want to enable your users to use UTF-8
353 as their default encoding but in your code still have eight-bit I/O streams
354 (such as images or zip files), you need to explicitly open() or binmode()
355 with C<:bytes> (see L<perlfunc/open> and L<perlfunc/binmode>), or you
356 can just use C<binmode(FH)> (nice for pre-5.8.0 backward compatibility).
360 File handles can translate character encodings from/to Perl's internal
361 Unicode form on read/write via the ":encoding()" layer.
365 File handles can be opened to "in memory" files held in Perl scalars via:
367 open($fh,'>', \$variable) || ...
371 Anonymous temporary files are available without need to
372 'use FileHandle' or other module via
374 open($fh,"+>", undef) || ...
376 That is a literal undef, not an undefined value.
380 The list form of C<open> is now implemented for pipes (at least on UNIX):
382 open($fh,"-|", 'cat', '/etc/motd')
384 creates a pipe, and runs the equivalent of exec('cat', '/etc/motd') in
389 If your locale environment variables (LANGUAGE, LC_ALL, LC_CTYPE, LANG)
390 contain the strings 'UTF-8' or 'UTF8' (case-insensitive matching),
391 the default encoding of your STDIN, STDOUT, and STDERR, and of
392 B<any subsequent file open>, is UTF-8.
396 =head2 Restricted Hashes
398 A restricted hash is restricted to a certain set of keys, no keys
399 outside the set can be added. Also individual keys can be restricted
400 so that the key cannot be deleted and the value cannot be changed.
401 No new syntax is involved: the Hash::Util module is the interface.
405 Perl used to be fragile in that signals arriving at inopportune moments
406 could corrupt Perl's internal state. Now Perl postpones handling of
407 signals until it's safe (between opcodes).
409 This change may have surprising side effects because signals no longer
410 interrupt Perl instantly. Perl will now first finish whatever it was
411 doing, like finishing an internal operation (like sort()) or an
412 external operation (like an I/O operation), and only then look at any
413 arrived signals (and before starting the next operation). No more corrupt
414 internal state since the current operation is always finished first,
415 but the signal may take more time to get heard. Note that breaking
416 out from potentially blocking operations should still work, though.
418 =head2 Unicode Overhaul
420 Unicode in general should be now much more usable than in Perl 5.6.0
421 (or even in 5.6.1). Unicode can be used in hash keys, Unicode in
422 regular expressions should work now, Unicode in tr/// should work now,
423 Unicode in I/O should work now. See L<perluniintro> for introduction
424 and L<perlunicode> for details.
430 The Unicode Character Database coming with Perl has been upgraded
431 to Unicode 3.2.0. For more information, see http://www.unicode.org/ .
432 [561+] (5.6.1 has UCD 3.0.1.)
436 For developers interested in enhancing Perl's Unicode capabilities:
437 almost all the UCD files are included with the Perl distribution in
438 the F<lib/unicore> subdirectory. The most notable omission, for space
439 considerations, is the Unihan database.
443 The properties \p{Blank} and \p{SpacePerl} have been added. "Blank" is like
444 C isblank(), that is, it contains only "horizontal whitespace" (the space
445 character is, the newline isn't), and the "SpacePerl" is the Unicode
446 equivalent of C<\s> (\p{Space} isn't, since that includes the vertical
447 tabulator character, whereas C<\s> doesn't.)
449 See "New Unicode Properties" earlier in this document for additional
450 information on changes with Unicode properties.
454 =head2 Understanding of Numbers
456 In general a lot of fixing has happened in the area of Perl's
457 understanding of numbers, both integer and floating point. Since in
458 many systems the standard number parsing functions like C<strtoul()>
459 and C<atof()> seem to have bugs, Perl tries to work around their
460 deficiencies. This results hopefully in more accurate numbers.
462 Perl now tries internally to use integer values in numeric conversions
463 and basic arithmetics (+ - * /) if the arguments are integers, and
464 tries also to keep the results stored internally as integers.
465 This change leads to often slightly faster and always less lossy
466 arithmetics. (Previously Perl always preferred floating point numbers
469 =head2 Arrays now always interpolate into double-quoted strings [561]
471 In double-quoted strings, arrays now interpolate, no matter what. The
472 behavior in earlier versions of perl 5 was that arrays would interpolate
473 into strings if the array had been mentioned before the string was
474 compiled, and otherwise Perl would raise a fatal compile-time error.
475 In versions 5.000 through 5.003, the error was
477 Literal @example now requires backslash
479 In versions 5.004_01 through 5.6.0, the error was
481 In string, @example now must be written as \@example
483 The idea here was to get people into the habit of writing
484 C<"fred\@example.com"> when they wanted a literal C<@> sign, just as
485 they have always written C<"Give me back my \$5"> when they wanted a
488 Starting with 5.6.1, when Perl now sees an C<@> sign in a
489 double-quoted string, it I<always> attempts to interpolate an array,
490 regardless of whether or not the array has been used or declared
491 already. The fatal error has been downgraded to an optional warning:
493 Possible unintended interpolation of @example in string
495 This warns you that C<"fred@example.com"> is going to turn into
496 C<fred.com> if you don't backslash the C<@>.
497 See http://www.plover.com/~mjd/perl/at-error.html for more details
498 about the history here.
500 =head2 Miscellaneous Changes
506 AUTOLOAD is now lvaluable, meaning that you can add the :lvalue attribute
507 to AUTOLOAD subroutines and you can assign to the AUTOLOAD return value.
511 The $Config{byteorder} (and corresponding BYTEORDER in config.h) was
512 previously wrong in platforms if sizeof(long) was 4, but sizeof(IV)
513 was 8. The byteorder was only sizeof(long) bytes long (1234 or 4321),
514 but now it is correctly sizeof(IV) bytes long, (12345678 or 87654321).
515 (This problem didn't affect Windows platforms.)
517 Also, $Config{byteorder} is now computed dynamically--this is more
518 robust with "fat binaries" where an executable image contains binaries
519 for more than one binary platform, and when cross-compiling.
523 C<perl -d:Module=arg,arg,arg> now works (previously one couldn't pass
524 in multiple arguments.)
528 C<do> followed by a bareword now ensures that this bareword isn't
529 a keyword (to avoid a bug where C<do q(foo.pl)> tried to call a
530 subroutine called C<q>). This means that for example instead of
531 C<do format()> you must write C<do &format()>.
535 The builtin dump() now gives an optional warning
536 C<dump() better written as CORE::dump()>,
537 meaning that by default C<dump(...)> is resolved as the builtin
538 dump() which dumps core and aborts, not as (possibly) user-defined
539 C<sub dump>. To call the latter, qualify the call as C<&dump(...)>.
540 (The whole dump() feature is to considered deprecated, and possibly
541 removed/changed in future releases.)
545 chomp() and chop() are now overridable. Note, however, that their
546 prototype (as given by C<prototype("CORE::chomp")> is undefined,
547 because it cannot be expressed and therefore one cannot really write
548 replacements to override these builtins.
552 END blocks are now run even if you exit/die in a BEGIN block.
553 Internally, the execution of END blocks is now controlled by
554 PL_exit_flags & PERL_EXIT_DESTRUCT_END. This enables the new
555 behaviour for Perl embedders. This will default in 5.10. See
560 Formats now support zero-padded decimal fields.
564 Lvalue subroutines can now return C<undef> in list context. However,
565 the lvalue subroutine feature still remains experimental. [561+]
569 A lost warning "Can't declare ... dereference in my" has been
570 restored (Perl had it earlier but it became lost in later releases.)
574 A new special regular expression variable has been introduced:
575 C<$^N>, which contains the most-recently closed group (submatch).
579 C<no Module;> does not produce an error even if Module does not have an
580 unimport() method. This parallels the behavior of C<use> vis-a-vis
585 The numerical comparison operators return C<undef> if either operand
586 is a NaN. Previously the behaviour was unspecified.
590 C<our> can now have an experimental optional attribute C<unique> that
591 affects how global variables are shared among multiple interpreters,
596 The following builtin functions are now overridable: each(), keys(),
597 pop(), push(), shift(), splice(), unshift(). [561]
601 C<pack() / unpack()> can now group template letters with C<()> and then
602 apply repetition/count modifiers on the groups.
606 C<pack() / unpack()> can now process the Perl internal numeric types:
607 IVs, UVs, NVs-- and also long doubles, if supported by the platform.
608 The template letters are C<j>, C<J>, C<F>, and C<D>.
612 C<pack('U0a*', ...)> can now be used to force a string to UTF8.
616 my __PACKAGE__ $obj now works. [561]
620 POSIX::sleep() now returns the number of I<unslept> seconds
621 (as the POSIX standard says), as opposed to CORE::sleep() which
622 returns the number of slept seconds.
626 The printf() and sprintf() now support parameter reordering using the
627 C<%\d+\$> and C<*\d+\$> syntaxes. For example
629 print "%2\$s %1\$s\n", "foo", "bar";
631 will print "bar foo\n". This feature helps in writing
632 internationalised software, and in general when the order
633 of the parameters can vary.
637 The (\&) prototype now works properly. [561]
641 prototype(\[$@%&]) is now available to implicitly create references
642 (useful for example if you want to emulate the tie() interface).
646 A new command-line option, C<-t> is available. It is the
647 little brother of C<-T>: instead of dying on taint violations,
648 lexical warnings are given. B<This is only meant as a temporary
649 debugging aid while securing the code of old legacy applications.
650 This is not a substitute for -T.>
654 In other taint news, the C<exec LIST> and C<system LIST> have now been
655 considered too risky (think C<exec @ARGV>: it can start any program
656 with any arguments), and now the said forms cause a warning under
657 lexical warnings. You should carefully launder the arguments to
658 guarantee their validity. In future releases of Perl the forms will
659 become fatal errors so consider starting laundering now.
663 Tied hash interfaces are now required to have the EXISTS and DELETE
664 methods (either own or inherited).
668 If tr/// is just counting characters, it doesn't attempt to
673 untie() will now call an UNTIE() hook if it exists. See L<perltie>
678 L<utime> now supports C<utime undef, undef, @files> to change the
679 file timestamps to the current time.
683 The rules for allowing underscores (underbars) in numeric constants
684 have been relaxed and simplified: now you can have an underscore
685 simply B<between digits>.
689 Rather than relying on C's argv[0] (which may not contain a full pathname)
690 where possible $^X is now set by asking the operating system.
691 (eg by reading F</proc/self/exe> on Linux, F</proc/curproc/file> on FreeBSD)
695 A new variable, C<${^TAINT}>, indicates whether taint mode is enabled.
699 You can now override the readline() builtin, and this overrides also
700 the <FILEHANDLE> angle bracket operator.
704 The command-line options -s and -F are now recognized on the shebang
709 Use of the C</c> match modifier without an accompanying C</g> modifier
710 elicits a new warning: C<Use of /c modifier is meaningless without /g>.
