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 64-bit platforms and malloc
51 If your pointers are 64 bits wide, the Perl malloc is no longer being
52 used because it does not work well with 8-byte pointers. Also,
53 usually the system mallocs on such platforms are much better optimized
54 for such large memory models than the Perl malloc. Some memory-hungry
55 Perl applications like the PDL don't work well with Perl's malloc.
56 Finally, other applications than Perl (like modperl) tend to prefer
57 the system malloc. Such platforms include Alpha and 64-bit HPPA,
60 =head2 AIX Dynaloading
62 The AIX dynaloading now uses in AIX releases 4.3 and newer the native
63 dlopen interface of AIX instead of the old emulated interface. This
64 change will probably break backward compatibility with compiled
65 modules. The change was made to make Perl more compliant with other
66 applications like modperl which are using the AIX native interface.
68 =head2 Attributes for C<my> variables now handled at run-time.
70 The C<my EXPR : ATTRS> syntax now applies variable attributes at
71 run-time. (Subroutine and C<our> variables still get attributes applied
72 at compile-time.) See L<attributes> for additional details. In particular,
73 however, this allows variable attributes to be useful for C<tie> interfaces,
74 which was a deficiency of earlier releases. Note that the new semantics
75 doesn't work with the Attribute::Handlers module (as of version 0.76).
77 =head2 Socket Extension Dynamic in VMS
79 The Socket extension is now dynamically loaded instead of being
80 statically built in. This may or may not be a problem with ancient
81 TCP/IP stacks of VMS: we do not know since we weren't able to test
82 Perl in such configurations.
84 =head2 IEEE-format Floating Point Default on OpenVMS Alpha
86 Perl now uses IEEE format (T_FLOAT) as the default internal floating
87 point format on OpenVMS Alpha, potentially breaking binary compatibility
88 with external libraries or existing data. G_FLOAT is still available as
89 a configuration option. The default on VAX (D_FLOAT) has not changed.
91 =head2 New Unicode Properties
93 Unicode I<scripts> are now supported. Scripts are similar to (and superior
94 to) Unicode I<blocks>. The difference between scripts and blocks is that
95 scripts are the glyphs used by a language or a group of languages, while
96 the blocks are more artificial groupings of (mostly) 256 characters based
97 on the Unicode numbering.
99 In general, scripts are more inclusive, but not universally so. For
100 example, while the script C<Latin> includes all the Latin characters and
101 their various diacritic-adorned versions, it does not include the various
102 punctuation or digits (since they are not solely C<Latin>).
104 A number of other properties are now supported, including C<\p{L&}>,
105 C<\p{Any}> C<\p{Assigned}>, C<\p{Unassigned}>, C<\p{Blank}> and
106 C<\p{SpacePerl}> (along with their C<\P{...}> versions, of course).
107 See L<perlunicode> for details, and more additions.
109 The C<In> or C<Is> prefix to names used with the C<\p{...}> and C<\P{...}>
110 are now almost always optional. The only exception is that a C<In> prefix
111 is required to signify a Unicode block when a block name conflicts with a
112 script name. For example, C<\p{Tibetan}> refers to the script, while
113 C<\p{InTibetan}> refers to the block. When there is no name conflict, you
114 can omit the C<In> from the block name (e.g. C<\p{BraillePatterns}>), but
115 to be safe, it's probably best to always use the C<In>).
117 =head2 REF(...) Instead Of SCALAR(...)
119 A reference to a reference now stringifies as "REF(0x81485ec)" instead
120 of "SCALAR(0x81485ec)" in order to be more consistent with the return
123 =head2 pack/unpack D/F recycled
125 The undocumented pack/unpack template letters D/F have been recycled
126 for better use: now they stand for long double (if supported by the
127 platform) and NV (Perl internal floating point type). (They used
128 to be aliases for d/f, but you never knew that.)
136 The semantics of bless(REF, REF) were unclear and until someone proves
137 it to make some sense, it is forbidden.
141 The obsolete chat2 library that should never have been allowed
142 to escape the laboratory has been decommissioned.
