3 perldelta - what is new for perl v5.8.0
7 This document describes differences between the 5.6.0 release
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
14 If you are upgrading from Perl 5.005_03, you might also want
15 to read L<perl56delta>.
17 =head1 Highlights In 5.8.0
23 Better Unicode support
27 New Thread Implementation
35 Better Numeric Accuracy
43 More Extensive Regression Testing
47 =head1 Incompatible Changes
49 =head2 Binary Incompatibility
51 B<Perl 5.8 is not binary compatible with earlier releases of Perl.>
53 B<You have to recompile your XS modules.>
55 (Pure Perl modules should continue to work.)
57 The major reason for the discontinuity is the new IO architecture
58 called PerlIO. PerlIO is the default configuration because without
59 it many new features of Perl 5.8 cannot be used. In other words:
60 you just have to recompile your modules containing XS code, sorry
63 In future releases of Perl, non-PerlIO aware XS modules may become
64 completely unsupported. This shouldn't be too difficult for module
65 authors, however: PerlIO has been designed as a drop-in replacement
66 (at the source code level) for the stdio interface.
68 Depending on your platform, there are also other reasons why
69 we decided to break binary compatibility, please read on.
71 =head2 64-bit platforms and malloc
73 If your pointers are 64 bits wide, the Perl malloc is no longer being
74 used because it does not work well with 8-byte pointers. Also,
75 usually the system mallocs on such platforms are much better optimized
76 for such large memory models than the Perl malloc. Some memory-hungry
77 Perl applications like the PDL don't work well with Perl's malloc.
78 Finally, other applications than Perl (such as mod_perl) tend to prefer
79 the system malloc. Such platforms include Alpha and 64-bit HPPA,
82 =head2 AIX Dynaloading
84 The AIX dynaloading now uses in AIX releases 4.3 and newer the native
85 dlopen interface of AIX instead of the old emulated interface. This
86 change will probably break backward compatibility with compiled
87 modules. The change was made to make Perl more compliant with other
88 applications like mod_perl which are using the AIX native interface.
90 =head2 Attributes for C<my> variables now handled at run-time.
92 The C<my EXPR : ATTRS> syntax now applies variable attributes at
93 run-time. (Subroutine and C<our> variables still get attributes applied
94 at compile-time.) See L<attributes> for additional details. In particular,
95 however, this allows variable attributes to be useful for C<tie> interfaces,
96 which was a deficiency of earlier releases. Note that the new semantics
97 doesn't work with the Attribute::Handlers module (as of version 0.76).
99 =head2 Socket Extension Dynamic in VMS
101 The Socket extension is now dynamically loaded instead of being
102 statically built in. This may or may not be a problem with ancient
103 TCP/IP stacks of VMS: we do not know since we weren't able to test
104 Perl in such configurations.
106 =head2 IEEE-format Floating Point Default on OpenVMS Alpha
108 Perl now uses IEEE format (T_FLOAT) as the default internal floating
109 point format on OpenVMS Alpha, potentially breaking binary compatibility
110 with external libraries or existing data. G_FLOAT is still available as
111 a configuration option. The default on VAX (D_FLOAT) has not changed.
113 =head2 New Unicode Properties
115 Unicode I<scripts> are now supported. Scripts are similar to (and superior
116 to) Unicode I<blocks>. The difference between scripts and blocks is that
117 scripts are the glyphs used by a language or a group of languages, while
118 the blocks are more artificial groupings of (mostly) 256 characters based
119 on the Unicode numbering.
121 In general, scripts are more inclusive, but not universally so. For
122 example, while the script C<Latin> includes all the Latin characters and
123 their various diacritic-adorned versions, it does not include the various
124 punctuation or digits (since they are not solely C<Latin>).
126 A number of other properties are now supported, including C<\p{L&}>,
127 C<\p{Any}> C<\p{Assigned}>, C<\p{Unassigned}>, C<\p{Blank}> and
128 C<\p{SpacePerl}> (along with their C<\P{...}> versions, of course).
129 See L<perlunicode> for details, and more additions.
131 The C<In> or C<Is> prefix to names used with the C<\p{...}> and C<\P{...}>
132 are now almost always optional. The only exception is that a C<In> prefix
133 is required to signify a Unicode block when a block name conflicts with a
134 script name. For example, C<\p{Tibetan}> refers to the script, while
135 C<\p{InTibetan}> refers to the block. When there is no name conflict, you
136 can omit the C<In> from the block name (e.g. C<\p{BraillePatterns}>), but
137 to be safe, it's probably best to always use the C<In>).
139 =head2 REF(...) Instead Of SCALAR(...)
141 A reference to a reference now stringifies as "REF(0x81485ec)" instead
142 of "SCALAR(0x81485ec)" in order to be more consistent with the return
145 =head2 pack/unpack D/F recycled
147 The undocumented pack/unpack template letters D/F have been recycled
148 for better use: now they stand for long double (if supported by the
149 platform) and NV (Perl internal floating point type). (They used
150 to be aliases for d/f, but you never knew that.)
158 The semantics of bless(REF, REF) were unclear and until someone proves
159 it to make some sense, it is forbidden.
163 The obsolete chat2 library that should never have been allowed
164 to escape the laboratory has been decommissioned.
168 The builtin dump() function has probably outlived most of its
169 usefulness. The core-dumping functionality will remain in future
170 available as an explicit call to C<CORE::dump()>, but in future
171 releases the behaviour of an unqualified C<dump()> call may change.
175 The very dusty examples in the eg/ directory have been removed.
176 Suggestions for new shiny examples welcome but the main issue is that
177 the examples need to be documented, tested and (most importantly)
182 The (bogus) escape sequences \8 and \9 now give an optional warning
183 ("Unrecognized escape passed through"). There is no need to \-escape
188 The list of filenames from glob() (or <...>) is now by default sorted
189 alphabetically to be csh-compliant (which is what happened before
190 in most UNIX platforms). (bsd_glob() does still sort platform
191 natively, ASCII or EBCDIC, unless GLOB_ALPHASORT is specified.)
195 Spurious syntax errors generated in certain situations, when glob()
196 caused File::Glob to be loaded for the first time, have been fixed.
200 Although "you shouldn't do that", it was possible to write code that
201 depends on Perl's hashed key order (Data::Dumper does this). The new
202 algorithm "One-at-a-Time" produces a different hashed key order.
203 More details are in L</"Performance Enhancements">.
207 lstat(FILEHANDLE) now gives a warning because the operation makes no sense.
208 In future releases this may become a fatal error.
212 The C<package;> syntax (C<package> without an argument) has been
213 deprecated. Its semantics were never that clear and its
214 implementation even less so. If you have used that feature to
215 disallow all but fully qualified variables, C<use strict;> instead.
219 The unimplemented POSIX regex features [[.cc.]] and [[=c=]] are still
220 recognised but now cause fatal errors. The previous behaviour of
221 ignoring them by default and warning if requested was unacceptable
222 since it, in a way, falsely promised that the features could be used.
226 The current user-visible implementation of pseudo-hashes (the weird
227 use of the first array element) is deprecated starting from Perl 5.8.0
228 and will be removed in Perl 5.10.0, and the feature will be
229 implemented differently. Not only is the current interface rather
230 ugly, but the current implementation slows down normal array and hash
231 use quite noticeably. The C<fields> pragma interface will remain
232 available. The I<restricted hashes> interface is expected to
233 be the replacement interface (see L<Hash::Util>).
237 The syntaxes C<< @a->[...] >> and C<< %h->{...} >> have now been deprecated.
241 After years of trying, suidperl is considered to be too complex to
242 ever be considered truly secure. The suidperl functionality is likely
243 to be removed in a future release.
247 The 5.005 threads model (module C<Thread>) is deprecated and expected
248 to be removed in Perl 5.10. Multithreaded code should be migrated to
249 the new ithreads model (see L<threads>, L<threads::shared> and
254 The long deprecated uppercase aliases for the string comparison
255 operators (EQ, NE, LT, LE, GE, GT) have now been removed.
259 The tr///C and tr///U features have been removed and will not return;
260 the interface was a mistake. Sorry about that. For similar
261 functionality, see pack('U0', ...) and pack('C0', ...).
265 Earlier Perls treated "sub foo (@bar)" as equivalent to "sub foo (@)".
266 The prototypes are now checked better at compile-time for invalid
267 syntax. An optional warning is generated ("Illegal character in
268 prototype...") but this may be upgraded to a fatal error in a future
273 The existing behaviour when localising tied arrays and hashes is wrong,
274 and will be changed in a future release, so do not rely on the existing
275 behaviour. See L<"Localising Tied Arrays and Hashes Is Broken">.
279 =head1 Core Enhancements
281 =head2 PerlIO is Now The Default
287 IO is now by default done via PerlIO rather than system's "stdio".
288 PerlIO allows "layers" to be "pushed" onto a file handle to alter the
289 handle's behaviour. Layers can be specified at open time via 3-arg
292 open($fh,'>:crlf :utf8', $path) || ...
294 or on already opened handles via extended C<binmode>:
296 binmode($fh,':encoding(iso-8859-7)');
298 The built-in layers are: unix (low level read/write), stdio (as in
299 previous Perls), perlio (re-implementation of stdio buffering in a
300 portable manner), crlf (does CRLF <=> "\n" translation as on Win32,
301 but available on any platform). A mmap layer may be available if
302 platform supports it (mostly UNIXes).
304 Layers to be applied by default may be specified via the 'open' pragma.
306 See L</"Installation and Configuration Improvements"> for the effects
307 of PerlIO on your architecture name.
311 File handles can be marked as accepting Perl's internal encoding of Unicode
312 (UTF-8 or UTF-EBCDIC depending on platform) by a pseudo layer ":utf8" :
314 open($fh,">:utf8","Uni.txt");
316 Note for EBCDIC users: the pseudo layer ":utf8" is erroneously named
317 for you since it's not UTF-8 what you will be getting but instead
318 UTF-EBCDIC. See L<perlunicode>, L<utf8>, and
319 http://www.unicode.org/unicode/reports/tr16/ for more information.
320 In future releases this naming may change.
324 File handles can translate character encodings from/to Perl's internal
325 Unicode form on read/write via the ":encoding()" layer.
329 File handles can be opened to "in memory" files held in Perl scalars via:
331 open($fh,'>', \$variable) || ...
335 Anonymous temporary files are available without need to
336 'use FileHandle' or other module via
338 open($fh,"+>", undef) || ...
340 That is a literal undef, not an undefined value.
344 The list form of C<open> is now implemented for pipes (at least on UNIX):
346 open($fh,"-|", 'cat', '/etc/motd')
348 creates a pipe, and runs the equivalent of exec('cat', '/etc/motd') in
353 If your locale environment variables (LANGUAGE, LC_ALL, LC_CTYPE, LANG)
354 contain the strings 'UTF-8' or 'UTF8' (case-insensitive matching),
355 the default encoding of your STDIN, STDOUT, and STDERR, and of
356 B<any subsequent file open>, is UTF-8.
360 =head2 Restricted Hashes
362 A restricted hash is restricted to a certain set of keys, no keys
363 outside the set can be added. Also individual keys can be restricted
364 so that the key cannot be deleted and the value cannot be changed.