712 Use of C</c> in substitutions, even with C</g>, elicits
713 C<Use of /c modifier is meaningless in s///>.
715 Use of C</g> with C<split> elicits C<Use of /g modifier is meaningless
720 Support for the C<CLONE> special subroutine had been added.
721 With ithreads, when a new thread is created, all Perl data is cloned,
722 however non-Perl data cannot be cloned automatically. In C<CLONE> you
723 can do whatever you need to do, like for example handle the cloning of
724 non-Perl data, if necessary. C<CLONE> will be executed once for every
725 package that has it defined or inherited. It will be called in the
726 context of the new thread, so all modifications are made in the new area.
732 =head1 Modules and Pragmata
734 =head2 New Modules and Pragmata
740 C<Attribute::Handlers> allows a class to define attribute handlers.
743 use Attribute::Handlers;
744 sub Wolf :ATTR(SCALAR) { print "howl!\n" }
746 # later, in some package using or inheriting from MyPack...
748 my MyPack $Fluffy : Wolf; # the attribute handler Wolf will be called
750 Both variables and routines can have attribute handlers. Handlers can
751 be specific to type (SCALAR, ARRAY, HASH, or CODE), or specific to the
752 exact compilation phase (BEGIN, CHECK, INIT, or END).
753 See L<Attribute::Handlers>.
757 C<B::Concise>, by Stephen McCamant, is a new compiler backend for
758 walking the Perl syntax tree, printing concise info about ops.
759 The output is highly customisable. See L<B::Concise>. [561+]
763 The new bignum, bigint, and bigrat pragmas, by Tels, implement
764 transparent bignum support (using the Math::BigInt, Math::BigFloat,
765 and Math::BigRat backends).
769 C<Class::ISA>, by Sean Burke, is a module for reporting the search
770 path for a class's ISA tree. See L<Class::ISA>.
774 C<Cwd> now has a split personality: if possible, an XS extension is
775 used, (this will hopefully be faster, more secure, and more robust)
776 but if not possible, the familiar Perl implementation is used.
780 C<Devel::PPPort>, originally by Kenneth Albanowski and now
781 maintained by Paul Marquess, has been added. It is primarily used
782 by C<h2xs> to enhance portability of XS modules between different
783 versions of Perl. See L<Devel::PPPort>.
787 C<Digest>, frontend module for calculating digests (checksums), from
788 Gisle Aas, has been added. See L<Digest>.
792 C<Digest::MD5> for calculating MD5 digests (checksums) as defined in
793 RFC 1321, from Gisle Aas, has been added. See L<Digest::MD5>.
795 use Digest::MD5 'md5_hex';
797 $digest = md5_hex("Thirsty Camel");
799 print $digest, "\n"; # 01d19d9d2045e005c3f1b80e8b164de1
801 NOTE: the C<MD5> backward compatibility module is deliberately not
802 included since its further use is discouraged.
806 C<Encode>, originally by Nick Ing-Simmons and now maintained by Dan
807 Kogai, provides a mechanism to translate between different character
808 encodings. Support for Unicode, ISO-8859-1, and ASCII are compiled in
809 to the module. Several other encodings (like the rest of the
810 ISO-8859, CP*/Win*, Mac, KOI8-R, three variants EBCDIC, Chinese,
811 Japanese, and Korean encodings) are included and can be loaded at
812 runtime. (For space considerations, the largest Chinese encodings
813 have been separated into their own CPAN module, Encode::HanExtra,
814 which Encode will use if available). See L<Encode>.
816 Any encoding supported by Encode module is also available to the
817 ":encoding()" layer if PerlIO is used.
821 C<Hash::Util> is the interface to the new I<restricted hashes>
822 feature. (Implemented by Jeffrey Friedl, Nick Ing-Simmons, and
823 Michael Schwern.) See L<Hash::Util>.
827 C<I18N::Langinfo> can be used to query locale information.
828 See L<I18N::Langinfo>.
832 C<I18N::LangTags>, by Sean Burke, has functions for dealing with
833 RFC3066-style language tags. See L<I18N::LangTags>.
837 C<ExtUtils::Constant>, by Nicholas Clark, is a new tool for extension
838 writers for generating XS code to import C header constants.
839 See L<ExtUtils::Constant>.
843 C<Filter::Simple>, by Damian Conway, is an easy-to-use frontend to
844 Filter::Util::Call. See L<Filter::Simple>.
850 use Filter::Simple sub {
851 while (my ($from, $to) = splice @_, 0, 2) {
860 use MyFilter qr/red/ => 'green';
862 print "red\n"; # this code is filtered, will print "green\n"
863 print "bored\n"; # this code is filtered, will print "bogreen\n"
867 print "red\n"; # this code is not filtered, will print "red\n"
871 C<File::Temp>, by Tim Jenness, allows one to create temporary files
872 and directories in an easy, portable, and secure way. See L<File::Temp>.
877 C<Filter::Util::Call>, by Paul Marquess, provides you with the
878 framework to write I<source filters> in Perl. For most uses, the
879 frontend Filter::Simple is to be preferred. See L<Filter::Util::Call>.
883 C<if>, by Ilya Zakharevich, is a new pragma for conditional inclusion
888 L<libnet>, by Graham Barr, is a collection of perl5 modules related
889 to network programming. See L<Net::FTP>, L<Net::NNTP>, L<Net::Ping>
890 (not part of libnet, but related), L<Net::POP3>, L<Net::SMTP>,
893 Perl installation leaves libnet unconfigured; use F<libnetcfg>
898 C<List::Util>, by Graham Barr, is a selection of general-utility
899 list subroutines, such as sum(), min(), first(), and shuffle().
904 C<Locale::Constants>, C<Locale::Country>, C<Locale::Currency>
905 C<Locale::Language>, and L<Locale::Script>, by Neil Bowers, have
906 been added. They provide the codes for various locale standards, such
907 as "fr" for France, "usd" for US Dollar, and "ja" for Japanese.
911 $country = code2country('jp'); # $country gets 'Japan'
912 $code = country2code('Norway'); # $code gets 'no'
914 See L<Locale::Constants>, L<Locale::Country>, L<Locale::Currency>,
915 and L<Locale::Language>.
919 C<Locale::Maketext>, by Sean Burke, is a localization framework. See
920 L<Locale::Maketext>, and L<Locale::Maketext::TPJ13>. The latter is an
921 article about software localization, originally published in The Perl
922 Journal #13, and republished here with kind permission.
926 C<Math::BigRat> for big rational numbers, to accompany Math::BigInt and
927 Math::BigFloat, from Tels. See L<Math::BigRat>.
931 C<Memoize> can make your functions faster by trading space for time,
932 from Mark-Jason Dominus. See L<Memoize>.
936 C<MIME::Base64>, by Gisle Aas, allows you to encode data in base64,
937 as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail
942 $encoded = encode_base64('Aladdin:open sesame');
943 $decoded = decode_base64($encoded);
945 print $encoded, "\n"; # "QWxhZGRpbjpvcGVuIHNlc2FtZQ=="
951 C<MIME::QuotedPrint>, by Gisle Aas, allows you to encode data
952 in quoted-printable encoding, as defined in RFC 2045 - I<MIME
953 (Multipurpose Internet Mail Extensions)>.
955 use MIME::QuotedPrint;
957 $encoded = encode_qp("Smiley in Unicode: \x{263a}");
958 $decoded = decode_qp($encoded);
960 print $encoded, "\n"; # "Smiley in Unicode: =263A"
962 MIME::QuotedPrint has been enhanced to provide the basic methods
963 necessary to use it with PerlIO::Via as in :
965 use MIME::QuotedPrint;
966 open($fh,">Via(MIME::QuotedPrint)",$path);
968 See L<MIME::QuotedPrint>.
972 C<NEXT>, by Damian Conway, is a pseudo-class for method redispatch.
977 C<open> is a new pragma for setting the default I/O disciplines
982 C<PerlIO::Scalar>, by Nick Ing-Simmons, provides the implementation
983 of IO to "in memory" Perl scalars as discussed above. It also serves
984 as an example of a loadable PerlIO layer. Other future possibilities
985 include PerlIO::Array and PerlIO::Code. See L<PerlIO::Scalar>.
989 C<PerlIO::Via>, by Nick Ing-Simmons, acts as a PerlIO layer and wraps
990 PerlIO layer functionality provided by a class (typically implemented
993 use MIME::QuotedPrint;
994 open($fh,">Via(MIME::QuotedPrint)",$path);
996 This will automatically convert everything output to C<$fh>
997 to Quoted-Printable. See L<PerlIO::Via>.
1001 C<Pod::ParseLink>, by Russ Allbery, has been added,
1002 to parse LZ<><> links in pods as described in the new
1007 C<Pod::Text::Overstrike>, by Joe Smith, has been added.
1008 It converts POD data to formatted overstrike text.
1009 See L<Pod::Text::Overstrike>. [561+]
1013 C<Scalar::Util> is a selection of general-utility scalar subroutines,
1014 such as blessed(), reftype(), and tainted(). See L<Scalar::Util>.
1018 C<sort> is a new pragma for controlling the behaviour of sort().
1022 C<Storable> gives persistence to Perl data structures by allowing the
1023 storage and retrieval of Perl data to and from files in a fast and
1024 compact binary format. Because in effect Storable does serialisation
1025 of Perl data structues, with it you can also clone deep, hierarchical
1026 datastructures. Storable was originally created by Raphael Manfredi,
1027 but it is now maintained by Abhijit Menon-Sen. Storable has been
1028 enhanced to understand the two new hash features, Unicode keys and
1029 restricted hashes. See L<Storable>.
1033 C<Switch>, by Damian Conway, has been added. Just by saying
1037 you have C<switch> and C<case> available in Perl.
1043 case 1 { print "number 1" }
1044 case "a" { print "string a" }
1045 case [1..10,42] { print "number in list" }
1046 case (@array) { print "number in list" }
1047 case /\w+/ { print "pattern" }
1048 case qr/\w+/ { print "pattern" }
1049 case (%hash) { print "entry in hash" }
1050 case (\%hash) { print "entry in hash" }
1051 case (\&sub) { print "arg to subroutine" }
1052 else { print "previous case not true" }
1059 C<Test::More>, by Michael Schwern, is yet another framework for writing
1060 test scripts, more extensive than Test::Simple. See L<Test::More>.
1064 C<Test::Simple>, by Michael Schwern, has basic utilities for writing
1065 tests. See L<Test::Simple>.
1069 C<Text::Balanced>, by Damian Conway, has been added, for extracting
1070 delimited text sequences from strings.
1072 use Text::Balanced 'extract_delimited';
1074 ($a, $b) = extract_delimited("'never say never', he never said", "'", '');
1076 $a will be "'never say never'", $b will be ', he never said'.
1078 In addition to extract_delimited(), there are also extract_bracketed(),
1079 extract_quotelike(), extract_codeblock(), extract_variable(),
1080 extract_tagged(), extract_multiple(), gen_delimited_pat(), and
1081 gen_extract_tagged(). With these, you can implement rather advanced
1082 parsing algorithms. See L<Text::Balanced>.