146 The builtin dump() function has probably outlived most of its
147 usefulness. The core-dumping functionality will remain in future
148 available as an explicit call to C<CORE::dump()>, but in future
149 releases the behaviour of an unqualified C<dump()> call may change.
153 The very dusty examples in the eg/ directory have been removed.
154 Suggestions for new shiny examples welcome but the main issue is that
155 the examples need to be documented, tested and (most importantly)
160 The (bogus) escape sequences \8 and \9 now give an optional warning
161 ("Unrecognized escape passed through"). There is no need to \-escape
166 The list of filenames from glob() (or <...>) is now by default sorted
167 alphabetically to be csh-compliant (which is what happened before
168 in most UNIX platforms). (bsd_glob() does still sort platform
169 natively, ASCII or EBCDIC, unless GLOB_ALPHASORT is specified.)
173 Spurious syntax errors generated in certain situations, when glob()
174 caused File::Glob to be loaded for the first time, have been fixed.
178 Although "you shouldn't do that", it was possible to write code that
179 depends on Perl's hashed key order (Data::Dumper does this). The new
180 algorithm "One-at-a-Time" produces a different hashed key order.
181 More details are in L</"Performance Enhancements">.
185 lstat(FILEHANDLE) now gives a warning because the operation makes no sense.
186 In future releases this may become a fatal error.
190 The C<package;> syntax (C<package> without an argument) has been
191 deprecated. Its semantics were never that clear and its
192 implementation even less so. If you have used that feature to
193 disallow all but fully qualified variables, C<use strict;> instead.
197 The unimplemented POSIX regex features [[.cc.]] and [[=c=]] are still
198 recognised but now cause fatal errors. The previous behaviour of
199 ignoring them by default and warning if requested was unacceptable
200 since it, in a way, falsely promised that the features could be used.
204 The current user-visible implementation of pseudo-hashes (the weird
205 use of the first array element) is deprecated starting from Perl 5.8.0
206 and will be removed in Perl 5.10.0, and the feature will be
207 implemented differently. Not only is the current interface rather
208 ugly, but the current implementation slows down normal array and hash
209 use quite noticeably. The C<fields> pragma interface will remain
210 available. The I<restricted hashes> interface is expected to
211 be the replacement interface (see L<Hash::Util>).
215 The syntaxes C<< @a->[...] >> and C<< %h->{...} >> have now been deprecated.
219 After years of trying the suidperl is considered to be too complex to
220 ever be considered truly secure. The suidperl functionality is likely
221 to be removed in a future release.
225 The long deprecated uppercase aliases for the string comparison
226 operators (EQ, NE, LT, LE, GE, GT) have now been removed.
230 The tr///C and tr///U features have been removed and will not return;
231 the interface was a mistake. Sorry about that. For similar
232 functionality, see pack('U0', ...) and pack('C0', ...).
236 Earlier Perls treated "sub foo (@bar)" as equivalent to "sub foo (@)".
237 The prototypes are now checked at compile-time for invalid characters.
238 An optional warning is generated ("Illegal character in prototype...")
239 but this may be upgraded to a fatal error in a future release.
243 =head1 Core Enhancements
245 =head2 PerlIO is Now The Default
251 IO is now by default done via PerlIO rather than system's "stdio".
252 PerlIO allows "layers" to be "pushed" onto a file handle to alter the
253 handle's behaviour. Layers can be specified at open time via 3-arg
256 open($fh,'>:crlf :utf8', $path) || ...
258 or on already opened handles via extended C<binmode>:
260 binmode($fh,':encoding(iso-8859-7)');
262 The built-in layers are: unix (low level read/write), stdio (as in
263 previous Perls), perlio (re-implementation of stdio buffering in a
264 portable manner), crlf (does CRLF <=> "\n" translation as on Win32,
265 but available on any platform). A mmap layer may be available if
266 platform supports it (mostly UNIXes).
268 Layers to be applied by default may be specified via the 'open' pragma.
270 See L</"Installation and Configuration Improvements"> for the effects
271 of PerlIO on your architecture name.