365 No new syntax is involved: the Hash::Util module is the interface.
369 Perl used to be fragile in that signals arriving at inopportune moments
370 could corrupt Perl's internal state. Now Perl postpones handling of
371 signals until it's safe (between opcodes).
373 This change may have surprising side effects because signals no longer
374 interrupt Perl instantly. Perl will now first finish whatever it was
375 doing, like finishing an internal operation (like sort()) or an
376 external operation (like an I/O operation), and only then look at any
377 arrived signals (and before starting the next operation). No more corrupt
378 internal state since the current operation is always finished first,
379 but the signal may take more time to get heard. Note that breaking
380 out from potentially blocking operations should still work, though.
382 =head2 Unicode Overhaul
384 Unicode in general should be now much more usable than in Perl 5.6.0
385 (or even in 5.6.1). Unicode can be used in hash keys, Unicode in
386 regular expressions should work now, Unicode in tr/// should work now,
387 Unicode in I/O should work now. See L<perluniintro> for introduction
388 and L<perlunicode> for details.
394 The Unicode Character Database coming with Perl has been upgraded
395 to Unicode 3.2.0. For more information, see http://www.unicode.org/ .
399 For developers interested in enhancing Perl's Unicode capabilities:
400 almost all the UCD files are included with the Perl distribution in
401 the F<lib/unicore> subdirectory. The most notable omission, for space
402 considerations, is the Unihan database.
406 The properties \p{Blank} and \p{SpacePerl} have been added. "Blank" is like
407 C isblank(), that is, it contains only "horizontal whitespace" (the space
408 character is, the newline isn't), and the "SpacePerl" is the Unicode
409 equivalent of C<\s> (\p{Space} isn't, since that includes the vertical
410 tabulator character, whereas C<\s> doesn't.)
412 See "New Unicode Properties" earlier in this document for additional
413 information on changes with Unicode properties.
417 =head2 Understanding of Numbers
419 In general a lot of fixing has happened in the area of Perl's
420 understanding of numbers, both integer and floating point. Since in
421 many systems the standard number parsing functions like C<strtoul()>
422 and C<atof()> seem to have bugs, Perl tries to work around their
423 deficiencies. This results hopefully in more accurate numbers.
425 Perl now tries internally to use integer values in numeric conversions
426 and basic arithmetics (+ - * /) if the arguments are integers, and
427 tries also to keep the results stored internally as integers.
428 This change leads to often slightly faster and always less lossy
429 arithmetics. (Previously Perl always preferred floating point numbers
432 =head2 Miscellaneous Changes
438 AUTOLOAD is now lvaluable, meaning that you can add the :lvalue attribute
439 to AUTOLOAD subroutines and you can assign to the AUTOLOAD return value.
443 The $Config{byteorder} (and corresponding BYTEORDER in config.h) was
444 previously wrong in platforms if sizeof(long) was 4, but sizeof(IV)
445 was 8. The byteorder was only sizeof(long) bytes long (1234 or 4321),
446 but now it is correctly sizeof(IV) bytes long, (12345678 or 87654321).
447 (This problem didn't affect Windows platforms.)
449 Also, $Config{byteorder} is now computed dynamically--this is more
450 robust with "fat binaries" where an executable image contains binaries
451 for more than one binary platform, and when cross-compiling.
455 C<perl -d:Module=arg,arg,arg> now works (previously one couldn't pass
456 in multiple arguments.)
460 The builtin dump() now gives an optional warning
461 C<dump() better written as CORE::dump()>,
462 meaning that by default C<dump(...)> is resolved as the builtin
463 dump() which dumps core and aborts, not as (possibly) user-defined
464 C<sub dump>. To call the latter, qualify the call as C<&dump(...)>.
465 (The whole dump() feature is to considered deprecated, and possibly
466 removed/changed in future releases.)
470 chomp() and chop() are now overridable. Note, however, that their
471 prototype (as given by C<prototype("CORE::chomp")> is undefined,
472 because it cannot be expressed and therefore one cannot really write
473 replacements to override these builtins.
477 END blocks are now run even if you exit/die in a BEGIN block.
478 Internally, the execution of END blocks is now controlled by
479 PL_exit_flags & PERL_EXIT_DESTRUCT_END. This enables the new
480 behaviour for Perl embedders. This will default in 5.10. See
485 Formats now support zero-padded decimal fields.
489 Lvalue subroutines can now return C<undef> in list context.
490 However, the lvalue subroutine feature still remains experimental.
494 A lost warning "Can't declare ... dereference in my" has been
495 restored (Perl had it earlier but it became lost in later releases.)
499 A new special regular expression variable has been introduced:
500 C<$^N>, which contains the most-recently closed group (submatch).
504 C<no Module;> now works even if there is no "sub unimport" in the Module.
508 The numerical comparison operators return C<undef> if either operand
509 is a NaN. Previously the behaviour was unspecified.
513 The following builtin functions are now overridable: each(), keys(),
514 pop(), push(), shift(), splice(), unshift().
518 C<pack() / unpack()> now can group template letters with C<()> and then
519 apply repetition/count modifiers on the groups.
523 C<pack() / unpack()> can now process the Perl internal numeric types:
524 IVs, UVs, NVs-- and also long doubles, if supported by the platform.
525 The template letters are C<j>, C<J>, C<F>, and C<D>.
529 C<pack('U0a*', ...)> can now be used to force a string to UTF8.
533 my __PACKAGE__ $obj now works.
537 POSIX::sleep() now returns the number of I<unslept> seconds
538 (as the POSIX standard says), as opposed to CORE::sleep() which
539 returns the number of slept seconds.
543 The printf() and sprintf() now support parameter reordering using the
544 C<%\d+\$> and C<*\d+\$> syntaxes. For example
546 print "%2\$s %1\$s\n", "foo", "bar";
548 will print "bar foo\n". This feature helps in writing
549 internationalised software, and in general when the order
550 of the parameters can vary.
554 prototype(\&) is now available.
558 prototype(\[$@%&]) is now available to implicitly create references
559 (useful for example if you want to emulate the tie() interface).
563 A new command-line option, C<-t> is available. It is the
564 little brother of C<-T>: instead of dying on taint violations,
565 lexical warnings are given. B<This is only meant as a temporary
566 debugging aid while securing the code of old legacy applications.
567 This is not a substitute for -T.>
571 In other taint news, the C<exec LIST> and C<system LIST> have now been
572 considered too risky (think C<exec @ARGV>: it can start any program
573 with any arguments), and now the said forms cause a warning.
574 You should carefully launder the arguments to guarantee their
575 validity. In future releases of Perl the forms will become fatal
576 errors so consider starting laundering now.
580 Tied hash interfaces are now required to have the EXISTS and DELETE
581 methods (either own or inherited).
585 If tr/// is just counting characters, it doesn't attempt to
590 untie() will now call an UNTIE() hook if it exists. See L<perltie>
595 L<utime> now supports C<utime undef, undef, @files> to change the
596 file timestamps to the current time.
600 The rules for allowing underscores (underbars) in numeric constants
601 have been relaxed and simplified: now you can have an underscore
602 simply B<between digits>.
606 Rather than relying on C's argv[0] (which may not contain a full pathname)
607 where possible $^X is now set by asking the operating system.
608 (eg by reading F</proc/self/exe> on Linux, F</proc/curproc/file> on FreeBSD)
612 A new variable, C<${^TAINT}>, indicates whether taint mode is enabled.
616 You can now override the readline() builtin, and this overrides also
617 the <FILEHANDLE> angle bracket operator.
621 The command-line options -s and -F are now recognized on the shebang
626 Use of the C</c> match modifier without an accompanying C</g> modifier
627 elicits a new warning: C<Use of /c modifier is meaningless without /g>.
629 Use of C</c> in substitutions, even with C</g>, elicits
630 C<Use of /c modifier is meaningless in s///>.
632 Use of C</g> with C<split> elicits C<Use of /g modifier is meaningless
637 =head1 Modules and Pragmata
639 =head2 New Modules and Pragmata
645 C<Attribute::Handlers> allows a class to define attribute handlers.
648 use Attribute::Handlers;
649 sub Wolf :ATTR(SCALAR) { print "howl!\n" }
651 # later, in some package using or inheriting from MyPack...
653 my MyPack $Fluffy : Wolf; # the attribute handler Wolf will be called
655 Both variables and routines can have attribute handlers. Handlers can
656 be specific to type (SCALAR, ARRAY, HASH, or CODE), or specific to the
657 exact compilation phase (BEGIN, CHECK, INIT, or END).
658 See L<Attribute::Handlers>.
662 C<B::Concise>, by Stephen McCamant, is a new compiler backend for
663 walking the Perl syntax tree, printing concise info about ops.
664 The output is highly customisable. See L<B::Concise>.
668 The new bignum, bigint, and bigrat pragmas, by Tels, implement
669 transparent bignum support (using the Math::BigInt, Math::BigFloat,
670 and Math::BigRat backends).
674 C<Class::ISA>, by Sean Burke, is a module for reporting the search
675 path for a class's ISA tree. See L<Class::ISA>.
679 C<Cwd> now has a split personality: if possible, an XS extension is
680 used, (this will hopefully be faster, more secure, and more robust)
681 but if not possible, the familiar Perl implementation is used.
685 C<Devel::PPPort>, originally by Kenneth Albanowski and now
686 maintained by Paul Marquess, has been added. It is primarily used
687 by C<h2xs> to enhance portability of XS modules between different
688 versions of Perl. See L<Devel::PPPort>.
692 C<Digest>, frontend module for calculating digests (checksums), from
693 Gisle Aas, has been added. See L<Digest>.
697 C<Digest::MD5> for calculating MD5 digests (checksums) as defined in
698 RFC 1321, from Gisle Aas, has been added. See L<Digest::MD5>.
700 use Digest::MD5 'md5_hex';
702 $digest = md5_hex("Thirsty Camel");
704 print $digest, "\n"; # 01d19d9d2045e005c3f1b80e8b164de1
706 NOTE: the C<MD5> backward compatibility module is deliberately not
707 included since its further use is discouraged.
711 C<Encode>, originally by Nick Ing-Simmons and now maintained by Dan
712 Kogai, provides a mechanism to translate between different character
713 encodings. Support for Unicode, ISO-8859-1, and ASCII are compiled in
714 to the module. Several other encodings (like the rest of the
715 ISO-8859, CP*/Win*, Mac, KOI8-R, three variants EBCDIC, Chinese,
716 Japanese, and Korean encodings) are included and can be loaded at
717 runtime. (For space considerations, the largest Chinese encodings
718 have been separated into their own CPAN module, Encode::HanExtra,
719 which Encode will use if available). See L<Encode>.
721 Any encoding supported by Encode module is also available to the
722 ":encoding()" layer if PerlIO is used.
726 C<Hash::Util> is the interface to the new I<restricted hashes>
727 feature. (Implemented by Jeffrey Friedl, Nick Ing-Simmons, and
728 Michael Schwern.) See L<Hash::Util>.
732 C<I18N::Langinfo> can be used to query locale information.
733 See L<I18N::Langinfo>.
737 C<I18N::LangTags>, by Sean Burke, has functions for dealing with
738 RFC3066-style language tags. See L<I18N::LangTags>.
742 C<ExtUtils::Constant>, by Nicholas Clark, is a new tool for extension
743 writers for generating XS code to import C header constants.