1086 C<threads>, by Arthur Bergman, is an interface to interpreter threads.
1087 Interpreter threads (ithreads) is the new thread model introduced in
1088 Perl 5.6 but only available as an internal interface for extension
1089 writers (and for Win32 Perl for C<fork()> emulation). See L<threads>,
1090 L<threads::shared>, and L<perlthrtut>.
1094 C<threads::shared>, by Arthur Bergman, allows data sharing for
1095 interpreter threads. In the ithreads model any data sharing between
1096 threads must be explicit, as opposed to the old 5.005 thread model
1097 where data sharing was implicit. See L<threads::shared>.
1101 C<Tie::File>, by Mark-Jason Dominus, associates a Perl array with the
1102 lines of a file. See L<Tie::File>.
1106 C<Tie::Memoize>, by Ilya Zakharevich, provides on-demand loaded hashes.
1107 See L<Tie::Memoize>.
1111 C<Tie::RefHash::Nestable>, by Edward Avis, allows storing hash
1112 references (unlike the standard Tie::RefHash) The module is contained
1113 within Tie::RefHash. See L<Tie::RefHash>.
1117 C<Time::HiRes>, by Douglas E. Wegscheid, provides high resolution
1118 timing (ualarm, usleep, and gettimeofday). See L<Time::HiRes>.
1122 C<Unicode::UCD> offers a querying interface to the Unicode Character
1123 Database. See L<Unicode::UCD>.
1127 C<Unicode::Collate>, by SADAHIRO Tomoyuki, implements the UCA
1128 (Unicode Collation Algorithm) for sorting Unicode strings.
1129 See L<Unicode::Collate>.
1133 C<Unicode::Normalize>, by SADAHIRO Tomoyuki, implements the various
1134 Unicode normalization forms. See L<Unicode::Normalize>.
1138 C<XS::APItest>, by Tim Jenness, is a test extension that exercises XS
1139 APIs. Currently only C<printf()> is tested: how to output various
1140 basic data types from XS.
1144 C<XS::Typemap>, by Tim Jenness, is a test extension that exercises
1145 XS typemaps. Nothing gets installed, but the code is worth studying
1146 for extension writers.
1150 =head2 Updated And Improved Modules and Pragmata
1156 The following independently supported modules have been updated to the
1157 newest versions from CPAN: CGI, CPAN, DB_File, File::Spec, File::Temp,
1158 Getopt::Long, Math::BigFloat, Math::BigInt, the podlators bundle
1159 (Pod::Man, Pod::Text), Pod::LaTeX [561+], Pod::Parser, Storable,
1160 Term::ANSIColor, Test, Text-Tabs+Wrap.
1164 attributes::reftype() now works on tied arguments.
1168 AutoLoader can now be disabled with C<no AutoLoader;>.
1172 B::Deparse has been significantly enhanced by Robin Houston. It can
1173 now deparse almost all of the standard test suite (so that the tests
1174 still succeed). There is a make target "test.deparse" for trying this
1179 Carp now has better interface documentation, and the @CARP_NOT
1180 interface has been added to get optional control over where errors
1181 are reported independently of @ISA, by Ben Tilly.
1185 Class::Struct can now define the classes in compile time.
1189 Class::Struct now assigns the array/hash element if the accessor
1190 is called with an array/hash element as the B<sole> argument.
1194 The return value of Cwd::fastcwd() is now tainted.
1198 Data::Dumper now has an option to sort hashes.
1202 Data::Dumper now has an option to dump code references
1207 DB_File now supports newer Berkeley DB versions, among
1212 Devel::Peek now has an interface for the Perl memory statistics
1213 (this works only if you are using perl's malloc, and if you have
1214 compiled with debugging).
1218 The English module can now be used without the infamous performance
1221 use English '-no_match_vars';
1223 (Assuming, of course, that you don't need the troublesome variables
1224 C<$`>, C<$&>, or C<$'>.) Also, introduced C<@LAST_MATCH_START> and
1225 C<@LAST_MATCH_END> English aliases for C<@-> and C<@+>.
1229 ExtUtils::MakeMaker now uses File::Spec internally, which hopefully
1230 leads to better portability.
1234 Fcntl, Socket, and Sys::Syslog have been rewritten by Nicholas Clark
1235 to use the new-style constant dispatch section (see L<ExtUtils::Constant>).
1236 This means that they will be more robust and hopefully faster.
1240 File::Find now chdir()s correctly when chasing symbolic links. [561]
1244 File::Find now has pre- and post-processing callbacks. It also
1245 correctly changes directories when chasing symbolic links. Callbacks
1246 (naughtily) exiting with "next;" instead of "return;" now work.
1250 File::Find is now (again) reentrant. It also has been made
1255 The warnings issued by File::Find now belong to their own category.
1256 You can enable/disable them with C<use/no warnings 'File::Find';>.
1260 File::Glob::glob() has been renamed to File::Glob::bsd_glob()
1261 because the name clashes with the builtin glob(). The older
1262 name is still available for compatibility, but is deprecated. [561]
1266 File::Glob now supports C<GLOB_LIMIT> constant to limit the size of
1267 the returned list of filenames.
1271 IPC::Open3 now allows the use of numeric file descriptors.
1275 IO::Socket now has an atmark() method, which returns true if the socket
1276 is positioned at the out-of-band mark. The method is also exportable
1277 as a sockatmark() function.
1281 IO::Socket::INET failed to open the specified port if the service name
1282 was not known. It now correctly uses the supplied port number as is. [561]
1286 IO::Socket::INET has support for the ReusePort option (if your
1287 platform supports it). The Reuse option now has an alias, ReuseAddr.
1288 For clarity, you may want to prefer ReuseAddr.
1292 IO::Socket::INET now supports a value of zero for C<LocalPort>
1293 (usually meaning that the operating system will make one up.)
1297 'use lib' now works identically to @INC. Removing directories
1298 with 'no lib' now works.
1302 Math::BigFloat and Math::BigInt have undergone a full rewrite by Tels.
1303 They are now magnitudes faster, and they support various bignum
1304 libraries such as GMP and PARI as their backends.
1308 Math::Complex handles inf, NaN etc., better.
1312 Net::Ping has been considerably enhanced by Rob Brown: multihoming is
1313 now supported, Win32 functionality is better, there is now time
1314 measuring functionality (optionally high-resolution using
1315 Time::HiRes), and there is now "external" protocol which uses
1316 Net::Ping::External module which runs your external ping utility and
1317 parses the output. A version of Net::Ping::External is available in
1320 Note that some of the Net::Ping tests are disabled when running
1321 under the Perl distribution since one cannot assume one or more
1322 of the following: enabled echo port at localhost, full Internet
1323 connectivity, or sympathetic firewalls. You can set the environment
1324 variable PERL_TEST_Net_Ping to "1" (one) before running the Perl test
1325 suite to enable all the Net::Ping tests.
1329 POSIX::sigaction() is now much more flexible and robust.
1330 You can now install coderef handlers, 'DEFAULT', and 'IGNORE'
1331 handlers, installing new handlers was not atomic.
1335 In Safe, C<%INC> is now localised in a Safe compartment so that
1340 In SDBM_File on dosish platforms, some keys went missing because of
1341 lack of support for files with "holes". A workaround for the problem
1346 In Search::Dict one can now have a pre-processing hook for the
1347 lines being searched.
1351 The Shell module now has an OO interface.
1355 In Sys::Syslog there is now a failover mechanism that will go
1356 through alternative connection mechanisms until the message
1357 is successfully logged.
1361 The Test module has been significantly enhanced.
1365 Time::Local::timelocal() does not handle fractional seconds anymore.
1366 The rationale is that neither does localtime(), and timelocal() and
1367 localtime() are supposed to be inverses of each other.
1371 The vars pragma now supports declaring fully qualified variables.
1372 (Something that C<our()> does not and will not support.)
1376 The C<utf8::> name space (as in the pragma) provides various
1377 Perl-callable functions to provide low level access to Perl's
1378 internal Unicode representation. At the moment only length()
1379 has been implemented.
1383 =head1 Utility Changes
1389 Emacs perl mode (emacs/cperl-mode.el) has been updated to version
1394 F<emacs/e2ctags.pl> is now much faster.
1398 C<enc2xs> is a tool for people adding their own encodings to the
1403 C<h2ph> now supports C trigraphs.
1407 C<h2xs> now produces a template README.
1411 C<h2xs> now uses C<Devel::PPPort> for better portability between
1412 different versions of Perl.
1416 C<h2xs> uses the new L<ExtUtils::Constant|ExtUtils::Constant> module
1417 which will affect newly created extensions that define constants.
1418 Since the new code is more correct (if you have two constants where the
1419 first one is a prefix of the second one, the first constant B<never>
1420 got defined), less lossy (it uses integers for integer constant,
1421 as opposed to the old code that used floating point numbers even for
1422 integer constants), and slightly faster, you might want to consider
1423 regenerating your extension code (the new scheme makes regenerating
1424 easy). L<h2xs> now also supports C trigraphs.
1428 C<libnetcfg> has been added to configure libnet.
1432 C<perlbug> is now much more robust. It also sends the bug report to
1433 perl.org, not perl.com.
1437 C<perlcc> has been rewritten and its user interface (that is,
1438 command line) is much more like that of the UNIX C compiler, cc.
1439 (The perlbc tools has been removed. Use C<perlcc -B> instead.)
1440 B<Note that perlcc is still considered very experimental and
1445 C<perlivp> is a new Installation Verification Procedure utility
1446 for running any time after installing Perl.
1450 C<piconv> is an implementation of the character conversion utility
1451 C<iconv>, demonstrating the new Encode module.
1455 C<pod2html> now allows specifying a cache directory.
1459 C<pod2html> now produces XHTML 1.0.
1463 C<pod2html> now understands POD written using different line endings
1464 (PC-like CRLF versus UNIX-like LF versus MacClassic-like CR).
1468 C<s2p> has been completely rewritten in Perl. (It is in fact a full
1469 implementation of sed in Perl: you can use the sed functionality by
1470 using the C<psed> utility.)
1474 C<xsubpp> now understands POD documentation embedded in the *.xs
1479 C<xsubpp> now supports the OUT keyword.
1483 =head1 New Documentation
1489 perl56delta details the changes between the 5.005 release and the
1494 perlclib documents the internal replacements for standard C library
1495 functions. (Interesting only for extension writers and Perl core
1500 perldebtut is a Perl debugging tutorial. [561+]
1504 perlebcdic contains considerations for running Perl on EBCDIC
1509 perlintro is a gentle introduction to Perl.
1513 perliol documents the internals of PerlIO with layers.
1517 perlmodstyle is a style guide for writing modules.
1521 perlnewmod tells about writing and submitting a new module. [561+]
1525 perlpacktut is a pack() tutorial.
1529 perlpod has been rewritten to be clearer and to record the best
1530 practices gathered over the years.
1534 perlpodspec is a more formal specification of the pod format,
1535 mainly of interest for writers of pod applications, not to
1536 people writing in pod.
1540 perlretut is a regular expression tutorial. [561+]
1544 perlrequick is a regular expressions quick-start guide.
1545 Yes, much quicker than perlretut. [561]
1549 perltodo has been updated.
1553 perltootc has been renamed as perltooc (to not to conflict
1554 with perltoot in filesystems restricted to "8.3" names).