275 File handles can be marked as accepting Perl's internal encoding of Unicode
276 (UTF-8 or UTF-EBCDIC depending on platform) by a pseudo layer ":utf8" :
278 open($fh,">:utf8","Uni.txt");
280 Note for EBCDIC users: the pseudo layer ":utf8" is erroneously named
281 for you since it's not UTF-8 what you will be getting but instead
282 UTF-EBCDIC. See L<perlunicode>, L<utf8>, and
283 http://www.unicode.org/unicode/reports/tr16/ for more information.
284 In future releases this naming may change.
288 File handles can translate character encodings from/to Perl's internal
289 Unicode form on read/write via the ":encoding()" layer.
293 File handles can be opened to "in memory" files held in Perl scalars via:
295 open($fh,'>', \$variable) || ...
299 Anonymous temporary files are available without need to
300 'use FileHandle' or other module via
302 open($fh,"+>", undef) || ...
304 That is a literal undef, not an undefined value.
308 The list form of C<open> is now implemented for pipes (at least on UNIX):
310 open($fh,"-|", 'cat', '/etc/motd')
312 creates a pipe, and runs the equivalent of exec('cat', '/etc/motd') in
317 If your locale environment variables (LANGUAGE, LC_ALL, LC_CTYPE, LANG)
318 contain the strings 'UTF-8' or 'UTF8' (case-insensitive matching),
319 the default encoding of your STDIN, STDOUT, and STDERR, and of
320 B<any subsequent file open>, is UTF-8.
326 Perl used to be fragile in that signals arriving at inopportune moments
327 could corrupt Perl's internal state. Now Perl postpones handling of
328 signals until it's safe (between opcodes).
330 This change may have surprising side effects because signals no longer
331 interrupt Perl instantly. Perl will now first finish whatever it was
332 doing, like finishing an internal operation (like sort()) or an
333 external operation (like an I/O operation), and only then look at any
334 arrived signals (and before starting the next operation). No more corrupt
335 internal state since the current operation is always finished first,
336 but the signal may take more time to get heard. Note that breaking
337 out from potentially blocking operations should still work, though.
339 =head2 Unicode Overhaul
341 Unicode in general should be now much more usable than in Perl 5.6.0
342 (or even in 5.6.1). Unicode can be used in hash keys, Unicode in
343 regular expressions should work now, Unicode in tr/// should work now,
344 Unicode in I/O should work now. See L<perluniintro> for introduction
345 and L<perlunicode> for details.
351 The Unicode Character Database coming with Perl has been upgraded
352 to Unicode 3.2.0. For more information, see http://www.unicode.org/ .
356 For developers interested in enhancing Perl's Unicode capabilities:
357 almost all the UCD files are included with the Perl distribution in
358 the F<lib/unicore subdirectory>. The most notable omission, for space
359 considerations, is the Unihan database.
363 The properties \p{Blank} and \p{SpacePerl} have been added. "Blank" is like
364 C isblank(), that is, it contains only "horizontal whitespace" (the space
365 character is, the newline isn't), and the "SpacePerl" is the Unicode
366 equivalent of C<\s> (\p{Space} isn't, since that includes the vertical
367 tabulator character, whereas C<\s> doesn't.)
369 See "New Unicode Properties" earlier in this document for additional
370 information on changes with Unicode properties.
374 =head2 Understanding of Numbers
376 In general a lot of fixing has happened in the area of Perl's
377 understanding of numbers, both integer and floating point. Since in
378 many systems the standard number parsing functions like C<strtoul()>
379 and C<atof()> seem to have bugs, Perl tries to work around their
380 deficiencies. This results hopefully in more accurate numbers.
382 Perl now tries internally to use integer values in numeric conversions
383 and basic arithmetics (+ - * /) if the arguments are integers, and
384 tries also to keep the results stored internally as integers.
385 This change leads to often slightly faster and always less lossy
386 arithmetics. (Previously Perl always preferred floating point numbers
389 =head2 Miscellaneous Changes
395 AUTOLOAD is now lvaluable, meaning that you can add the :lvalue attribute
396 to AUTOLOAD subroutines and you can assign to the AUTOLOAD return value.
400 C<perl -d:Module=arg,arg,arg> now works (previously one couldn't pass
401 in multiple arguments.)