744 See L<ExtUtils::Constant>.
748 C<Filter::Simple>, by Damian Conway, is an easy-to-use frontend to
749 Filter::Util::Call. See L<Filter::Simple>.
755 use Filter::Simple sub {
756 while (my ($from, $to) = splice @_, 0, 2) {
765 use MyFilter qr/red/ => 'green';
767 print "red\n"; # this code is filtered, will print "green\n"
768 print "bored\n"; # this code is filtered, will print "bogreen\n"
772 print "red\n"; # this code is not filtered, will print "red\n"
776 C<File::Temp>, by Tim Jenness, allows one to create temporary files and
777 directories in an easy, portable, and secure way. See L<File::Temp>.
781 C<Filter::Util::Call>, by Paul Marquess, provides you with the
782 framework to write I<source filters> in Perl. For most uses, the
783 frontend Filter::Simple is to be preferred. See L<Filter::Util::Call>.
787 C<if>, by Ilya Zakharevich, is a new pragma for conditional inclusion
792 L<libnet>, by Graham Barr, is a collection of perl5 modules related
793 to network programming. See L<Net::FTP>, L<Net::NNTP>, L<Net::Ping>
794 (not part of libnet, but related), L<Net::POP3>, L<Net::SMTP>,
797 Perl installation leaves libnet unconfigured; use F<libnetcfg>
802 C<List::Util>, by Graham Barr, is a selection of general-utility
803 list subroutines, such as sum(), min(), first(), and shuffle().
808 C<Locale::Constants>, C<Locale::Country>, C<Locale::Currency>
809 C<Locale::Language>, and L<Locale::Script>, by Neil Bowers, have
810 been added. They provide the codes for various locale standards, such
811 as "fr" for France, "usd" for US Dollar, and "ja" for Japanese.
815 $country = code2country('jp'); # $country gets 'Japan'
816 $code = country2code('Norway'); # $code gets 'no'
818 See L<Locale::Constants>, L<Locale::Country>, L<Locale::Currency>,
819 and L<Locale::Language>.
823 C<Locale::Maketext>, by Sean Burke, is a localization framework. See
824 L<Locale::Maketext>, and L<Locale::Maketext::TPJ13>. The latter is an
825 article about software localization, originally published in The Perl
826 Journal #13, and republished here with kind permission.
830 C<Math::BigRat> for big rational numbers, to accompany Math::BigInt and
831 Math::BigFloat, from Tels. See L<Math::BigRat>.
835 C<Memoize> can make your functions faster by trading space for time,
836 from Mark-Jason Dominus. See L<Memoize>.
840 C<MIME::Base64>, by Gisle Aas, allows you to encode data in base64,
841 as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail
846 $encoded = encode_base64('Aladdin:open sesame');
847 $decoded = decode_base64($encoded);
849 print $encoded, "\n"; # "QWxhZGRpbjpvcGVuIHNlc2FtZQ=="
855 C<MIME::QuotedPrint>, by Gisle Aas, allows you to encode data
856 in quoted-printable encoding, as defined in RFC 2045 - I<MIME
857 (Multipurpose Internet Mail Extensions)>.
859 use MIME::QuotedPrint;
861 $encoded = encode_qp("Smiley in Unicode: \x{263a}");
862 $decoded = decode_qp($encoded);
864 print $encoded, "\n"; # "Smiley in Unicode: =263A"
866 MIME::QuotedPrint has been enhanced to provide the basic methods
867 necessary to use it with PerlIO::Via as in :
869 use MIME::QuotedPrint;
870 open($fh,">Via(MIME::QuotedPrint)",$path);
872 See L<MIME::QuotedPrint>.
876 C<NEXT>, by Damian Conway, is a pseudo-class for method redispatch.
881 C<open> is a new pragma for setting the default I/O disciplines
886 C<PerlIO::Scalar>, by Nick Ing-Simmons, provides the implementation
887 of IO to "in memory" Perl scalars as discussed above. It also serves
888 as an example of a loadable PerlIO layer. Other future possibilities
889 include PerlIO::Array and PerlIO::Code. See L<PerlIO::Scalar>.
893 C<PerlIO::Via>, by Nick Ing-Simmons, acts as a PerlIO layer and wraps
894 PerlIO layer functionality provided by a class (typically implemented
897 use MIME::QuotedPrint;
898 open($fh,">Via(MIME::QuotedPrint)",$path);
900 This will automatically convert everything output to C<$fh>
901 to Quoted-Printable. See L<PerlIO::Via>.
905 C<Pod::ParseLink>, by Russ Allbery, has been added,
906 to parse LZ<><> links in pods as described in the new
911 C<Pod::Text::Overstrike>, by Joe Smith, has been added.
912 It converts POD data to formatted overstrike text.
913 See L<Pod::Text::Overstrike>.
917 C<Scalar::Util> is a selection of general-utility scalar subroutines,
918 such as blessed(), reftype(), and tainted(). See L<Scalar::Util>.
922 C<sort> is a new pragma for controlling the behaviour of sort().
926 C<Storable> gives persistence to Perl data structures by allowing the
927 storage and retrieval of Perl data to and from files in a fast and
928 compact binary format. Because in effect Storable does serialisation
929 of Perl data structues, with it you can also clone deep, hierarchical
930 datastructures. Storable was originally created by Raphael Manfredi,
931 but it is now maintained by Abhijit Menon-Sen. Storable has been
932 enhanced to understand the two new hash features, Unicode keys and
933 restricted hashes. See L<Storable>.
937 C<Switch>, by Damian Conway, has been added. Just by saying
941 you have C<switch> and C<case> available in Perl.
947 case 1 { print "number 1" }
948 case "a" { print "string a" }
949 case [1..10,42] { print "number in list" }
950 case (@array) { print "number in list" }
951 case /\w+/ { print "pattern" }
952 case qr/\w+/ { print "pattern" }
953 case (%hash) { print "entry in hash" }
954 case (\%hash) { print "entry in hash" }
955 case (\&sub) { print "arg to subroutine" }
956 else { print "previous case not true" }
963 C<Test::More>, by Michael Schwern, is yet another framework for writing
964 test scripts, more extensive than Test::Simple. See L<Test::More>.
968 C<Test::Simple>, by Michael Schwern, has basic utilities for writing
969 tests. See L<Test::Simple>.
973 C<Text::Balanced>, by Damian Conway, has been added, for extracting
974 delimited text sequences from strings.
976 use Text::Balanced 'extract_delimited';
978 ($a, $b) = extract_delimited("'never say never', he never said", "'", '');
980 $a will be "'never say never'", $b will be ', he never said'.
982 In addition to extract_delimited(), there are also extract_bracketed(),
983 extract_quotelike(), extract_codeblock(), extract_variable(),
984 extract_tagged(), extract_multiple(), gen_delimited_pat(), and
985 gen_extract_tagged(). With these, you can implement rather advanced
986 parsing algorithms. See L<Text::Balanced>.
990 C<threads>, by Arthur Bergman, is an interface to interpreter threads.
991 Interpreter threads (ithreads) is the new thread model introduced in
992 Perl 5.6 but only available as an internal interface for extension
993 writers (and for Win32 Perl for C<fork()> emulation). See L<threads>,
994 L<threads::shared>, and L<perlthrtut>.
998 C<threads::shared>, by Arthur Bergman, allows data sharing for
999 interpreter threads. In the ithreads model any data sharing between
1000 threads must be explicit, as opposed to the old 5.005 thread model
1001 where data sharing was implicit. See L<threads::shared>.
1005 C<Tie::File>, by Mark-Jason Dominus, associates a Perl array with the
1006 lines of a file. See L<Tie::File>.
1010 C<Tie::Memoize>, by Ilya Zakharevich, provides on-demand loaded hashes.
1011 See L<Tie::Memoize>.
1015 C<Tie::RefHash::Nestable>, by Edward Avis, allows storing hash
1016 references (unlike the standard Tie::RefHash) The module is contained
1017 within Tie::RefHash. See L<Tie::RefHash>.
1021 C<Time::HiRes>, by Douglas E. Wegscheid, provides high resolution
1022 timing (ualarm, usleep, and gettimeofday). See L<Time::HiRes>.
1026 C<Unicode::UCD> offers a querying interface to the Unicode Character
1027 Database. See L<Unicode::UCD>.
1031 C<Unicode::Collate>, by SADAHIRO Tomoyuki, implements the UCA
1032 (Unicode Collation Algorithm) for sorting Unicode strings.
1033 See L<Unicode::Collate>.
1037 C<Unicode::Normalize>, by SADAHIRO Tomoyuki, implements the various
1038 Unicode normalization forms. See L<Unicode::Normalize>.
1042 C<XS::Typemap>, by Tim Jenness, is a test extension that exercises
1043 XS typemaps. Nothing gets installed, but the code is worth studying
1044 for extension writers.
1048 =head2 Updated And Improved Modules and Pragmata
1054 The following independently supported modules have been updated to the
1055 newest versions from CPAN: CGI, CPAN, DB_File, File::Spec, File::Temp,
1056 Getopt::Long, Math::BigFloat, Math::BigInt, the podlators bundle
1057 (Pod::Man, Pod::Text), Pod::LaTeX, Pod::Parser, Storable,
1058 Term::ANSIColor, Test, Text-Tabs+Wrap.
1062 attributes::reftype() now works on tied arguments.
1066 AutoLoader can now be disabled with C<no AutoLoader;>.
1070 B::Deparse has been significantly enhanced. It now can deparse almost
1071 all of the standard test suite (so that the tests still succeed).
1072 There is a make target "test.deparse" for trying this out.
1076 Class::Struct can now define the classes in compile time.
1080 Class::Struct now assigns the array/hash element if the accessor
1081 is called with an array/hash element as the B<sole> argument.
1085 The return value of Cwd::fastcwd() is now tainted.
1089 Data::Dumper now has an option to sort hashes.
1093 Data::Dumper now has an option to dump code references
1098 DB_File now supports newer Berkeley DB versions, among
1103 Devel::Peek now has an interface for the Perl memory statistics
1104 (this works only if you are using perl's malloc, and if you have
1105 compiled with debugging).
1109 The English module can now be used without the infamous performance
1112 use English '-no_match_vars';
1114 (Assuming, of course, that you don't need the troublesome variables
1115 C<$`>, C<$&>, or C<$'>.) Also, introduced C<@LAST_MATCH_START> and
1116 C<@LAST_MATCH_END> English aliases for C<@-> and C<@+>.
1120 ExtUtils::MakeMaker now uses File::Spec internally, which hopefully
1121 leads to better portability.
1125 Fcntl, Socket, and Sys::Syslog have been rewritten to use the
1126 new-style constant dispatch section (see L<ExtUtils::Constant>).
1127 This means that they will be more robust and hopefully faster.
1131 File::Find now chdir()s correctly when chasing symbolic links.
1135 File::Find now has pre- and post-processing callbacks. It also
1136 correctly changes directories when chasing symbolic links. Callbacks
1137 (naughtily) exiting with "next;" instead of "return;" now work.
1141 File::Find is now (again) reentrant. It also has been made
1146 The warnings issued by File::Find now belong to their own category.
1147 You can enable/disable them with C<use/no warnings 'File::Find';>.
1151 File::Glob::glob() renamed to File::Glob::bsd_glob() to avoid
1152 prototype mismatch with CORE::glob().