1558 perluniintro is an introduction to using Unicode in Perl.
1559 (perlunicode is more of a detailed reference and background
1564 perlutil explains the command line utilities packaged with the Perl
1565 distribution. [561+]
1569 The following platform-specific documents are available before
1570 the installation as README.I<platform>, and after the installation
1573 perlaix perlamiga perlapollo perlbeos perlbs2000
1574 perlce perlcygwin perldgux perldos perlepoc perlfreebsd perlhpux
1575 perlhurd perlirix perlmachten perlmacos perlmint perlmpeix
1576 perlnetware perlos2 perlos390 perlplan9 perlqnx perlsolaris
1577 perltru64 perluts perlvmesa perlvms perlvos perlwin32
1579 These documents usually detail one or more of the following subjects:
1580 configuring, building, testing, installing, and sometimes also using
1581 Perl on the said platform.
1583 Eastern Asian Perl users are now welcomed in their own languages:
1584 README.jp (Japanese), README.ko (Korean), README.cn (simplified
1585 Chinese) and README.tw (traditional Chinese), which are written in
1586 normal pod but encoded in EUC-JP, EUC-KR, EUC-CN and Big5. These
1587 will get installed as
1589 perljp perlko perlcn perltw
1595 The documentation for the POSIX-BC platform is called "BS2000", to avoid
1596 confusion with the Perl POSIX module.
1600 The documentation for the WinCE platform is called perlce (README.ce
1601 in the source code kit), to avoid confusion with the perlwin32
1602 documentation on 8.3-restricted filesystems.
1606 =head1 Performance Enhancements
1612 map() could get pathologically slow when the result list it generates
1613 is larger than the source list. The performance has been improved for
1614 common scenarios. [561]
1618 sort() is also fully reentrant, in the sense that the sort function
1619 can itself call sort(). This did not work reliably in previous
1624 sort() has been changed to use primarily mergesort internally as
1625 opposed to the earlier quicksort. For very small lists this may
1626 result in slightly slower sorting times, but in general the speedup
1627 should be at least 20%. Additional bonuses are that the worst case
1628 behaviour of sort() is now better (in computer science terms it now
1629 runs in time O(N log N), as opposed to quicksort's Theta(N**2)
1630 worst-case run time behaviour), and that sort() is now stable
1631 (meaning that elements with identical keys will stay ordered as they
1632 were before the sort). See the C<sort> pragma for information.
1634 The story in more detail: suppose you want to serve yourself a little
1637 @digits = ( 3,1,4,1,5,9 );
1639 A numerical sort of the digits will yield (1,1,3,4,5,9), as expected.
1640 Which C<1> comes first is hard to know, since one C<1> looks pretty
1641 much like any other. You can regard this as totally trivial,
1642 or somewhat profound. However, if you just want to sort the even
1643 digits ahead of the odd ones, then what will
1645 sort { ($a % 2) <=> ($b % 2) } @digits;
1647 yield? The only even digit, C<4>, will come first. But how about
1648 the odd numbers, which all compare equal? With the quicksort algorithm
1649 used to implement Perl 5.6 and earlier, the order of ties is left up
1650 to the sort. So, as you add more and more digits of Pi, the order
1651 in which the sorted even and odd digits appear will change.
1652 and, for sufficiently large slices of Pi, the quicksort algorithm
1653 in Perl 5.8 won't return the same results even if reinvoked with the
1654 same input. The justification for this rests with quicksort's
1655 worst case behavior. If you run
1657 sort { $a <=> $b } ( 1 .. $N , 1 .. $N );
1659 (something you might approximate if you wanted to merge two sorted
1660 arrays using sort), doubling $N doesn't just double the quicksort time,
1661 it I<quadruples> it. Quicksort has a worst case run time that can
1662 grow like N**2, so-called I<quadratic> behaviour, and it can happen
1663 on patterns that may well arise in normal use. You won't notice this
1664 for small arrays, but you I<will> notice it with larger arrays,
1665 and you may not live long enough for the sort to complete on arrays
1666 of a million elements. So the 5.8 quicksort scrambles large arrays
1667 before sorting them, as a statistical defence against quadratic behaviour.
1668 But that means if you sort the same large array twice, ties may be
1669 broken in different ways.
1671 Because of the unpredictability of tie-breaking order, and the quadratic
1672 worst-case behaviour, quicksort was I<almost> replaced completely with
1673 a stable mergesort. I<Stable> means that ties are broken to preserve
1674 the original order of appearance in the input array. So
1676 sort { ($a % 2) <=> ($b % 2) } (3,1,4,1,5,9);
1678 will yield (4,3,1,1,5,9), guaranteed. The even and odd numbers
1679 appear in the output in the same order they appeared in the input.
1680 Mergesort has worst case O(N log N) behaviour, the best value
1681 attainable. And, ironically, this mergesort does particularly
1682 well where quicksort goes quadratic: mergesort sorts (1..$N, 1..$N)
1683 in O(N) time. But quicksort was rescued at the last moment because
1684 it is faster than mergesort on certain inputs and platforms.
1685 For example, if you really I<don't> care about the order of even
1686 and odd digits, quicksort will run in O(N) time; it's very good
1687 at sorting many repetitions of a small number of distinct elements.
1688 The quicksort divide and conquer strategy works well on platforms
1689 with relatively small, very fast, caches. Eventually, the problem gets
1690 whittled down to one that fits in the cache, from which point it
1691 benefits from the increased memory speed.
1693 Quicksort was rescued by implementing a sort pragma to control aspects
1694 of the sort. The B<stable> subpragma forces stable behaviour,
1695 regardless of algorithm. The B<_quicksort> and B<_mergesort>
1696 subpragmas are heavy-handed ways to select the underlying implementation.
1697 The leading C<_> is a reminder that these subpragmas may not survive
1698 beyond 5.8. More appropriate mechanisms for selecting the implementation
1699 exist, but they wouldn't have arrived in time to save quicksort.
1703 Hashes now use Bob Jenkins "One-at-a-Time" hashing key algorithm
1704 ( http://burtleburtle.net/bob/hash/doobs.html ). This algorithm is
1705 reasonably fast while producing a much better spread of values than
1706 the old hashing algorithm (originally by Chris Torek, later tweaked by
1707 Ilya Zakharevich). Hash values output from the algorithm on a hash of
1708 all 3-char printable ASCII keys comes much closer to passing the
1709 DIEHARD random number generation tests. According to perlbench, this
1710 change has not affected the overall speed of Perl.
1714 unshift() should now be noticeably faster.
1718 =head1 Installation and Configuration Improvements
1720 =head2 Generic Improvements
1726 INSTALL now explains how you can configure Perl to use 64-bit
1727 integers even on non-64-bit platforms.
1731 Policy.sh policy change: if you are reusing a Policy.sh file
1732 (see INSTALL) and you use Configure -Dprefix=/foo/bar and in the old
1733 Policy $prefix eq $siteprefix and $prefix eq $vendorprefix, all of
1734 them will now be changed to the new prefix, /foo/bar. (Previously
1735 only $prefix changed.) If you do not like this new behaviour,
1736 specify prefix, siteprefix, and vendorprefix explicitly.
1740 A new optional location for Perl libraries, otherlibdirs, is available.
1741 It can be used for example for vendor add-ons without disturbing Perl's
1742 own library directories.
1746 In many platforms, the vendor-supplied 'cc' is too stripped-down to
1747 build Perl (basically, 'cc' doesn't do ANSI C). If this seems
1748 to be the case and 'cc' does not seem to be the GNU C compiler
1749 'gcc', an automatic attempt is made to find and use 'gcc' instead.
1753 gcc needs to closely track the operating system release to avoid
1754 build problems. If Configure finds that gcc was built for a different
1755 operating system release than is running, it now gives a clearly visible
1756 warning that there may be trouble ahead.
1760 Since Perl 5.8 is not binary-compatible with previous releases
1761 of Perl, Configure no longer suggests including the 5.005
1766 Configure C<-S> can now run non-interactively. [561]
1770 Configure support for pdp11-style memory models has been removed due
1771 to obsolescence. [561]
1775 configure.gnu now works with options with whitespace in them.
1779 installperl now outputs everything to STDERR.
1783 Because PerlIO is now the default on most platforms, "-perlio" doesn't
1784 get appended to the $Config{archname} (also known as $^O) anymore.
1785 Instead, if you explicitly choose not to use perlio (Configure command
1786 line option -Uuseperlio), you will get "-stdio" appended.
1790 Another change related to the architecture name is that "-64all"
1791 (-Duse64bitall, or "maximally 64-bit") is appended only if your
1792 pointers are 64 bits wide. (To be exact, the use64bitall is ignored.)
1796 In AFS installations, one can configure the root of the AFS to be
1797 somewhere else than the default F</afs> by using the Configure
1798 parameter C<-Dafsroot=/some/where/else>.
1802 APPLLIB_EXP, a lesser-known configuration-time definition, has been
1803 documented. It can be used to prepend site-specific directories
1804 to Perl's default search path (@INC); see INSTALL for information.
1808 The version of Berkeley DB used when the Perl (and, presumably, the
1809 DB_File extension) was built is now available as
1810 C<@Config{qw(db_version_major db_version_minor db_version_patch)}>
1811 from Perl and as C<DB_VERSION_MAJOR_CFG DB_VERSION_MINOR_CFG
1812 DB_VERSION_PATCH_CFG> from C.
1816 Building Berkeley DB3 for compatibility modes for DB, NDBM, and ODBM
1817 has been documented in INSTALL.
1821 If you have CPAN access (either network or a local copy such as a
1822 CD-ROM) you can during specify extra modules to Configure to build and
1823 install with Perl using the -Dextras=... option. See INSTALL for
1828 In addition to config.over, a new override file, config.arch, is
1829 available. This file is supposed to be used by hints file writers
1830 for architecture-wide changes (as opposed to config.over which is
1831 for site-wide changes).
1835 If your file system supports symbolic links, you can build Perl outside
1836 of the source directory by
1838 mkdir /tmp/perl/build/directory
1839 cd /tmp/perl/build/directory
1840 sh /path/to/perl/source/Configure -Dmksymlinks ...
1842 This will create in /tmp/perl/build/directory a tree of symbolic links
1843 pointing to files in /path/to/perl/source. The original files are left
1844 unaffected. After Configure has finished, you can just say
1848 and Perl will be built and tested, all in /tmp/perl/build/directory.
1853 For Perl developers, several new make targets for profiling
1854 and debugging have been added; see L<perlhack>.
1860 Use of the F<gprof> tool to profile Perl has been documented in
1861 L<perlhack>. There is a make target called "perl.gprof" for
1862 generating a gprofiled Perl executable.
1866 If you have GCC 3, there is a make target called "perl.gcov" for
1867 creating a gcoved Perl executable for coverage analysis. See
1872 If you are on IRIX or Tru64 platforms, new profiling/debugging options
1873 have been added; see L<perlhack> for more information about pixie and
1880 Guidelines of how to construct minimal Perl installations have
1881 been added to INSTALL.