1156 File::Glob now supports C<GLOB_LIMIT> constant to limit the size of
1157 the returned list of filenames.
1161 IPC::Open3 now allows the use of numeric file descriptors.
1165 IO::Socket now has an atmark() method, which returns true if the socket
1166 is positioned at the out-of-band mark. The method is also exportable
1167 as a sockatmark() function.
1171 IO::Socket::INET has support for the ReusePort option (if your
1172 platform supports it). The Reuse option now has an alias, ReuseAddr.
1173 For clarity, you may want to prefer ReuseAddr.
1177 IO::Socket::INET now supports a value of zero for C<LocalPort>
1178 (usually meaning that the operating system will make one up.)
1182 'use lib' now works identically to @INC. Removing directories
1183 with 'no lib' now works.
1187 Math::BigFloat and Math::BigInt have undergone a full rewrite.
1188 They are now magnitudes faster, and they support various
1189 bignum libraries such as GMP and PARI as their backends.
1193 Math::Complex handles inf, NaN etc., better.
1197 Net::Ping has been considerably enhanced: multihoming is now supported,
1198 Win32 functionality is better, there is now time measuring
1199 functionality (optionally high-resolution using Time::HiRes),
1200 and there is now "external" protocol which uses Net::Ping::External
1201 module which runs your external ping utility and parses the output.
1202 A version of Net::Ping::External is available in CPAN.
1204 Note that some of the Net::Ping tests are disabled when running
1205 under the Perl distribution since one cannot assume one or more
1206 of the following: enabled echo port at localhost, full Internet
1207 connectivity, or sympathetic firewalls. You can set the environment
1208 variable PERL_TEST_Net_Ping to "1" (one) before running the Perl test
1209 suite to enable all the Net::Ping tests.
1213 POSIX::sigaction() is now much more flexible and robust.
1214 You can now install coderef handlers, 'DEFAULT', and 'IGNORE'
1215 handlers, installing new handlers was not atomic.
1219 In Safe, C<%INC> is now localised in a Safe compartment so that
1224 In SDBM_File on dosish platforms, some keys went missing because of
1225 lack of support for files with "holes". A workaround for the problem
1230 In Search::Dict one can now have a pre-processing hook for the
1231 lines being searched.
1235 The Shell module now has an OO interface.
1239 In Sys::Syslog there is now a failover mechanism that will go
1240 through alternative connection mechanisms until the message
1241 is successfully logged.
1245 The Test module has been significantly enhanced.
1249 Time::Local::timelocal() does not handle fractional seconds anymore.
1250 The rationale is that neither does localtime(), and timelocal() and
1251 localtime() are supposed to be inverses of each other.
1255 The vars pragma now supports declaring fully qualified variables.
1256 (Something that C<our()> does not and will not support.)
1260 The C<utf8::> name space (as in the pragma) provides various
1261 Perl-callable functions to provide low level access to Perl's
1262 internal Unicode representation. At the moment only length()
1263 has been implemented.
1267 =head1 Utility Changes
1273 Emacs perl mode (emacs/cperl-mode.el) has been updated to version
1278 F<emacs/e2ctags.pl> is now much faster.
1282 C<enc2xs> is a tool for people adding their own encodings to the
1287 C<h2ph> now supports C trigraphs.
1291 C<h2xs> now produces a template README.
1295 C<h2xs> now uses C<Devel::PPPort> for better portability between
1296 different versions of Perl.
1300 C<h2xs> uses the new L<ExtUtils::Constant|ExtUtils::Constant> module
1301 which will affect newly created extensions that define constants.
1302 Since the new code is more correct (if you have two constants where the
1303 first one is a prefix of the second one, the first constant B<never>
1304 got defined), less lossy (it uses integers for integer constant,
1305 as opposed to the old code that used floating point numbers even for
1306 integer constants), and slightly faster, you might want to consider
1307 regenerating your extension code (the new scheme makes regenerating
1308 easy). L<h2xs> now also supports C trigraphs.
1312 C<libnetcfg> has been added to configure libnet.
1316 C<perlbug> is now much more robust. It also sends the bug report to
1317 perl.org, not perl.com.
1321 C<perlcc> has been rewritten and its user interface (that is,
1322 command line) is much more like that of the UNIX C compiler, cc.
1323 (The perlbc tools has been removed. Use C<perlcc -B> instead.)
1324 B<Note that perlcc is still considered very experimental and
1329 C<perlivp> is a new Installation Verification Procedure utility
1330 for running any time after installing Perl.
1334 C<piconv> is an implementation of the character conversion utility
1335 C<iconv>, demonstrating the new Encode module.
1339 C<pod2html> now allows specifying a cache directory.
1343 C<pod2html> now produces XHTML 1.0.
1347 C<pod2html> now understands POD written using different line endings
1348 (PC-like CRLF versus UNIX-like LF versus MacClassic-like CR).
1352 C<s2p> has been completely rewritten in Perl. (It is in fact a full
1353 implementation of sed in Perl: you can use the sed functionality by
1354 using the C<psed> utility.)
1358 C<xsubpp> now understands POD documentation embedded in the *.xs files.
1362 C<xsubpp> now supports the OUT keyword.
1366 =head1 New Documentation
1372 perl56delta details the changes between the 5.005 release and the
1377 perlclib documents the internal replacements for standard C library
1378 functions. (Interesting only for extension writers and Perl core
1383 perldebtut is a Perl debugging tutorial.
1387 perlebcdic contains considerations for running Perl on EBCDIC platforms.
1391 perlintro is a gentle introduction to Perl.
1395 perliol documents the internals of PerlIO with layers.
1399 perlmodstyle is a style guide for writing modules.
1403 perlnewmod tells about writing and submitting a new module.
1407 perlpacktut is a pack() tutorial.
1411 perlpod has been rewritten to be clearer and to record the best
1412 practices gathered over the years.
1416 perlpodspec is a more formal specification of the pod format,
1417 mainly of interest for writers of pod applications, not to
1418 people writing in pod.
1422 perlretut is a regular expression tutorial.
1426 perlrequick is a regular expressions quick-start guide.
1427 Yes, much quicker than perlretut.
1431 perltodo has been updated.
1435 perltootc has been renamed as perltooc (to not to conflict
1436 with perltoot in filesystems restricted to "8.3" names).
1440 perluniintro is an introduction to using Unicode in Perl.
1441 (perlunicode is more of a detailed reference and background
1446 perlutil explains the command line utilities packaged with the Perl
1451 The following platform-specific documents are available before
1452 the installation as README.I<platform>, and after the installation
1455 perlaix perlamiga perlapollo perlbeos perlbs2000
1456 perlce perlcygwin perldgux perldos perlepoc perlhpux
1457 perlhurd perlmachten perlmacos perlmint perlmpeix
1458 perlnetware perlos2 perlos390 perlplan9 perlqnx perlsolaris
1459 perltru64 perluts perlvmesa perlvms perlvos perlwin32
1461 Eastern Asian Perl users are now welcomed in their own languages:
1462 README.jp (Japanese), README.ko (Korean), README.cn (simplified
1463 Chinese) and README.tw (traditional Chinese), which are written in
1464 normal pod but encoded in EUC-JP, EUC-KR, EUC-CN and Big5. These
1465 will get installed as
1467 perljp perlko perlcn perltw
1473 The documentation for the POSIX-BC platform is called "BS2000", to avoid
1474 confusion with the Perl POSIX module.
1478 The documentation for the WinCE platform is called perlce (README.ce
1479 in the source code kit), to avoid confusion with the perlwin32
1480 documentation on 8.3-restricted filesystems.
1484 =head1 Performance Enhancements
1490 map() could get pathologically slow when the result list it generates
1491 is larger than the source list. The performance has been improved for
1496 sort() has been changed to use primarily mergesort internally as
1497 opposed to the earlier quicksort. For very small lists this may
1498 result in slightly slower sorting times, but in general the speedup
1499 should be at least 20%. Additional bonuses are that the worst case
1500 behaviour of sort() is now better (in computer science terms it now
1501 runs in time O(N log N), as opposed to quicksort's Theta(N**2)
1502 worst-case run time behaviour), and that sort() is now stable
1503 (meaning that elements with identical keys will stay ordered as they
1504 were before the sort). See the C<sort> pragma for information.
1506 The story in more detail: suppose you want to serve yourself a little
1509 @digits = ( 3,1,4,1,5,9 );
1511 A numerical sort of the digits will yield (1,1,3,4,5,9), as expected.
1512 Which C<1> comes first is hard to know, since one C<1> looks pretty
1513 much like any other. You can regard this as totally trivial,
1514 or somewhat profound. However, if you just want to sort the even
1515 digits ahead of the odd ones, then what will
1517 sort { ($a % 2) <=> ($b % 2) } @digits;
1519 yield? The only even digit, C<4>, will come first. But how about
1520 the odd numbers, which all compare equal? With the quicksort algorithm
1521 used to implement Perl 5.6 and earlier, the order of ties is left up
1522 to the sort. So, as you add more and more digits of Pi, the order
1523 in which the sorted even and odd digits appear will change.
1524 and, for sufficiently large slices of Pi, the quicksort algorithm
1525 in Perl 5.8 won't return the same results even if reinvoked with the
1526 same input. The justification for this rests with quicksort's
1527 worst case behavior. If you run
1529 sort { $a <=> $b } ( 1 .. $N , 1 .. $N );
1531 (something you might approximate if you wanted to merge two sorted
1532 arrays using sort), doubling $N doesn't just double the quicksort time,
1533 it I<quadruples> it. Quicksort has a worst case run time that can
1534 grow like N**2, so-called I<quadratic> behaviour, and it can happen
1535 on patterns that may well arise in normal use. You won't notice this
1536 for small arrays, but you I<will> notice it with larger arrays,
1537 and you may not live long enough for the sort to complete on arrays
1538 of a million elements. So the 5.8 quicksort scrambles large arrays
1539 before sorting them, as a statistical defence against quadratic behaviour.
1540 But that means if you sort the same large array twice, ties may be
1541 broken in different ways.
1543 Because of the unpredictability of tie-breaking order, and the quadratic
1544 worst-case behaviour, quicksort was I<almost> replaced completely with
1545 a stable mergesort. I<Stable> means that ties are broken to preserve
1546 the original order of appearance in the input array. So
1548 sort { ($a % 2) <=> ($b % 2) } (3,1,4,1,5,9);
1550 will yield (4,3,1,1,5,9), guaranteed. The even and odd numbers
1551 appear in the output in the same order they appeared in the input.
1552 Mergesort has worst case O(N log N) behaviour, the best value
1553 attainable. And, ironically, this mergesort does particularly
1554 well where quicksort goes quadratic: mergesort sorts (1..$N, 1..$N)
1555 in O(N) time. But quicksort was rescued at the last moment because
1556 it is faster than mergesort on certain inputs and platforms.
1557 For example, if you really I<don't> care about the order of even
1558 and odd digits, quicksort will run in O(N) time; it's very good
1559 at sorting many repetitions of a small number of distinct elements.
1560 The quicksort divide and conquer strategy works well on platforms
1561 with relatively small, very fast, caches. Eventually, the problem gets
1562 whittled down to one that fits in the cache, from which point it
1563 benefits from the increased memory speed.