1885 The Thread extension is now not built at all under ithreads
1886 (C<Configure -Duseithreads>) because it wouldn't work anyway (the
1887 Thread extension requires being Configured with C<-Duse5005threads>).
1889 But note that the Thread.pm interface is now shared by both
1894 The Gconvert macro ($Config{d_Gconvert}) used by perl for stringifying
1895 floating-point numbers is now more picky about using sprintf %.*g
1896 rules for the conversion. Some platforms that used to use gcvt may
1897 now resort to the slower sprintf.
1901 The obsolete method of making a special (e.g., debugging) flavor
1904 make LIBPERL=libperld.a
1906 has been removed. Use -DDEBUGGING instead.
1910 =head2 New Or Improved Platforms
1912 For the list of platforms known to support Perl,
1913 see L<perlport/"Supported Platforms">.
1919 AIX dynamic loading should be now better supported.
1923 AIX should now work better with gcc, threads, and 64-bitness. Also the
1924 long doubles support in AIX should be better now. See L<perlaix>.
1928 AtheOS ( http://www.atheos.cx/ ) is a new platform.
1932 BeOS has been reclaimed.
1936 The DG/UX platform now supports 5.005-style threads.
1941 The DYNIX/ptx platform (a.k.a. dynixptx) is supported at or near
1946 EBCDIC platforms (z/OS (also known as OS/390), POSIX-BC, and VM/ESA)
1947 have been regained. Many test suite tests still fail and the
1948 co-existence of Unicode and EBCDIC isn't quite settled, but the
1949 situation is much better than with Perl 5.6. See L<perlos390>,
1950 L<perlbs2000> (for POSIX-BC), and L<perlvmesa> for more information.
1954 Building perl with -Duseithreads or -Duse5005threads now works under
1955 HP-UX 10.20 (previously it only worked under 10.30 or later). You will
1956 need a thread library package installed. See README.hpux. [561]
1960 Mac OS Classic is now supported in the mainstream source package
1961 (MacPerl has of course been available since perl 5.004 but now the
1962 source code bases of standard Perl and MacPerl have been synchronised)
1967 Mac OS X (or Darwin) should now be able to build Perl even on HFS+
1968 filesystems. (The case-insensitivity used to confuse the Perl build
1973 NCR MP-RAS is now supported. [561]
1977 All the NetBSD specific patches (except for the installation
1978 specific ones) have been merged back to the main distribution.
1982 NetWare from Novell is now supported. See L<perlnetware>.
1986 NonStop-UX is now supported. [561]
1990 NEC SUPER-UX is now supported.
1994 All the OpenBSD specific patches (except for the installation
1995 specific ones) have been merged back to the main distribution.
1999 Perl has been tested with the GNU pth userlevel thread package
2000 ( http://www.gnu.org/software/pth/pth.html ). All thread tests
2001 of Perl now work, but not without adding some yield()s to the tests,
2002 so while pth (and other userlevel thread implementations) can be
2003 considered to be "working" with Perl ithreads, keep in mind the
2004 possible non-preemptability of the underlying thread implementation.
2008 Stratus VOS is now supported using Perl's native build method
2009 (Configure). This is the recommended method to build Perl on
2010 VOS. The older methods, which build miniperl, are still
2011 available. See L<perlvos>. [561+]
2015 The Amdahl UTS UNIX mainframe platform is now supported. [561]
2019 WinCE is now supported. See L<perlce>.
2023 z/OS (formerly known as OS/390, formerly known as MVS OE) now has
2024 support for dynamic loading. This is not selected by default,
2025 however, you must specify -Dusedl in the arguments of Configure. [561]
2029 =head1 Selected Bug Fixes
2031 Numerous memory leaks and uninitialized memory accesses have been
2032 hunted down. Most importantly, anonymous subs used to leak quite
2039 The autouse pragma didn't work for Multi::Part::Function::Names.
2043 caller() could cause core dumps in certain situations. Carp was
2044 sometimes affected by this problem. In particular, caller() now
2045 returns a subroutine name of C<(unknown)> for subroutines that have
2046 been removed from the symbol table.
2050 chop(@list) in list context returned the characters chopped in
2051 reverse order. This has been reversed to be in the right order. [561]
2055 Configure no longer includes the DBM libraries (dbm, gdbm, db, ndbm)
2056 when building the Perl binary. The only exception to this is SunOS 4.x,
2057 which needs them. [561]
2061 The behaviour of non-decimal but numeric string constants such as
2062 "0x23" was platform-dependent: in some platforms that was seen as 35,
2063 in some as 0, in some as a floating point number (don't ask). This
2064 was caused by Perl's using the operating system libraries in a situation
2065 where the result of the string to number conversion is undefined: now
2066 Perl consistently handles such strings as zero in numeric contexts.
2070 The order of DESTROYs has been made more predictable.
2074 Several debugger fixes: exit code now reflects the script exit code,
2075 condition C<"0"> now treated correctly, the C<d> command now checks
2076 line number, C<$.> no longer gets corrupted, and all debugger output
2077 now goes correctly to the socket if RemotePort is set. [561]
2081 Perl 5.6.0 could emit spurious warnings about redefinition of
2082 dl_error() when statically building extensions into perl.
2083 This has been corrected. [561]
2087 L<dprofpp> -R didn't work.
2091 C<*foo{FORMAT}> now works.
2095 Infinity is now recognized as a number.
2099 UNIVERSAL::isa no longer caches methods incorrectly. (This broke
2100 the Tk extension with 5.6.0.) [561]
2104 Lexicals I: lexicals outside an eval "" weren't resolved
2105 correctly inside a subroutine definition inside the eval "" if they
2106 were not already referenced in the top level of the eval""ed code.
2110 Lexicals II: lexicals leaked at file scope into subroutines that
2111 were declared before the lexicals.
2115 Lexical warnings now propagating correctly between scopes
2116 and into C<eval "...">.
2120 C<use warnings qw(FATAL all)> did not work as intended. This has been
2125 warnings::enabled() now reports the state of $^W correctly if the caller
2126 isn't using lexical warnings. [561]
2130 Line renumbering with eval and C<#line> now works. [561]
2134 Fixed numerous memory leaks, especially in eval "".
2138 Localised tied variables no longer leak memory
2141 tie my %tied_hash => 'Tie::StdHash';
2145 # Used to leak memory every time local() was called;
2146 # in a loop, this added up.
2147 local($tied_hash{Foo}) = 1;
2151 Localised hash elements (and %ENV) are correctly unlocalised to not
2152 exist, if they didn't before they were localised.
2156 tie my %tied_hash => 'Tie::StdHash';
2160 # Nothing has set the FOO element so far
2162 { local $tied_hash{FOO} = 'Bar' }
2164 # This used to print, but not now.
2165 print "exists!\n" if exists $tied_hash{FOO};
2167 As a side effect of this fix, tied hash interfaces B<must> define
2168 the EXISTS and DELETE methods.
2172 mkdir() now ignores trailing slashes in the directory name,
2173 as mandated by POSIX.
2177 Some versions of glibc have a broken modfl(). This affects builds
2178 with C<-Duselongdouble>. This version of Perl detects this brokenness
2179 and has a workaround for it. The glibc release 2.2.2 is known to have
2180 fixed the modfl() bug.
2184 Modulus of unsigned numbers now works (4063328477 % 65535 used to
2185 return 27406, instead of 27047). [561]
2189 Some "not a number" warnings introduced in 5.6.0 eliminated to be
2190 more compatible with 5.005. Infinity is now recognised as a number. [561]
2194 Numeric conversions did not recognize changes in the string value
2195 properly in certain circumstances. [561]
2199 Attributes (such as :shared) didn't work with our().
2203 our() variables will not cause bogus "Variable will not stay shared"
2208 "our" variables of the same name declared in two sibling blocks
2209 resulted in bogus warnings about "redeclaration" of the variables.
2210 The problem has been corrected. [561]
2214 pack "Z" now correctly terminates the string with "\0".
2218 Fix password routines which in some shadow password platforms
2219 (e.g. HP-UX) caused getpwent() to return every other entry.
2223 The PERL5OPT environment variable (for passing command line arguments
2224 to Perl) didn't work for more than a single group of options. [561]
2228 PERL5OPT with embedded spaces didn't work.
2232 printf() no longer resets the numeric locale to "C".
2236 C<qw(a\\b)> now parses correctly as C<'a\\b'>: that is, as three
2237 characters, not four. [561]
2241 pos() did not return the correct value within s///ge in earlier
2242 versions. This is now handled correctly. [561]
2246 Printing quads (64-bit integers) with printf/sprintf now works
2247 without the q L ll prefixes (assuming you are on a quad-capable platform).
2251 Regular expressions on references and overloaded scalars now work. [561+]
2255 Right-hand side magic (GMAGIC) could in many cases such as string
2256 concatenation be invoked too many times.
2260 scalar() now forces scalar context even when used in void context.
2264 SOCKS support is now much more robust.
2268 sort() arguments are now compiled in the right wantarray context
2269 (they were accidentally using the context of the sort() itself).
2270 The comparison block is now run in scalar context, and the arguments
2271 to be sorted are always provided list context. [561]
2275 Changed the POSIX character class C<[[:space:]]> to include the (very
2276 rarely used) vertical tab character. Added a new POSIX-ish character
2277 class C<[[:blank:]]> which stands for horizontal whitespace
2278 (currently, the space and the tab).
2282 The tainting behaviour of sprintf() has been rationalized. It does
2283 not taint the result of floating point formats anymore, making the
2284 behaviour consistent with that of string interpolation. [561]
2288 Some cases of inconsistent taint propagation (such as within hash
2289 values) have been fixed.
2293 The RE engine found in Perl 5.6.0 accidentally pessimised certain kinds
2294 of simple pattern matches. These are now handled better. [561]
2298 Regular expression debug output (whether through C<use re 'debug'>
2299 or via C<-Dr>) now looks better. [561]
2303 Multi-line matches like C<"a\nxb\n" =~ /(?!\A)x/m> were flawed. The
2304 bug has been fixed. [561]
2308 Use of $& could trigger a core dump under some situations. This
2309 is now avoided. [561]
2313 The regular expression captured submatches ($1, $2, ...) are now
2314 more consistently unset if the match fails, instead of leaving false
2315 data lying around in them. [561]
2319 readline() on files opened in "slurp" mode could return an extra
2320 "" (blank line) at the end in certain situations. This has been
2325 Autovivification of symbolic references of special variables described
2326 in L<perlvar> (as in C<${$num}>) was accidentally disabled. This works
2331 Sys::Syslog ignored the C<LOG_AUTH> constant.
2335 $AUTOLOAD, sort(), lock(), and spawning subprocesses
2336 in multiple threads simultaneously are now thread-safe.
2340 Tie::Array's SPLICE method was broken.
2344 Allow a read-only string on the left-hand side of a non-modifying tr///.
2348 If C<STDERR> is tied, warnings caused by C<warn> and C<die> now
2349 correctly pass to it.
2353 Several Unicode fixes.
2359 BOMs (byte order marks) at the beginning of Perl files
2360 (scripts, modules) should now be transparently skipped.
2361 UTF-16 and UCS-2 encoded Perl files should now be read correctly.
2365 The character tables have been updated to Unicode 3.2.0.
2369 Comparing with utf8 data does not magically upgrade non-utf8 data
2370 into utf8. (This was a problem for example if you were mixing data
2371 from I/O and Unicode data: your output might have got magically encoded
2376 Generating illegal Unicode code points such as U+FFFE, or the UTF-16
2377 surrogates, now also generates an optional warning.
2381 C<IsAlnum>, C<IsAlpha>, and C<IsWord> now match titlecase.
2385 Concatenation with the C<.> operator or via variable interpolation,
2386 C<eq>, C<substr>, C<reverse>, C<quotemeta>, the C<x> operator,
2387 substitution with C<s///>, single-quoted UTF8, should now work.
2391 The C<tr///> operator now works. Note that the C<tr///CU>
2392 functionality has been removed (but see pack('U0', ...)).
2396 C<eval "v200"> now works.
2400 Perl 5.6.0 parsed m/\x{ab}/ incorrectly, leading to spurious warnings.
2401 This has been corrected. [561]
2405 Zero entries were missing from the Unicode classes such as C<IsDigit>.
2411 Large unsigned numbers (those above 2**31) could sometimes lose their
2412 unsignedness, causing bogus results in arithmetic operations. [561]
2416 The Perl parser has been stress tested using both random input and
2417 Markov chain input and the few found crashes and lockups have been
2422 =head2 Platform Specific Changes and Fixes
2430 Perl now works on post-4.0 BSD/OSes.
2436 Setting C<$0> now works (as much as possible; see L<perlvar> for details).
2442 Numerous updates; currently synchronised with Cygwin 1.3.10.
2446 Previously DYNIX/ptx had problems in its Configure probe for non-blocking I/O.
2452 EPOC now better supported. See README.epoc. [561]
2458 Perl now works on post-3.0 FreeBSDs.
2464 README.hpux updated; C<Configure -Duse64bitall> now works;
2465 now uses HP-UX malloc instead of Perl malloc.
2471 Numerous compilation flag and hint enhancements; accidental mixing
2472 of 32-bit and 64-bit libraries (a doomed attempt) made much harder.
2482 Long doubles should now work (see INSTALL). [561]
2486 Linux previously had problems related to sockaddrlen when using
2487 accept(), recvfrom() (in Perl: recv()), getpeername(), and
2496 Compilation of the standard Perl distribution in Mac OS Classic should
2497 now work if you have the Metrowerks development environment and the
2498 missing Mac-specific toolkit bits. Contact the macperl mailing list
2505 MPE/iX update after Perl 5.6.0. See README.mpeix. [561]
2509 NetBSD/threads: try installing the GNU pth (should be in the
2510 packages collection, or http://www.gnu.org/software/pth/),
2511 and Configure with -Duseithreads.
2517 Perl now works on NetBSD/sparc.
2523 Now works with usethreads (see INSTALL). [561]
2529 64-bitness using the Sun Workshop compiler now works.
2535 The native build method requires at least VOS Release 14.5.0
2536 and GNU C++/GNU Tools 2.0.1 or later. The Perl pack function
2537 now maps overflowed values to +infinity and underflowed values
2542 Tru64 (aka Digital UNIX, aka DEC OSF/1)
2544 The operating system version letter now recorded in $Config{osvers}.
2545 Allow compiling with gcc (previously explicitly forbidden). Compiling
2546 with gcc still not recommended because buggy code results, even with
2553 Fixed various alignment problems that lead into core dumps either
2554 during build or later; no longer dies on math errors at runtime;
2555 now using full quad integers (64 bits), previously was using
2556 only 46 bit integers for speed.
2562 See L</"Socket Extension Dynamic in VMS"> and L</"IEEE-format Floating Point
2563 Default on OpenVMS Alpha"> for important changes not otherwise listed here.
2565 chdir() now works better despite a CRT bug; now works with MULTIPLICITY
2566 (see INSTALL); now works with Perl's malloc.
2568 The tainting of C<%ENV> elements via C<keys> or C<values> was previously
2569 unimplemented. It now works as documented.
2571 The C<waitpid> emulation has been improved. The worst bug (now fixed)
2572 was that a pid of -1 would cause a wildcard search of all processes on
2575 POSIX-style signals are now emulated much better on VMS versions prior
2578 The C<system> function and backticks operator have improved
2579 functionality and better error handling. [561]
2581 File access tests now use current process privileges rather than the
2582 user's default privileges, which could sometimes result in a mismatch
2583 between reported access and actual access. This improvement is only
2584 available on VMS v6.0 and later.
2586 There is a new C<kill> implementation based on C<sys$sigprc> that allows
2587 older VMS systems (pre-7.0) to use C<kill> to send signals rather than
2588 simply force exit. This implementation also allows later systems to
2589 call C<kill> from within a signal handler.
2591 Iterative logical name translations are now limited to 10 iterations in
2592 imitation of SHOW LOGICAL and other OpenVMS facilities.
2602 Signal handling now works better than it used to. It is now implemented
2603 using a Windows message loop, and is therefore less prone to random
2608 fork() emulation is now more robust, but still continues to have a few
2609 esoteric bugs and caveats. See L<perlfork> for details. [561+]
2613 A failed (pseudo)fork now returns undef and sets errno to EAGAIN. [561]
2617 The following modules now work on Windows:
2619 ExtUtils::Embed [561]
2626 IO::File::new_tmpfile() is no longer limited to 32767 invocations
2631 Better chdir() return value for a non-existent directory.
2635 Compiling perl using the 64-bit Platform SDK tools is now supported.
2639 The Win32::SetChildShowWindow() builtin can be used to control the
2640 visibility of windows created by child processes. See L<Win32> for
2645 Non-blocking waits for child processes (or pseudo-processes) are
2646 supported via C<waitpid($pid, &POSIX::WNOHANG)>.
2650 The behavior of system() with multiple arguments has been rationalized.
2651 Each unquoted argument will be automatically quoted to protect whitespace,
2652 and any existing whitespace in the arguments will be preserved. This
2653 improves the portability of system(@args) by avoiding the need for
2654 Windows C<cmd> shell specific quoting in perl programs.
2656 Note that this means that some scripts that may have relied on earlier
2657 buggy behavior may no longer work correctly. For example,
2658 C<system("nmake /nologo", @args)> will now attempt to run the file
2659 C<nmake /nologo> and will fail when such a file isn't found.
2660 On the other hand, perl will now execute code such as
2661 C<system("c:/Program Files/MyApp/foo.exe", @args)> correctly.
2665 The perl header files no longer suppress common warnings from the
2666 Microsoft Visual C++ compiler. This means that additional warnings may
2667 now show up when compiling XS code.
2671 Borland C++ v5.5 is now a supported compiler that can build Perl.
2672 However, the generated binaries continue to be incompatible with those
2673 generated by the other supported compilers (GCC and Visual C++). [561]
2677 Duping socket handles with open(F, ">&MYSOCK") now works under Windows 9x.
2682 Current directory entries in %ENV are now correctly propagated to child
2687 New %ENV entries now propagate to subprocesses. [561]
2691 Win32::GetCwd() correctly returns C:\ instead of C: when at the drive root.
2692 Other bugs in chdir() and Cwd::cwd() have also been fixed. [561]
2696 The makefiles now default to the features enabled in ActiveState ActivePerl
2697 (a popular Win32 binary distribution). [561]
2701 HTML files will now be installed in c:\perl\html instead of
2702 c:\perl\lib\pod\html
2706 REG_EXPAND_SZ keys are now allowed in registry settings used by perl. [561]
2710 Can now send() from all threads, not just the first one. [561]
2714 ExtUtils::MakeMaker now uses $ENV{LIB} to search for libraries. [561]
2718 Less stack reserved per thread so that more threads can run
2719 concurrently. (Still 16M per thread.) [561]
2723 C<< File::Spec->tmpdir() >> now prefers C:/temp over /tmp
2724 (works better when perl is running as service).
2728 Better UNC path handling under ithreads. [561]
2732 wait(), waitpid(), and backticks now return the correct exit status
2733 under Windows 9x. [561]
2737 A socket handle leak in accept() has been fixed. [561]
2743 =head1 New or Changed Diagnostics
2749 The lexical warnings category "deprecated" is no longer a sub-category
2750 of the "syntax" category. It is now a top-level category in its own
2755 All regular expression compilation error messages are now hopefully
2756 easier to understand both because the error message now comes before
2757 the failed regex and because the point of failure is now clearly
2758 marked by a C<E<lt>-- HERE> marker.
2762 The various "opened only for", "on closed", "never opened" warnings
2763 drop the C<main::> prefix for filehandles in the C<main> package,
2764 for example C<STDIN> instead of C<main::STDIN>.
2768 The "Unrecognized escape" warning has been extended to include C<\8>,
2769 C<\9>, and C<\_>. There is no need to escape any of the C<\w> characters.
2773 Two new debugging options have been added: if you have compiled your
2774 Perl with debugging, you can use the -DT [561] and -DR options to trace
2775 tokenising and to add reference counts to displaying variables,
2780 The debugger (perl5db.pl) has been modified to present a more
2781 consistent commands interface, via (CommandSet=580). perl5db.t was
2782 also added to test the changes, and as a placeholder for further tests.
2788 The debugger has a new C<dumpDepth> option to control the maximum
2789 depth to which nested structures are dumped. The C<x> command has
2790 been extended so that C<x N EXPR> dumps out the value of I<EXPR> to a
2791 depth of at most I<N> levels.
2795 The debugger can now show lexical variables if you have the CPAN
2796 module PadWalker installed.
2800 If an attempt to use a (non-blessed) reference as an array index
2801 is made, a warning is given.
2805 C<push @a;> and C<unshift @a;> (with no values to push or unshift)
2806 now give a warning. This may be a problem for generated and evaled
2811 If you try to L<perlfunc/pack> a number less than 0 or larger than 255
2812 using the C<"C"> format you will get an optional warning. Similarly
2813 for the C<"c"> format and a number less than -128 or more than 127.
2817 Certain regex modifiers such as C<(?o)> make sense only if applied to
2818 the entire regex. You will get an optional warning if you try to do
2823 Using arrays or hashes as references (e.g. C<< %foo->{bar} >>
2824 has been deprecated for a while. Now you will get an optional warning.
2828 Using C<sort> in scalar context now issues an optional warning.
2829 This didn't do anything useful, as the sort was not performed.
2833 =head1 Changed Internals
2839 perlapi.pod (a companion to perlguts) now attempts to document the
2844 You can now build a really minimal perl called microperl.
2845 Building microperl does not require even running Configure;
2846 C<make -f Makefile.micro> should be enough. Beware: microperl makes
2847 many assumptions, some of which may be too bold; the resulting
2848 executable may crash or otherwise misbehave in wondrous ways.
2849 For careful hackers only.
2853 Added rsignal(), whichsig(), do_join(), op_clear, op_null,
2854 ptr_table_clear(), ptr_table_free(), sv_setref_uv(), and several UTF-8
2855 interfaces to the publicised API. For the full list of the available
2856 APIs see L<perlapi>.
2860 Made possible to propagate customised exceptions via croak()ing.
2864 Now xsubs can have attributes just like subs. (Well, at least the
2865 built-in attributes.)