1565 Quicksort was rescued by implementing a sort pragma to control aspects
1566 of the sort. The B<stable> subpragma forces stable behaviour,
1567 regardless of algorithm. The B<_quicksort> and B<_mergesort>
1568 subpragmas are heavy-handed ways to select the underlying implementation.
1569 The leading C<_> is a reminder that these subpragmas may not survive
1570 beyond 5.8. More appropriate mechanisms for selecting the implementation
1571 exist, but they wouldn't have arrived in time to save quicksort.
1575 Hashes now use Bob Jenkins "One-at-a-Time" hashing key algorithm
1576 ( http://burtleburtle.net/bob/hash/doobs.html ). This algorithm is
1577 reasonably fast while producing a much better spread of values than
1578 the old hashing algorithm (originally by Chris Torek, later tweaked by
1579 Ilya Zakharevich). Hash values output from the algorithm on a hash of
1580 all 3-char printable ASCII keys comes much closer to passing the
1581 DIEHARD random number generation tests. According to perlbench, this
1582 change has not affected the overall speed of Perl.
1586 unshift() should now be noticeably faster.
1590 =head1 Installation and Configuration Improvements
1592 =head2 Generic Improvements
1598 INSTALL now explains how you can configure Perl to use 64-bit
1599 integers even on non-64-bit platforms.
1603 Policy.sh policy change: if you are reusing a Policy.sh file
1604 (see INSTALL) and you use Configure -Dprefix=/foo/bar and in the old
1605 Policy $prefix eq $siteprefix and $prefix eq $vendorprefix, all of
1606 them will now be changed to the new prefix, /foo/bar. (Previously
1607 only $prefix changed.) If you do not like this new behaviour,
1608 specify prefix, siteprefix, and vendorprefix explicitly.
1612 A new optional location for Perl libraries, otherlibdirs, is available.
1613 It can be used for example for vendor add-ons without disturbing Perl's
1614 own library directories.
1618 In many platforms, the vendor-supplied 'cc' is too stripped-down to
1619 build Perl (basically, 'cc' doesn't do ANSI C). If this seems
1620 to be the case and 'cc' does not seem to be the GNU C compiler
1621 'gcc', an automatic attempt is made to find and use 'gcc' instead.
1625 gcc needs to closely track the operating system release to avoid
1626 build problems. If Configure finds that gcc was built for a different
1627 operating system release than is running, it now gives a clearly visible
1628 warning that there may be trouble ahead.
1632 Since Perl 5.8 is not binary-compatible with previous releases
1633 of Perl, Configure no longer suggests including the 5.005
1638 Configure C<-S> can now run non-interactively.
1642 Configure support for pdp11-style memory models has been removed due
1647 configure.gnu now works with options with whitespace in them.
1651 installperl now outputs everything to STDERR.
1655 Because PerlIO is now the default on most platforms, "-perlio" doesn't
1656 get appended to the $Config{archname} (also known as $^O) anymore.
1657 Instead, if you explicitly choose not to use perlio (Configure command
1658 line option -Uuseperlio), you will get "-stdio" appended.
1662 Another change related to the architecture name is that "-64all"
1663 (-Duse64bitall, or "maximally 64-bit") is appended only if your
1664 pointers are 64 bits wide. (To be exact, the use64bitall is ignored.)
1668 In AFS installations, one can configure the root of the AFS to be
1669 somewhere else than the default F</afs> by using the Configure
1670 parameter C<-Dafsroot=/some/where/else>.
1674 APPLLIB_EXP, a lesser-known configuration-time definition, has been
1675 documented. It can be used to prepend site-specific directories
1676 to Perl's default search path (@INC); see INSTALL for information.
1680 The version of Berkeley DB used when the Perl (and, presumably, the
1681 DB_File extension) was built is now available as
1682 C<@Config{qw(db_version_major db_version_minor db_version_patch)}>
1683 from Perl and as C<DB_VERSION_MAJOR_CFG DB_VERSION_MINOR_CFG
1684 DB_VERSION_PATCH_CFG> from C.
1688 Building Berkeley DB3 for compatibility modes for DB, NDBM, and ODBM
1689 has been documented in INSTALL.
1693 If you have CPAN access (either network or a local copy such as a
1694 CD-ROM) you can during specify extra modules to Configure to build and
1695 install with Perl using the -Dextras=... option. See INSTALL for
1700 In addition to config.over, a new override file, config.arch, is
1701 available. This file is supposed to be used by hints file writers
1702 for architecture-wide changes (as opposed to config.over which is
1703 for site-wide changes).
1707 If your file system supports symbolic links, you can build Perl outside
1708 of the source directory by
1710 mkdir /tmp/perl/build/directory
1711 cd /tmp/perl/build/directory
1712 sh /path/to/perl/source/Configure -Dmksymlinks ...
1714 This will create in /tmp/perl/build/directory a tree of symbolic links
1715 pointing to files in /path/to/perl/source. The original files are left
1716 unaffected. After Configure has finished, you can just say
1720 and Perl will be built and tested, all in /tmp/perl/build/directory.
1724 For Perl developers, several new make targets for profiling
1725 and debugging have been added; see L<perlhack>.
1731 Use of the F<gprof> tool to profile Perl has been documented in
1732 L<perlhack>. There is a make target called "perl.gprof" for
1733 generating a gprofiled Perl executable.
1737 If you have GCC 3, there is a make target called "perl.gcov" for
1738 creating a gcoved Perl executable for coverage analysis. See
1743 If you are on IRIX or Tru64 platforms, new profiling/debugging options
1744 have been added; see L<perlhack> for more information about pixie and
1751 Guidelines of how to construct minimal Perl installations have
1752 been added to INSTALL.
1756 The Thread extension is now not built at all under ithreads
1757 (C<Configure -Duseithreads>) because it wouldn't work anyway (the
1758 Thread extension requires being Configured with C<-Duse5005threads>).
1760 But note that the Thread.pm interface is now shared by both
1765 The Gconvert macro ($Config{d_Gconvert}) used by perl for stringifying
1766 floating-point numbers is now more picky about using sprintf %.*g
1767 rules for the conversion. Some platforms that used to use gcvt may
1768 now resort to the slower sprintf.
1772 The obsolete method of making a special (e.g., debugging) flavor
1775 make LIBPERL=libperld.a
1777 has been removed. Use -DDEBUGGING instead.
1781 =head2 New Or Improved Platforms
1783 For the list of platforms known to support Perl,
1784 see L<perlport/"Supported Platforms">.
1790 AIX dynamic loading should be now better supported.
1794 AIX should now work better with gcc, threads, and 64-bitness. Also the
1795 long doubles support in AIX should be better now. See L<perlaix>.
1799 AtheOS ( http://www.atheos.cx/ ) is a new platform.
1803 BeOS has been reclaimed.
1807 The DG/UX platform now supports 5.005-style threads.
1812 The DYNIX/ptx platform (a.k.a. dynixptx) is supported at or near
1817 EBCDIC platforms (z/OS (also known as OS/390), POSIX-BC, and VM/ESA)
1818 have been regained. Many test suite tests still fail and the
1819 co-existence of Unicode and EBCDIC isn't quite settled, but the
1820 situation is much better than with Perl 5.6. See L<perlos390>,
1821 L<perlbs2000> (for POSIX-BC), and L<perlvmesa> for more information.
1825 Building perl with -Duseithreads or -Duse5005threads now works under
1826 HP-UX 10.20 (previously it only worked under 10.30 or later). You will
1827 need a thread library package installed. See README.hpux.
1831 MacOS Classic (MacPerl has of course been available since
1832 perl 5.004 but now the source code bases of standard Perl
1833 and MacPerl have been synchronised)
1837 MacOS X (or Darwin) should now be able to build Perl even on HFS+
1838 filesystems. (The case-insensitivity used to confuse the Perl build
1843 NCR MP-RAS is now supported.
1847 All the NetBSD specific patches (except for the installation
1848 specific ones) have been merged back to the main distribution.
1852 NetWare from Novell is now supported. See L<perlnetware>.
1856 NonStop-UX is now supported.
1860 NEC SUPER-UX is now supported.
1864 All the OpenBSD specific patches (except for the installation
1865 specific ones) have been merged back to the main distribution.
1869 Perl has been tested with the GNU pth userlevel thread package
1870 ( http://www.gnu.org/software/pth/pth.html ) . All but one thread
1871 test worked, and that one failure was because of test results arriving
1872 in unexpected order.
1876 Stratus VOS is now supported using Perl's native build method
1877 (Configure). This is the recommended method to build Perl on
1878 VOS. The older methods, which build miniperl, are still
1879 available. See L<perlvos>.
1883 The Amdahl UTS UNIX mainframe platform is now supported.
1887 WinCE is now supported. See L<perlce>.
1891 z/OS (formerly known as OS/390, formerly known as MVS OE) now has
1892 support for dynamic loading. This is not selected by default,
1893 however, you must specify -Dusedl in the arguments of Configure.
1897 =head1 Selected Bug Fixes
1899 Numerous memory leaks and uninitialized memory accesses have been
1900 hunted down. Most importantly, anonymous subs used to leak quite
1907 The autouse pragma didn't work for Multi::Part::Function::Names.
1911 caller() could cause core dumps in certain situations. Carp was sometimes
1912 affected by this problem. In particular, caller() now returns a
1913 subroutine name of C<(unknown)> for subroutines that have been removed
1914 from the symbol table.
1918 chop(@list) in list context returned the characters chopped in
1919 reverse order. This has been reversed to be in the right order.
1923 Configure no longer includes the DBM libraries (dbm, gdbm, db, ndbm)
1924 when building the Perl binary. The only exception to this is SunOS 4.x,
1929 The behaviour of non-decimal but numeric string constants such as
1930 "0x23" was platform-dependent: in some platforms that was seen as 35,
1931 in some as 0, in some as a floating point number (don't ask). This
1932 was caused by Perl's using the operating system libraries in a situation
1933 where the result of the string to number conversion is undefined: now
1934 Perl consistently handles such strings as zero in numeric contexts.
1938 The order of DESTROYs has been made more predictable.
1942 Several debugger fixes: exit code now reflects the script exit code,
1943 condition C<"0"> now treated correctly, the C<d> command now checks
1944 line number, C<$.> no longer gets corrupted, and all debugger output
1945 now goes correctly to the socket if RemotePort is set.
1949 Perl 5.6.0 could emit spurious warnings about redefinition of dl_error()
1950 when statically building extensions into perl. This has been corrected.
1954 L<dprofpp> -R didn't work.
1958 C<*foo{FORMAT}> now works.
1962 Infinity is now recognized as a number.
1966 UNIVERSAL::isa no longer caches methods incorrectly. (This broke
1967 the Tk extension with 5.6.0.)
1971 Lexicals I: lexicals outside an eval "" weren't resolved
1972 correctly inside a subroutine definition inside the eval "" if they
1973 were not already referenced in the top level of the eval""ed code.
1977 Lexicals II: lexicals leaked at file scope into subroutines that
1978 were declared before the lexicals.
1982 Lexical warnings now propagating correctly between scopes
1983 and into C<eval "...">.
1987 C<use warnings qw(FATAL all)> did not work as intended. This has been
1992 warnings::enabled() now reports the state of $^W correctly if the caller
1993 isn't using lexical warnings.