2869 dTHR and djSP have been obsoleted; the former removed (because it's
2870 a no-op) and the latter replaced with dSP.
2874 PERL_OBJECT has been completely removed.
2878 The MAGIC constants (e.g. C<'P'>) have been macrofied
2879 (e.g. C<PERL_MAGIC_TIED>) for better source code readability
2880 and maintainability.
2884 The regex compiler now maintains a structure that identifies nodes in
2885 the compiled bytecode with the corresponding syntactic features of the
2886 original regex expression. The information is attached to the new
2887 C<offsets> member of the C<struct regexp>. See L<perldebguts> for more
2888 complete information.
2892 The C code has been made much more C<gcc -Wall> clean. Some warning
2893 messages still remain in some platforms, so if you are compiling with
2894 gcc you may see some warnings about dubious practices. The warnings
2895 are being worked on.
2899 F<perly.c>, F<sv.c>, and F<sv.h> have now been extensively commented.
2903 Documentation on how to use the Perl source repository has been added
2904 to F<Porting/repository.pod>.
2908 There are now several profiling make targets.
2912 =head1 Security Vulnerability Closed [561]
2914 (This change was already made in 5.7.0 but bears repeating here.)
2916 A potential security vulnerability in the optional suidperl component
2917 of Perl was identified in August 2000. suidperl is neither built nor
2918 installed by default. As of November 2001 the only known vulnerable
2919 platform is Linux, most likely all Linux distributions. CERT and
2920 various vendors and distributors have been alerted about the vulnerability.
2921 See http://www.cpan.org/src/5.0/sperl-2000-08-05/sperl-2000-08-05.txt
2922 for more information.
2924 The problem was caused by Perl trying to report a suspected security
2925 exploit attempt using an external program, /bin/mail. On Linux
2926 platforms the /bin/mail program had an undocumented feature which
2927 when combined with suidperl gave access to a root shell, resulting in
2928 a serious compromise instead of reporting the exploit attempt. If you
2929 don't have /bin/mail, or if you have 'safe setuid scripts', or if
2930 suidperl is not installed, you are safe.
2932 The exploit attempt reporting feature has been completely removed from
2933 Perl 5.8.0 (and the maintenance release 5.6.1, and it was removed also
2934 from all the Perl 5.7 releases), so that particular vulnerability
2935 isn't there anymore. However, further security vulnerabilities are,
2936 unfortunately, always possible. The suidperl functionality is most
2937 probably going to be removed in Perl 5.10. In any case, suidperl
2938 should only be used by security experts who know exactly what they are
2939 doing and why they are using suidperl instead of some other solution
2940 such as sudo ( see http://www.courtesan.com/sudo/ ).
2944 Several new tests have been added, especially for the F<lib> and F<ext>
2945 subsections. There are now about 65 000 individual tests (spread over
2946 about 700 test scripts), in the regression suite (5.6.1 has about
2947 11700 tests, in 258 test scripts) Many of the new tests are of course
2948 introduced by the new modules, but still in general Perl is now more
2951 Because of the large number of tests, running the regression suite
2952 will take considerably longer time than it used to: expect the suite
2953 to take up to 4-5 times longer to run than in perl 5.6. On a really
2954 fast machine you can hope to finish the suite in about 6-8 minutes
2957 The tests are now reported in a different order than in earlier Perls.
2958 (This happens because the test scripts from under t/lib have been moved
2959 to be closer to the library/extension they are testing.)
2961 =head1 Known Problems
2969 If using the AIX native make command, instead of just "make" issue
2970 "make all". In some setups the former has been known to spuriously
2971 also try to run "make install". Alternatively, you may want to use
2976 In AIX 4.2, Perl extensions that use C++ functions that use statics
2977 may have problems in that the statics are not getting initialized.
2978 In newer AIX releases, this has been solved by linking Perl with
2979 the libC_r library, but unfortunately in AIX 4.2 the said library
2980 has an obscure bug where the various functions related to time
2981 (such as time() and gettimeofday()) return broken values, and
2982 therefore in AIX 4.2 Perl is not linked against libC_r.
2986 vac 5.0.0.0 May Produce Buggy Code For Perl
2988 The AIX C compiler vac version 5.0.0.0 may produce buggy code,
2989 resulting in a few random tests failing when run as part of "make
2990 test", but when the failing tests are run by hand, they succeed.
2991 We suggest upgrading to at least vac version 5.0.1.0, that has been
2992 known to compile Perl correctly. "lslpp -L|grep vac.C" will tell
2993 you the vac version. See README.aix.
2997 If building threaded Perl, you may get compilation warning from pp_sys.c:
2999 "pp_sys.c", line 4651.39: 1506-280 (W) Function argument assignment between types "unsigned char*" and "const void*" is not allowed.
3001 This is harmless; it is caused by the getnetbyaddr() and getnetbyaddr_r()
3002 having slightly different types for their first argument.
3006 =head2 Alpha systems with old gccs fail several tests
3008 If you see op/pack, op/pat, op/regexp, or ext/Storable tests failing
3009 in a Linux/alpha or *BSD/Alpha, it's probably time to upgrade your gcc.
3010 gccs prior to 2.95.3 are definitely not good enough, and gcc 3.1 may
3011 be even better. (RedHat Linux/alpha with gcc 3.1 reported no problems,
3012 as did Linux 2.4.18 with gcc 2.95.4.) (In Tru64, it is preferable to
3013 use the bundled C compiler.)
3017 Perl 5.8.0 doesn't build in AmigaOS. It broke at some point during
3018 the ithreads work and we could not find Amiga experts to unbreak the
3019 problems. Perl 5.6.1 still works for AmigaOS (as does the the 5.7.2
3020 development release).
3024 The following tests fail on 5.8.0 Perl in BeOS Personal 5.03:
3026 t/op/lfs............................FAILED at test 17
3027 t/op/magic..........................FAILED at test 24
3028 ext/POSIX/t/sigaction...............FAILED at test 13
3029 ext/POSIX/t/waitpid.................FAILED at test 1
3031 See L<perlbeos> (README.beos) for more details.
3033 =head2 Cygwin "unable to remap"
3035 For example when building the Tk extension for Cygwin,
3036 you may get an error message saying "unable to remap".
3037 This is known problem with Cygwin, and a workaround is
3038 detailed in here: http://sources.redhat.com/ml/cygwin/2001-12/msg00894.html
3040 =head2 ext/threads/t/libc
3042 If this test fails, it indicates that your libc (C library) is not
3043 threadsafe. This particular test stress tests the localtime() call to
3044 find out whether it is threadsafe. See L<perlthrtut> for more information.
3046 =head2 FreeBSD built with ithreads coredumps reading large directories
3048 This is a known bug in FreeBSD's readdir_r() (see L<perlfreebsd>
3049 (README.freebsd)), which hopefully will be fixed in FreeBSD 4.6.
3051 =head2 FreeBSD Failing locale Test 117 For ISO8859-15 Locales
3053 The ISO8859-15 locales may fail the locale test 117 in FreeBSD.
3054 This is caused by the characters \xFF (y with diaeresis) and \xBE
3055 (Y with diaeresis) not behaving correctly when being matched
3058 =head2 IRIX fails ext/List/Util/t/shuffle.t
3060 IRIX with MIPSpro 7.3.1.3m compiler may fail the said List::Util test
3061 by dumping core. This seems to be a compiler error since if compiled
3062 with gcc no core dump ensues, and no failures on the said test on any
3065 =head2 Modifying $_ Inside for(..)
3069 works without complaint. It shouldn't. (You should be able to
3070 modify only lvalue elements inside the loops.) You can see the
3071 correct behaviour by replacing the 1..5 with 1, 2, 3, 4, 5.
3073 =head2 mod_perl 1.26 Doesn't Build With Threaded Perl
3075 Use mod_perl 1.27 or higher.
3077 =head2 lib/ftmp-security tests warn 'system possibly insecure'
3079 Don't panic. Read the 'make test' section of INSTALL instead.
3081 =head2 HP-UX lib/posix Subtest 9 Fails When LP64-Configured
3083 If perl is configured with -Duse64bitall, the successful result of the
3084 subtest 10 of lib/posix may arrive before the successful result of the
3085 subtest 9, which confuses the test harness so much that it thinks the
3088 =head2 Linux with glibc 2.2.5 fails t/op/int subtest #6 with -Duse64bitint
3090 This is a known bug in the glibc 2.2.5 with long long integers.
3091 ( http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=65612 )
3093 =head2 Linux With Sfio Fails op/misc Test 48
3097 =head2 libwww-perl (LWP) fails base/date #51
3099 Use libwww-perl 5.65 or later.
3103 Please remember to set your environment variable LC_ALL to "C"
3104 (setenv LC_ALL C) before running "make test" to avoid a lot of
3105 warnings about the broken locales of Mac OS X.
3107 The following tests are known to fail in Mac OS X 10.1.5 because of
3108 buggy (old) implementations of Berkeley DB included in Mac OS X:
3110 Failed Test Stat Wstat Total Fail Failed List of Failed
3111 -------------------------------------------------------------------------
3112 ../ext/DB_File/t/db-btree.t 0 11 ?? ?? % ??
3113 ../ext/DB_File/t/db-recno.t 149 3 2.01% 61 63 65
3115 If you are building on a UFS partition, you will also probably see
3116 t/op/stat.t subtest #9 fail. This is caused by Darwin's UFS not
3117 supporting inode change time.
3119 Also the ext/POSIX/t/posix.t subtest #10 fails but it is skipped for
3120 now because the failure is Apple's fault, not Perl's (blocked signals
3123 If you Configure with ithreads, ext/threads/t/libc.t will fail. Again,
3124 this is not Perl's fault-- the libc of Mac OS X is not threadsafe
3125 (in this particular test, the localtime() call is found to be
3128 =head2 op/sprintf tests 91, 129, and 130
3130 The op/sprintf tests 91, 129, and 130 are known to fail on some platforms.
3131 Examples include any platform using sfio, and Compaq/Tandem's NonStop-UX.
3133 Test 91 is known to fail on QNX6 (nto), because C<sprintf '%e',0>
3134 incorrectly produces C<0.000000e+0> instead of C<0.000000e+00>.
3136 For tests 129 and 130, the failing platforms do not comply with
3137 the ANSI C Standard: lines 19ff on page 134 of ANSI X3.159 1989, to
3138 be exact. (They produce something other than "1" and "-1" when
3139 formatting 0.6 and -0.6 using the printf format "%.0f"; most often,
3140 they produce "0" and "-0".)
3144 In case you are still using Solaris 2.5 (aka SunOS 5.5), you may
3145 experience failures (the test core dumping) in lib/locale.t.
3146 The suggested cure is to upgrade your Solaris.
3148 =head2 SUPER-UX (NEC SX)
3150 The following tests are known to fail on SUPER-UX:
3152 op/64bitint...........................FAILED tests 29-30, 32-33, 35-36
3153 op/arith..............................FAILED tests 128-130
3154 op/pack...............................FAILED tests 25-5625
3155 op/pow................................
3156 op/taint..............................# msgsnd failed
3157 ../ext/IO/lib/IO/t/io_poll............FAILED tests 3-4
3158 ../ext/IPC/SysV/ipcsysv...............FAILED tests 2, 5-6
3159 ../ext/IPC/SysV/t/msg.................FAILED tests 2, 4-6
3160 ../ext/Socket/socketpair..............FAILED tests 12
3161 ../lib/IPC/SysV.......................FAILED tests 2, 5-6
3162 ../lib/warnings.......................FAILED tests 115-116, 118-119
3164 The op/pack failure ("Cannot compress negative numbers at op/pack.t line 126")
3165 is serious but as of yet unsolved. It points at some problems with the
3166 signedness handling of the C compiler, as do the 64bitint, arith, and pow
3167 failures. Most of the rest point at problems with SysV IPC.
3169 =head2 PDL failing some tests
3171 Use PDL 2.3.4 or later.
3173 =head2 Term::ReadKey not working on Win32
3175 Use Term::ReadKey 2.20 or later.
3177 =head2 Failure of Thread (5.005-style) tests
3179 B<Note that support for 5.005-style threading is deprecated,
3180 experimental and practically unsupported. In 5.10, it is expected
3183 The following tests are known to fail due to fundamental problems in
3184 the 5.005 threading implementation. These are not new failures--Perl
3185 5.005_0x has the same bugs, but didn't have these tests.
3187 ../ext/B/t/xref.t 255 65280 14 12 85.71% 3-14
3188 ../ext/List/Util/t/first.t 255 65280 7 4 57.14% 2 5-7
3189 ../lib/English.t 2 512 54 2 3.70% 2-3
3190 ../lib/ExtUtils/t/basic.t 1 256 17 1 5.88% 14
3191 ../lib/FileCache.t 5 1 20.00% 5
3192 ../lib/Filter/Simple/t/data.t 6 3 50.00% 1-3
3193 ../lib/Filter/Simple/t/filter_onl 9 3 33.33% 1-2 5
3194 ../lib/Tie/File/t/31_autodefer.t 255 65280 65 32 49.23% 34-65
3195 ../lib/autouse.t 10 1 10.00% 4
3196 op/flip.t 15 1 6.67% 15
3198 These failures are unlikely to get fixed as 5.005-style threads
3199 are considered fundamentally broken. (Basically what happens is that
3200 competing threads can corrupt shared global state.)
3202 =head2 Timing problems
3204 The following tests may fail intermittently because of timing
3205 problems, for example if the system is heavily loaded.
3208 ext/Time/HiRes/HiRes.t
3210 lib/Memoize/t/expmod_t.t
3211 lib/Memoize/t/speed.t
3213 In case of failure please try running them manually, for example
3215 ./perl -Ilib ext/Time/HiRes/HiRes.t
3219 ../lib/Math/Trig.t 26 1 3.85% 25
3220 ../lib/warnings.t 470 1 0.21% 429
3222 The Trig.t failure is caused by the slighly differing (from IEEE)
3223 floating point implementation of UNICOS. The warnings.t failure is
3224 also related: the test assumes a certain floating point output format;
3225 this assumption fails in UNICOS.
3233 During Configure, the test
3235 Guessing which symbols your C compiler and preprocessor define...
3237 will probably fail with error messages like
3239 CC-20 cc: ERROR File = try.c, Line = 3
3240 The identifier "bad" is undefined.
3242 bad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K
3245 CC-65 cc: ERROR File = try.c, Line = 3
3246 A semicolon is expected at this point.
3248 This is caused by a bug in the awk utility of UNICOS/mk. You can ignore
3249 the error, but it does cause a slight problem: you cannot fully
3250 benefit from the h2ph utility (see L<h2ph>) that can be used to
3251 convert C headers to Perl libraries, mainly used to be able to access
3252 from Perl the constants defined using C preprocessor, cpp. Because of
3253 the above error, parts of the converted headers will be invisible.
3254 Luckily, these days the need for h2ph is rare.
3258 If building Perl with interpreter threads (ithreads), the
3259 getgrent(), getgrnam(), and getgrgid() functions cannot return the
3260 list of the group members due to a bug in the multithreaded support of
3261 UNICOS/mk. What this means is that in list context the functions will
3262 return only three values, not four.
3268 There are a few known test failures, see L<perluts> (README.uts).
3270 =head2 VOS (Stratus)
3272 When Perl is built using the native build process on VOS Release
3273 14.5.0 and GNU C++/GNU Tools 2.0.1, all attempted tests either
3274 pass or result in TODO (ignored) failures.
3278 There should be no reported test failures with a default configuration,
3279 though there are a number of tests marked TODO that point to areas
3280 needing further debugging and/or porting work.
3284 In multi-CPU boxes, there are some problems with the I/O buffering:
3285 some output may appear twice.
3287 =head2 XML::Parser not working
3289 Use XML::Parser 2.31 or later.
3291 =head2 z/OS (OS/390)
3293 z/OS has rather many test failures but the situation is actually
3294 better than it was in 5.6.0; it's just that so many new modules and
3295 tests have been added.
3297 Failed Test Stat Wstat Total Fail Failed List of Failed
3298 ---------------------------------------------------------------------------
3299 ../ext/Data/Dumper/t/dumper.t 357 8 2.24% 311 314 325 327
3301 ../ext/IO/lib/IO/t/io_unix.t 5 4 80.00% 2-5
3302 ../ext/Storable/t/downgrade.t 12 3072 169 12 7.10% 14-15 46-47 78-79
3304 ../lib/ExtUtils/t/Constant.t 121 30976 48 48 100.00% 1-48
3305 ../lib/ExtUtils/t/Embed.t 9 9 100.00% 1-9
3306 op/pat.t 910 7 0.77% 665 776 785 832-
3308 op/sprintf.t 224 3 1.34% 98 100 136
3309 op/tr.t 97 5 5.15% 63 71-74
3310 uni/fold.t 780 6 0.77% 61 169 196 661
3313 The failures in dumper.t and downgrade.t are problems in the tests,
3314 those in io_unix and sprintf are problems in the USS (UDP sockets
3315 and printf formats). The pat, tr, and fold failures are genuine Perl
3316 problems caused by EBCDIC (and in the pat and fold cases, combining
3317 that with Unicode). The Constant and Embed are probably problems
3318 in the tests (since they test Perl's ability to build extensions,
3319 and that seems to be working reasonably well.)
3321 =head2 Localising Tied Arrays and Hashes Is Broken
3325 doesn't work as one would expect: the old value is restored
3326 incorrectly. This will be changed in a future release, but we don't
3327 know yet what the new semantics will exactly be. In any case, the
3328 change will break existing code that relies on the current
3329 (ill-defined) semantics, so just avoid doing this in general.
3331 =head2 Self-tying Problems
3333 Self-tying of arrays and hashes is broken in rather deep and
3334 hard-to-fix ways. As a stop-gap measure to avoid people from getting
3335 frustrated at the mysterious results (core dumps, most often), it is
3336 forbidden for now (you will get a fatal error even from an attempt).
3338 A change to self-tying of globs has caused them to be recursively
3339 referenced (see: L<perlobj/"Two-Phased Garbage Collection">). You
3340 will now need an explicit untie to destroy a self-tied glob. This
3341 behaviour may be fixed at a later date.
3343 Self-tying of scalars and IO thingies works.
3345 =head2 Tied/Magical Array/Hash Elements Do Not Autovivify
3347 For normal arrays C<$foo = \$bar[1]> will assign C<undef> to
3348 C<$bar[1]> (assuming that it didn't exist before), but for
3349 tied/magical arrays and hashes such autovivification does not happen
3350 because there is currently no way to catch the reference creation.
3351 The same problem affects slicing over non-existent indices/keys of
3352 a tied/magical array/hash.
3354 =head2 Building Extensions Can Fail Because Of Largefiles
3356 Some extensions like mod_perl are known to have issues with
3357 `largefiles', a change brought by Perl 5.6.0 in which file offsets
3358 default to 64 bits wide, where supported. Modules may fail to compile
3359 at all, or they may compile and work incorrectly. Currently, there
3360 is no good solution for the problem, but Configure now provides
3361 appropriate non-largefile ccflags, ldflags, libswanted, and libs
3362 in the %Config hash (e.g., $Config{ccflags_nolargefiles}) so the
3363 extensions that are having problems can try configuring themselves
3364 without the largefileness. This is admittedly not a clean solution,
3365 and the solution may not even work at all. One potential failure is
3366 whether one can (or, if one can, whether it's a good idea to) link
3367 together at all binaries with different ideas about file offsets;
3368 all this is platform-dependent.
3370 =head2 Unicode Support on EBCDIC Still Spotty
3372 Though mostly working, Unicode support still has problem spots on
3373 EBCDIC platforms. One such known spot are the C<\p{}> and C<\P{}>
3374 regular expression constructs for code points less than 256: the
3375 C<pP> are testing for Unicode code points, not knowing about EBCDIC.
3377 =head2 The Compiler Suite Is Still Very Experimental
3379 The compiler suite is slowly getting better but it continues to be
3380 highly experimental. Use in production environments is discouraged.
3382 =head2 The Long Double Support Is Still Experimental
3384 The ability to configure Perl's numbers to use "long doubles",
3385 floating point numbers of hopefully better accuracy, is still
3386 experimental. The implementations of long doubles are not yet
3387 widespread and the existing implementations are not quite mature
3388 or standardised, therefore trying to support them is a rare
3389 and moving target. The gain of more precision may also be offset
3390 by slowdown in computations (more bits to move around, and the
3391 operations are more likely to be executed by less optimised
3394 =head2 Seen In Perl 5.7 But Gone Now
3396 C<Time::Piece> (previously known as C<Time::Object>) was removed
3397 because it was felt that it didn't have enough value in it to be a
3398 core module. It is still a useful module, though, and is available
3401 Perl 5.8 unfortunately does not build anymore on AmigaOS; this broke
3402 accidentally at some point. Since there are not that many Amiga
3403 developers available, we could not get this fixed and tested in time
3404 for 5.8.0. Perl 5.6.1 still works for AmigaOS (as does the the 5.7.2
3405 development release).
3407 =head1 Reporting Bugs
3409 If you find what you think is a bug, you might check the articles
3410 recently posted to the comp.lang.perl.misc newsgroup and the perl
3411 bug database at http://bugs.perl.org/ . There may also be
3412 information at http://www.perl.com/ , the Perl Home Page.
3414 If you believe you have an unreported bug, please run the B<perlbug>
3415 program included with your release. Be sure to trim your bug down
3416 to a tiny but sufficient test case. Your bug report, along with the
3417 output of C<perl -V>, will be sent off to perlbug@perl.org to be
3418 analysed by the Perl porting team.
3422 The F<Changes> file for exhaustive details on what changed.
3424 The F<INSTALL> file for how to build Perl.
3426 The F<README> file for general stuff.
3428 The F<Artistic> and F<Copying> files for copyright information.
3432 Written by Jarkko Hietaniemi <F<jhi@iki.fi>>.