1997 Line renumbering with eval and C<#line> now works.
2001 Fixed numerous memory leaks, especially in eval "".
2005 Localised tied variables no longer leak memory
2008 tie my %tied_hash => 'Tie::StdHash';
2012 # Used to leak memory every time local() was called;
2013 # in a loop, this added up.
2014 local($tied_hash{Foo}) = 1;
2018 Localised hash elements (and %ENV) are correctly unlocalised to not
2019 exist, if they didn't before they were localised.
2023 tie my %tied_hash => 'Tie::StdHash';
2027 # Nothing has set the FOO element so far
2029 { local $tied_hash{FOO} = 'Bar' }
2031 # This used to print, but not now.
2032 print "exists!\n" if exists $tied_hash{FOO};
2034 As a side effect of this fix, tied hash interfaces B<must> define
2035 the EXISTS and DELETE methods.
2039 mkdir() now ignores trailing slashes in the directory name,
2040 as mandated by POSIX.
2044 Some versions of glibc have a broken modfl(). This affects builds
2045 with C<-Duselongdouble>. This version of Perl detects this brokenness
2046 and has a workaround for it. The glibc release 2.2.2 is known to have
2047 fixed the modfl() bug.
2051 Modulus of unsigned numbers now works (4063328477 % 65535 used to
2052 return 27406, instead of 27047).
2056 Some "not a number" warnings introduced in 5.6.0 eliminated to be
2057 more compatible with 5.005. Infinity is now recognised as a number.
2061 Numeric conversions did not recognize changes in the string value
2062 properly in certain circumstances.
2066 Attributes (such as :shared) didn't work with our().
2070 our() variables will not cause "will not stay shared" warnings.
2074 "our" variables of the same name declared in two sibling blocks
2075 resulted in bogus warnings about "redeclaration" of the variables.
2076 The problem has been corrected.
2080 pack "Z" now correctly terminates the string with "\0".
2084 Fix password routines which in some shadow password platforms
2085 (e.g. HP-UX) caused getpwent() to return every other entry.
2089 The PERL5OPT environment variable (for passing command line arguments
2090 to Perl) didn't work for more than a single group of options.
2094 PERL5OPT with embedded spaces didn't work.
2098 printf() no longer resets the numeric locale to "C".
2102 C<qw(a\\b)> now parses correctly as C<'a\\b'>: that is, as three
2103 characters, not four.
2107 pos() did not return the correct value within s///ge in earlier
2108 versions. This is now handled correctly.
2112 Printing quads (64-bit integers) with printf/sprintf now works
2113 without the q L ll prefixes (assuming you are on a quad-capable platform).
2117 Regular expressions on references and overloaded scalars now work.
2121 Right-hand side magic (GMAGIC) could in many cases such as string
2122 concatenation be invoked too many times.
2126 scalar() now forces scalar context even when used in void context.
2130 SOCKS support is now much more robust.
2134 sort() arguments are now compiled in the right wantarray context
2135 (they were accidentally using the context of the sort() itself).
2136 The comparison block is now run in scalar context, and the arguments
2137 to be sorted are always provided list context.
2141 Changed the POSIX character class C<[[:space:]]> to include the (very
2142 rarely used) vertical tab character. Added a new POSIX-ish character
2143 class C<[[:blank:]]> which stands for horizontal whitespace
2144 (currently, the space and the tab).
2148 The tainting behaviour of sprintf() has been rationalized. It does
2149 not taint the result of floating point formats anymore, making the
2150 behaviour consistent with that of string interpolation.
2154 Some cases of inconsistent taint propagation (such as within hash
2155 values) have been fixed.
2159 The RE engine found in Perl 5.6.0 accidentally pessimised certain kinds
2160 of simple pattern matches. These are now handled better.
2164 Regular expression debug output (whether through C<use re 'debug'>
2165 or via C<-Dr>) now looks better.
2169 Multi-line matches like C<"a\nxb\n" =~ /(?!\A)x/m> were flawed. The
2174 Use of $& could trigger a core dump under some situations. This
2179 The regular expression captured submatches ($1, $2, ...) are now
2180 more consistently unset if the match fails, instead of leaving false
2181 data lying around in them.
2185 readline() on files opened in "slurp" mode could return an extra
2186 "" (blank line) at the end in certain situations. This has been
2191 Autovivification of symbolic references of special variables described
2192 in L<perlvar> (as in C<${$num}>) was accidentally disabled. This works
2197 Sys::Syslog ignored the C<LOG_AUTH> constant.
2201 All but the first argument of the IO syswrite() method are now optional.
2205 $AUTOLOAD, sort(), lock(), and spawning subprocesses
2206 in multiple threads simultaneously are now thread-safe.
2210 Tie::Array's SPLICE method was broken.
2214 Allow a read-only string on the left-hand side of a non-modifying tr///.
2218 If C<STDERR> is tied, warnings caused by C<warn> and C<die> now
2219 correctly pass to it.
2223 Several Unicode fixes.
2229 BOMs (byte order marks) at the beginning of Perl files
2230 (scripts, modules) should now be transparently skipped.
2231 UTF-16 and UCS-2 encoded Perl files should now be read correctly.
2235 The character tables have been updated to Unicode 3.2.0.
2239 Comparing with utf8 data does not magically upgrade non-utf8 data
2240 into utf8. (This was a problem for example if you were mixing data
2241 from I/O and Unicode data: your output might have got magically encoded
2246 Generating illegal Unicode code points such as U+FFFE, or the UTF-16
2247 surrogates, now also generates an optional warning.
2251 C<IsAlnum>, C<IsAlpha>, and C<IsWord> now match titlecase.
2255 Concatenation with the C<.> operator or via variable interpolation,
2256 C<eq>, C<substr>, C<reverse>, C<quotemeta>, the C<x> operator,
2257 substitution with C<s///>, single-quoted UTF8, should now work.
2261 The C<tr///> operator now works. Note that the C<tr///CU>
2262 functionality has been removed (but see pack('U0', ...)).
2266 C<eval "v200"> now works.
2270 Perl 5.6.0 parsed m/\x{ab}/ incorrectly, leading to spurious warnings.
2271 This has been corrected.
2275 Zero entries were missing from the Unicode classes such as C<IsDigit>.
2281 Large unsigned numbers (those above 2**31) could sometimes lose their
2282 unsignedness, causing bogus results in arithmetic operations.
2286 The Perl parser has been stress tested using both random input and
2287 Markov chain input and the few found crashes and lockups have been
2292 =head2 Platform Specific Changes and Fixes
2300 Perl now works on post-4.0 BSD/OSes.
2306 Setting C<$0> now works (as much as possible; see L<perlvar> for details).
2312 Numerous updates; currently synchronised with Cygwin 1.3.10.
2316 Previously DYNIX/ptx had problems in its Configure probe for non-blocking I/O.
2322 EPOC update after Perl 5.6.0. See README.epoc.
2328 Perl now works on post-3.0 FreeBSDs.
2334 README.hpux updated; C<Configure -Duse64bitall> now works;
2335 now uses HP-UX malloc instead of Perl malloc.
2341 Numerous compilation flag and hint enhancements; accidental mixing
2342 of 32-bit and 64-bit libraries (a doomed attempt) made much harder.
2352 Long doubles should now work (see INSTALL).
2356 Linux previously had problems related to sockaddrlen when using
2357 accept(), recvfrom() (in Perl: recv()), getpeername(), and
2366 Compilation of the standard Perl distribution in MacOS Classic should
2367 now work if you have the Metrowerks development environment and
2368 the missing Mac-specific toolkit bits. Contact the macperl mailing
2375 MPE/iX update after Perl 5.6.0. See README.mpeix.
2379 NetBSD/threads: try installing the GNU pth (should be in the
2380 packages collection, or http://www.gnu.org/software/pth/),
2381 and Configure with -Duseithreads.
2387 Perl now works on NetBSD/sparc.
2393 Now works with usethreads (see INSTALL).
2399 64-bitness using the Sun Workshop compiler now works.
2405 The native build method requires at least VOS Release 14.5.0
2406 and GNU C++/GNU Tools 2.0.1 or later. The Perl pack function
2407 now maps overflowed values to +infinity and underflowed values
2412 Tru64 (aka Digital UNIX, aka DEC OSF/1)
2414 The operating system version letter now recorded in $Config{osvers}.
2415 Allow compiling with gcc (previously explicitly forbidden). Compiling
2416 with gcc still not recommended because buggy code results, even with
2423 Fixed various alignment problems that lead into core dumps either
2424 during build or later; no longer dies on math errors at runtime;
2425 now using full quad integers (64 bits), previously was using
2426 only 46 bit integers for speed.
2432 chdir() now works better despite a CRT bug; now works with MULTIPLICITY
2433 (see INSTALL); now works with Perl's malloc.
2435 The tainting of C<%ENV> elements via C<keys> or C<values> was previously
2436 unimplemented. It now works as documented.
2438 The C<waitpid> emulation has been improved. The worst bug (now fixed)
2439 was that a pid of -1 would cause a wildcard search of all processes on
2442 POSIX-style signals are now emulated much better on VMS versions prior
2445 The C<system> function and backticks operator have improved
2446 functionality and better error handling.
2448 File access tests now use current process privileges rather than the
2449 user's default privileges, which could sometimes result in a mismatch
2450 between reported access and actual access.
2452 There is a new C<kill> implementation based on C<sys$sigprc> that allows
2453 older VMS systems (pre-7.0) to use C<kill> to send signals rather than
2454 simply force exit. This implementation also allows later systems to
2455 call C<kill> from within a signal handler.
2457 Iterative logical name translations are now limited to 10 iterations in
2458 imitation of SHOW LOGICAL and other OpenVMS facilities.
2468 accept() no longer leaks memory.
2472 Borland C++ v5.5 is now a supported compiler that can build Perl.
2473 However, the generated binaries continue to be incompatible with those
2474 generated by the other supported compilers (GCC and Visual C++).
2478 Better chdir() return value for a non-existent directory.
2482 Duping socket handles with open(F, ">&MYSOCK") now works under Windows 9x.
2486 New %ENV entries now propagate to subprocesses.
2490 Current directory entries in %ENV are now correctly propagated to child
2495 $ENV{LIB} now used to search for libs under Visual C.
2499 fork() emulation has been improved in various ways, but still continues
2500 to be experimental. See L<perlfork> for known bugs and caveats.
2504 A failed (pseudo)fork now returns undef and sets errno to EAGAIN.
2508 Win32::GetCwd() correctly returns C:\ instead of C: when at the drive root.
2509 Other bugs in chdir() and Cwd::cwd() have also been fixed.
2513 HTML files will be installed in c:\perl\html instead of c:\perl\lib\pod\html
2517 The makefiles now provide a single switch to bulk-enable all the features
2518 enabled in ActiveState ActivePerl (a popular Win32 binary distribution).
2522 Allow REG_EXPAND_SZ keys in the registry.
2526 Can now send() from all threads, not just the first one.
2530 Fake signal handling reenabled, bugs and all.
2534 %SIG has been enabled under USE_ITHREADS, but its use is completely
2535 unsupported under all configurations.
2539 Less stack reserved per thread so that more threads can run
2540 concurrently. (Still 16M per thread.)
2544 C<< File::Spec->tmpdir() >> now prefers C:/temp over /tmp
2545 (works better when perl is running as service).
2549 Better UNC path handling under ithreads.
2553 wait(), waitpid(), and backticks now return the correct exit status
2558 Win64 compilation is now supported.
2562 winsock handle leak fixed.
2568 =head1 New or Changed Diagnostics
2574 The lexical warnings category "deprecated" is no longer a sub-category
2575 of the "syntax" category. It is now a top-level category in its own
2580 All regular expression compilation error messages are now hopefully
2581 easier to understand both because the error message now comes before
2582 the failed regex and because the point of failure is now clearly
2583 marked by a C<E<lt>-- HERE> marker.
2587 The various "opened only for", "on closed", "never opened" warnings
2588 drop the C<main::> prefix for filehandles in the C<main> package,
2589 for example C<STDIN> instead of C<main::STDIN>.
2593 The "Unrecognized escape" warning has been extended to include C<\8>,
2594 C<\9>, and C<\_>. There is no need to escape any of the C<\w> characters.
2598 Two new debugging options have been added: if you have compiled your
2599 Perl with debugging, you can use the -DT and -DR options to trace
2600 tokenising and to add reference counts to displaying variables,
2605 The debugger (perl5db.pl) has been modified to present a more
2606 consistent commands interface, via (CommandSet=580). perl5db.t was
2607 also added to test the changes, and as a placeholder for further tests.
2613 The debugger has a new C<dumpDepth> option to control the maximum
2614 depth to which nested structures are dumped. The C<x> command has
2615 been extended so that C<x N EXPR> dumps out the value of I<EXPR> to a
2616 depth of at most I<N> levels.
2620 The debugger can now show lexical variables if you have the CPAN
2621 module PadWalker installed.
2625 If an attempt to use a (non-blessed) reference as an array index
2626 is made, a warning is given.
2630 C<push @a;> and C<unshift @a;> (with no values to push or unshift)
2631 now give a warning. This may be a problem for generated and evaled
2636 If you try to L<perlfunc/pack> a number less than 0 or larger than 255
2637 using the C<"C"> format you will get an optional warning. Similarly
2638 for the C<"c"> format and a number less than -128 or more than 127.
2642 Certain regex modifiers such as C<(?o)> make sense only if applied to
2643 the entire regex. You will get an optional warning if you try to do
2648 Using arrays or hashes as references (e.g. C<< %foo->{bar} >>
2649 has been deprecated for a while. Now you will get an optional warning.
2653 Using C<sort> in scalar context now issues an optional warning.
2654 This didn't do anything useful, as the sort was not performed.
2658 =head1 Changed Internals
2664 perlapi.pod (a companion to perlguts) now attempts to document the
2669 You can now build a really minimal perl called microperl.
2670 Building microperl does not require even running Configure;
2671 C<make -f Makefile.micro> should be enough. Beware: microperl makes
2672 many assumptions, some of which may be too bold; the resulting
2673 executable may crash or otherwise misbehave in wondrous ways.
2674 For careful hackers only.
2678 Added rsignal(), whichsig(), do_join(), op_clear, op_null,
2679 ptr_table_clear(), ptr_table_free(), sv_setref_uv(), and several UTF-8
2680 interfaces to the publicised API. For the full list of the available
2681 APIs see L<perlapi>.
2685 Made possible to propagate customised exceptions via croak()ing.
2689 Now xsubs can have attributes just like subs. (Well, at least the
2690 built-in attributes.)
2694 dTHR and djSP have been obsoleted; the former removed (because it's
2695 a no-op) and the latter replaced with dSP.
2699 PERL_OBJECT has been completely removed.
2703 The MAGIC constants (e.g. C<'P'>) have been macrofied
2704 (e.g. C<PERL_MAGIC_TIED>) for better source code readability
2705 and maintainability.
2709 The regex compiler now maintains a structure that identifies nodes in
2710 the compiled bytecode with the corresponding syntactic features of the
2711 original regex expression. The information is attached to the new
2712 C<offsets> member of the C<struct regexp>. See L<perldebguts> for more
2713 complete information.
2717 The C code has been made much more C<gcc -Wall> clean. Some warning
2718 messages still remain in some platforms, so if you are compiling with
2719 gcc you may see some warnings about dubious practices. The warnings
2720 are being worked on.
2724 F<perly.c>, F<sv.c>, and F<sv.h> have now been extensively commented.
2728 Documentation on how to use the Perl source repository has been added
2729 to F<Porting/repository.pod>.
2733 There are now several profiling make targets.
2737 =head1 Security Vulnerability Closed
2739 (This change was already made in 5.7.0 but bears repeating here.)
2741 A potential security vulnerability in the optional suidperl component
2742 of Perl was identified in August 2000. suidperl is neither built nor
2743 installed by default. As of November 2001 the only known vulnerable
2744 platform is Linux, most likely all Linux distributions. CERT and
2745 various vendors and distributors have been alerted about the vulnerability.
2746 See http://www.cpan.org/src/5.0/sperl-2000-08-05/sperl-2000-08-05.txt
2747 for more information.
2749 The problem was caused by Perl trying to report a suspected security
2750 exploit attempt using an external program, /bin/mail. On Linux
2751 platforms the /bin/mail program had an undocumented feature which
2752 when combined with suidperl gave access to a root shell, resulting in
2753 a serious compromise instead of reporting the exploit attempt. If you
2754 don't have /bin/mail, or if you have 'safe setuid scripts', or if
2755 suidperl is not installed, you are safe.
2757 The exploit attempt reporting feature has been completely removed from
2758 Perl 5.8.0 (and the maintenance release 5.6.1, and it was removed also
2759 from all the Perl 5.7 releases), so that particular vulnerability
2760 isn't there anymore. However, further security vulnerabilities are,
2761 unfortunately, always possible. The suidperl functionality is most
2762 probably going to be removed in Perl 5.10. In any case, suidperl
2763 should only be used by security experts who know exactly what they are
2764 doing and why they are using suidperl instead of some other solution
2765 such as sudo ( see http://www.courtesan.com/sudo/ ).
2769 Several new tests have been added, especially for the F<lib> and F<ext>
2770 subsections. There are now about 65 000 individual tests (spread over
2771 about 700 test scripts), in the regression suite (5.6.1 has about
2772 11700 tests, in 258 test scripts) Many of the new tests are of course
2773 introduced by the new modules, but still in general Perl is now more
2776 Because of the large number of tests, running the regression suite
2777 will take considerably longer time than it used to: expect the suite
2778 to take up to 4-5 times longer to run than in perl 5.6. On a really
2779 fast machine you can hope to finish the suite in about 6-8 minutes
2782 The tests are now reported in a different order than in earlier Perls.
2783 (This happens because the test scripts from under t/lib have been moved
2784 to be closer to the library/extension they are testing.)
2786 =head1 Known Problems
2794 In AIX 4.2, Perl extensions that use C++ functions that use statics
2795 may have problems in that the statics are not getting initialized.
2796 In newer AIX releases, this has been solved by linking Perl with
2797 the libC_r library, but unfortunately in AIX 4.2 the said library
2798 has an obscure bug where the various functions related to time
2799 (such as time() and gettimeofday()) return broken values, and
2800 therefore in AIX 4.2 Perl is not linked against libC_r.
2804 vac 5.0.0.0 May Produce Buggy Code For Perl
2806 The AIX C compiler vac version 5.0.0.0 may produce buggy code,
2807 resulting in a few random tests failing when run as part of "make
2808 test", but when the failing tests are run by hand, they succeed.
2809 We suggest upgrading to at least vac version 5.0.1.0, that has been
2810 known to compile Perl correctly. "lslpp -L|grep vac.C" will tell
2811 you the vac version. See README.aix.
2815 If building threaded Perl, you may get compilation warning from pp_sys.c:
2817 "pp_sys.c", line 4651.39: 1506-280 (W) Function argument assignment between types "unsigned char*" and "const void*" is not allowed.
2819 This is harmless; it is caused by the getnetbyaddr() and getnetbyaddr_r()
2820 having slightly different types for their first argument.
2824 =head2 Alpha systems with old gccs fail several tests
2826 If you see op/pack, op/pat, op/regexp, or ext/Storable tests failing
2827 in a Linux/alpha or *BSD/Alpha, it's probably time to upgrade your gcc.
2828 gccs prior to 2.95.3 are definitely not good enough, and gcc 3.1 may
2829 be even better. (RedHat Linux/alpha with gcc 3.1 reported no problems,
2830 as did Linux 2.4.18 with gcc 2.95.4.) (In Tru64, it is preferable to
2831 use the bundled C compiler.)
2835 Perl 5.8.0 doesn't build in AmigaOS. It broke at some point
2836 during the ithreads work and we could not find Amiga experts
2837 to unbreak the problems.
2841 The following tests fail on 5.8.0 Perl in BeOS Personal 5.03:
2843 t/op/lfs............................FAILED at test 17
2844 t/op/magic..........................FAILED at test 24
2845 ext/POSIX/t/sigaction...............FAILED at test 13
2846 ext/POSIX/t/waitpid.................FAILED at test 1
2848 See L<perlbeos> (README.beos) for more details.
2850 =head2 Cygwin "unable to remap"
2852 For example when building the Tk extension for Cygwin,
2853 you may get an error message saying "unable to remap".
2854 This is known problem with Cygwin, and a workaround is
2855 detailed in here: http://sources.redhat.com/ml/cygwin/2001-12/msg00894.html
2857 =head2 ext/threads/t/libc
2859 If this test fails, it indicates that your libc (C library) is not
2860 threadsafe. This particular test stress tests the localtime() call to
2861 find out whether it is threadsafe. See L<perlthrtut> for more information.
2863 =head2 FreeBSD Failing locale Test 117 For ISO8859-15 Locales
2865 The ISO8859-15 locales may fail the locale test 117 in FreeBSD.
2866 This is caused by the characters \xFF (y with diaeresis) and \xBE
2867 (Y with diaeresis) not behaving correctly when being matched
2870 =head2 IRIX fails ext/List/Util/t/shuffle.t
2872 IRIX with MIPSpro 7.3.1.3m compiler may fail the said List::Util test
2873 by dumping core. This seems to be a compiler error since if compiled
2874 with gcc no core dump ensues, and no failures on the said test on any
2877 =head2 Modifying $_ Inside for(..)
2881 works without complaint. It shouldn't. (You should be able to
2882 modify only lvalue elements inside the loops.) You can see the
2883 correct behaviour by replacing the 1..5 with 1, 2, 3, 4, 5.
2885 =head2 mod_perl 1.26 Doesn't Build With Threaded Perl
2887 Use mod_perl 1.27 or higher.
2889 =head2 lib/ftmp-security tests warn 'system possibly insecure'
2891 Don't panic. Read the 'make test' section of INSTALL instead.
2893 =head2 HP-UX lib/posix Subtest 9 Fails When LP64-Configured
2895 If perl is configured with -Duse64bitall, the successful result of the
2896 subtest 10 of lib/posix may arrive before the successful result of the
2897 subtest 9, which confuses the test harness so much that it thinks the
2900 =head2 Linux with glibc 2.2.5 fails t/op/int subtest #6 with -Duse64bitint
2902 This is a known bug in the glibc 2.2.5 with long long integers.
2903 ( http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=65612 )
2905 =head2 Linux With Sfio Fails op/misc Test 48
2909 =head2 libwww-perl (LWP) fails base/date #51
2911 Use libwww-perl 5.65 or later.
2915 Please remember to set your environment variable LC_ALL to "C"
2916 (setenv LC_ALL C) before running "make test" to avoid a lot of
2917 warnings about the broken locales of Mac OS X.
2919 The following tests are known to fail in Mac OS X 10.1.4 because of
2920 buggy (old) implementations of Berkeley DB included in Mac OS X:
2922 Failed Test Stat Wstat Total Fail Failed List of Failed
2923 -------------------------------------------------------------------------
2924 ../ext/DB_File/t/db-btree.t 0 11 ?? ?? % ??
2925 ../ext/DB_File/t/db-recno.t 149 3 2.01% 61 63 65
2927 If you are building on a UFS partition, you will also probably see
2928 t/op/stat.t subtest #9 fail. This is caused by Darwin's UFS not
2929 supporting inode change time.
2931 Also the ext/POSIX/t/posix.t subtest #10 fails but it is skipped for
2932 now because the failure is Apple's fault, not Perl's (blocked signals
2935 If you Configure with ithreads, ext/threads/t/libc.t will fail. Again,
2936 this is not Perl's fault-- the libc of Mac OS X is not threadsafe
2937 (in this particular test, the localtime() call is found to be
2940 =head2 op/sprintf tests 91, 129, and 130
2942 The op/sprintf tests 91, 129, and 130 are known to fail on some platforms.
2943 Examples include any platform using sfio, and Compaq/Tandem's NonStop-UX.
2945 Test 91 is known to fail on QNX6 (nto), because C<sprintf '%e',0>
2946 incorrectly produces C<0.000000e+0> instead of C<0.000000e+00>.
2948 For tests 129 and 130, the failing platforms do not comply with
2949 the ANSI C Standard: lines 19ff on page 134 of ANSI X3.159 1989, to
2950 be exact. (They produce something other than "1" and "-1" when
2951 formatting 0.6 and -0.6 using the printf format "%.0f"; most often,
2952 they produce "0" and "-0".)
2956 In case you are still using Solaris 2.5 (aka SunOS 5.5), you may
2957 experience failures (the test core dumping) in lib/locale.t.
2958 The suggested cure is to upgrade your Solaris.
2962 When Perl is built using the native build process on VOS Release
2963 14.5.0 and GNU C++/GNU Tools 2.0.1, all attempted tests either
2964 pass or result in TODO (ignored) failures.
2966 =head2 Term::ReadKey not working on Win32
2968 Use Term::ReadKey 2.20 or later.
2970 =head2 Failure of Thread (5.005-style) tests
2972 B<Note that support for 5.005-style threading is deprecated,
2973 experimental and practically unsupported. In 5.10, it is expected
2976 The following tests are known to fail due to fundamental problems in
2977 the 5.005 threading implementation. These are not new failures--Perl
2978 5.005_0x has the same bugs, but didn't have these tests.
2980 ../ext/List/Util/t/first.t 255 65280 7 4 57.14% 2 5-7
2981 ../lib/English.t 2 512 54 2 3.70% 2-3
2982 ../lib/Filter/Simple/t/data.t 6 3 50.00% 1-3
2983 ../lib/Filter/Simple/t/filter_only 9 3 33.33% 1-2 5
2984 ../lib/autouse.t 10 1 10.00% 4
2985 op/flip.t 15 1 6.67% 15
2987 These failures are unlikely to get fixed as 5.005-style threads
2988 are considered fundamentally broken. (Basically what happens is that
2989 competing threads can corrupt shared global state.)
2991 =head2 Timing problems
2993 The following tests may fail intermittently because of timing
2994 problems, for example if the system is heavily loaded.
2997 ext/Time/HiRes/HiRes.t
2999 lib/Memoize/t/expmod_t.t
3000 lib/Memoize/t/speed.t
3002 In case of failure please try running them manually, for example
3004 ./perl -Ilib ext/Time/HiRes/HiRes.t
3008 ../lib/Math/Trig.t 26 1 3.85% 25
3009 ../lib/warnings.t 470 1 0.21% 429
3011 The Trig.t failure is caused by the slighly differing (from IEEE)
3012 floating point implementation of UNICOS. The warnings.t failure is
3013 also related: the test assumes a certain floating point output format;
3014 this assumption fails in UNICOS.
3022 During Configure, the test
3024 Guessing which symbols your C compiler and preprocessor define...
3026 will probably fail with error messages like
3028 CC-20 cc: ERROR File = try.c, Line = 3
3029 The identifier "bad" is undefined.
3031 bad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K
3034 CC-65 cc: ERROR File = try.c, Line = 3
3035 A semicolon is expected at this point.
3037 This is caused by a bug in the awk utility of UNICOS/mk. You can ignore
3038 the error, but it does cause a slight problem: you cannot fully
3039 benefit from the h2ph utility (see L<h2ph>) that can be used to
3040 convert C headers to Perl libraries, mainly used to be able to access
3041 from Perl the constants defined using C preprocessor, cpp. Because of
3042 the above error, parts of the converted headers will be invisible.
3043 Luckily, these days the need for h2ph is rare.
3047 If building Perl with interpreter threads (ithreads), the
3048 getgrent(), getgrnam(), and getgrgid() functions cannot return the
3049 list of the group members due to a bug in the multithreaded support of
3050 UNICOS/mk. What this means is that in list context the functions will
3051 return only three values, not four.
3057 There are a few known test failures, see L<perluts> (README.uts).
3061 There should be no reported test failures with a default configuration,
3062 though there are a number of tests marked TODO that point to areas
3063 needing further debugging and/or porting work.
3067 In multi-CPU boxes, there are some problems with the I/O buffering:
3068 some output may appear twice.
3070 =head2 XML::Parser not working
3072 Use XML::Parser 2.31 or later.
3074 =head2 z/OS (OS/390)
3076 z/OS has rather many test failures but the situation is actually
3077 better than it was in 5.6.0; it's just that so many new modules and
3078 tests have been added.
3080 Failed Test Stat Wstat Total Fail Failed List of Failed
3081 ---------------------------------------------------------------------------
3082 ../ext/Data/Dumper/t/dumper.t 357 8 2.24% 311 314 325 327
3084 ../ext/IO/lib/IO/t/io_unix.t 5 4 80.00% 2-5
3085 ../ext/Storable/t/downgrade.t 12 3072 169 12 7.10% 14-15 46-47 78-79
3087 ../lib/ExtUtils/t/Constant.t 121 30976 48 48 100.00% 1-48
3088 ../lib/ExtUtils/t/Embed.t 9 9 100.00% 1-9
3089 op/pat.t 910 7 0.77% 665 776 785 832-
3091 op/sprintf.t 224 3 1.34% 98 100 136
3092 op/tr.t 97 5 5.15% 63 71-74
3093 uni/fold.t 780 6 0.77% 61 169 196 661
3096 The failures in dumper.t and downgrade.t are problems in the tests,
3097 those in io_unix and sprintf are problems in the USS (UDP sockets
3098 and printf formats). The pat, tr, and fold failures are genuine Perl
3099 problems caused by EBCDIC (and in the pat and fold cases, combining
3100 that with Unicode). The Constant and Embed are probably problems
3101 in the tests (since they test Perl's ability to build extensions,
3102 and that seems to be working reasonably well.)
3104 =head2 Localising Tied Arrays and Hashes Is Broken
3108 doesn't work as one would expect: the old value is restored
3109 incorrectly. This will be changed in a future release, but we don't
3110 know yet what the new semantics will exactly be. In any case, the
3111 change will break existing code that relies on the current
3112 (ill-defined) semantics, so just avoid doing this in general.
3114 =head2 Self-tying Problems
3116 Self-tying of arrays and hashes is broken in rather deep and
3117 hard-to-fix ways. As a stop-gap measure to avoid people from getting
3118 frustrated at the mysterious results (core dumps, most often), it is
3119 forbidden for now (you will get a fatal error even from an attempt).
3121 A change to self-tying of globs has caused them to be recursively
3122 referenced (see: L<perlobj/"Two-Phased Garbage Collection">). You
3123 will now need an explicit untie to destroy a self-tied glob. This
3124 behaviour may be fixed at a later date.
3126 Self-tying of scalars and IO thingies works.
3128 =head2 Building Extensions Can Fail Because Of Largefiles
3130 Some extensions like mod_perl are known to have issues with
3131 `largefiles', a change brought by Perl 5.6.0 in which file offsets
3132 default to 64 bits wide, where supported. Modules may fail to compile
3133 at all, or they may compile and work incorrectly. Currently, there
3134 is no good solution for the problem, but Configure now provides
3135 appropriate non-largefile ccflags, ldflags, libswanted, and libs
3136 in the %Config hash (e.g., $Config{ccflags_nolargefiles}) so the
3137 extensions that are having problems can try configuring themselves
3138 without the largefileness. This is admittedly not a clean solution,
3139 and the solution may not even work at all. One potential failure is
3140 whether one can (or, if one can, whether it's a good idea to) link
3141 together at all binaries with different ideas about file offsets;
3142 all this is platform-dependent.
3144 =head2 Unicode Support on EBCDIC Still Spotty
3146 Though mostly working, Unicode support still has problem spots on
3147 EBCDIC platforms. One such known spot are the C<\p{}> and C<\P{}>
3148 regular expression constructs for code points less than 256: the
3149 C<pP> are testing for Unicode code points, not knowing about EBCDIC.
3151 =head2 The Compiler Suite Is Still Very Experimental
3153 The compiler suite is slowly getting better but it continues to be
3154 highly experimental. Use in production environments is discouraged.
3156 =head2 The Long Double Support Is Still Experimental
3158 The ability to configure Perl's numbers to use "long doubles",
3159 floating point numbers of hopefully better accuracy, is still
3160 experimental. The implementations of long doubles are not yet
3161 widespread and the existing implementations are not quite mature
3162 or standardised, therefore trying to support them is a rare
3163 and moving target. The gain of more precision may also be offset
3164 by slowdown in computations (more bits to move around, and the
3165 operations are more likely to be executed by less optimised
3168 =head2 Seen In Perl 5.7 But Gone Now
3170 C<Time::Piece> (previously known as C<Time::Object>) was removed
3171 because it was felt that it didn't have enough value in it to be a
3172 core module. It is still a useful module, though, and is available
3175 Perl 5.8 unfortunately does not build anymore on AmigaOS;
3176 this broke accidentally at some point. Since there are not that many
3177 Amiga developers available, we could not get this fixed and tested in
3180 =head1 Reporting Bugs
3182 If you find what you think is a bug, you might check the articles
3183 recently posted to the comp.lang.perl.misc newsgroup and the perl
3184 bug database at http://bugs.perl.org/ . There may also be
3185 information at http://www.perl.com/ , the Perl Home Page.
3187 If you believe you have an unreported bug, please run the B<perlbug>
3188 program included with your release. Be sure to trim your bug down
3189 to a tiny but sufficient test case. Your bug report, along with the
3190 output of C<perl -V>, will be sent off to perlbug@perl.org to be
3191 analysed by the Perl porting team.
3195 The F<Changes> file for exhaustive details on what changed.
3197 The F<INSTALL> file for how to build Perl.
3199 The F<README> file for general stuff.
3201 The F<Artistic> and F<Copying> files for copyright information.
3205 Written by Jarkko Hietaniemi <F<jhi@iki.fi>>.