405 The builtin dump() now gives an optional warning
406 C<dump() better written as CORE::dump()>,
407 meaning that by default C<dump(...)> is resolved as the builtin
408 dump() which dumps core and aborts, not as (possibly) user-defined
409 C<sub dump>. To call the latter, qualify the call as C<&dump(...)>.
410 (The whole dump() feature is to considered deprecated, and possibly
411 removed/changed in future releases.)
415 chomp() and chop() are now overridable. Note, however, that their
416 prototype (as given by C<prototype("CORE::chomp")> is undefined,
417 because it cannot be expressed and therefore one cannot really write
418 replacements to override these builtins.
422 END blocks are now run even if you exit/die in a BEGIN block.
423 Internally, the execution of END blocks is now controlled by
424 PL_exit_flags & PERL_EXIT_DESTRUCT_END. This enables the new
425 behaviour for Perl embedders. This will default in 5.10. See
430 Formats now support zero-padded decimal fields.
434 Lvalue subroutines can now return C<undef> in list context.
435 However, the lvalue subroutine feature still remains experimental.
439 A lost warning "Can't declare ... dereference in my" has been
440 restored (Perl had it earlier but it became lost in later releases.)
444 A new special regular expression variable has been introduced:
445 C<$^N>, which contains the most-recently closed group (submatch).
449 C<no Module;> now works even if there is no "sub unimport" in the Module.
453 The numerical comparison operators return C<undef> if either operand
454 is a NaN. Previously the behaviour was unspecified.
458 The following builtin functions are now overridable: each(), keys(),
459 pop(), push(), shift(), splice(), unshift().
463 C<pack() / unpack()> now can group template letters with C<()> and then
464 apply repetition/count modifiers on the groups.
468 C<pack() / unpack()> can now process the Perl internal numeric types:
469 IVs, UVs, NVs-- and also long doubles, if supported by the platform.
470 The template letters are C<j>, C<J>, C<F>, and C<D>.
474 C<pack('U0a*', ...)> can now be used to force a string to UTF8.
478 my __PACKAGE__ $obj now works.
482 The printf() and sprintf() now support parameter reordering using the
483 C<%\d+\$> and C<*\d+\$> syntaxes. For example
485 print "%2\$s %1\$s\n", "foo", "bar";
487 will print "bar foo\n". This feature helps in writing
488 internationalised software, and in general when the order
489 of the parameters can vary.
493 prototype(\&) is now available.
497 prototype(\[$@%&]) is now available to implicitly create references
498 (useful for example if you want to emulate the tie() interface).
502 A new command-line option, C<-t> is available. It is the
503 little brother of C<-T>: instead of dying on taint violations,
504 lexical warnings are given. B<This is only meant as a temporary
505 debugging aid while securing the code of old legacy applications.
506 This is not a substitute for -T.>
510 In other taint news, the C<exec LIST> and C<system LIST> have now been
511 considered too risky (think C<exec @ARGV>: it can start any program
512 with any arguments), and now the said forms cause a warning.
513 You should carefully launder the arguments to guarantee their
514 validity. In future releases of Perl the forms will become fatal
515 errors so consider starting laundering now.
519 If tr/// is just counting characters, it doesn't attempt to
524 untie() will now call an UNTIE() hook if it exists. See L<perltie>
529 L<utime> now supports C<utime undef, undef, @files> to change the
530 file timestamps to the current time.
534 The rules for allowing underscores (underbars) in numeric constants
535 have been relaxed and simplified: now you can have an underscore
536 simply B<between digits>.
540 Rather than relying on C's argv[0] (which may not contain a full pathname)
541 where possible $^X is now set by asking the operating system.
542 (eg by reading F</proc/self/exe> on Linux, F</proc/curproc/file> on FreeBSD)
546 A new variable, C<${^TAINT}>, indicates whether taint mode is enabled.
550 You can now override the readline() builtin, and this overrides also
551 the <FILEHANDLE> angle bracket operator.
555 The command-line options -s and -F are now recognized on the shebang
560 Use of the C</c> match modifier without an accompanying C</g> modifier
561 elicits a new warning: C<Use of /c modifier is meaningless without /g>.
563 Use of C</c> in substitutions, even with C</g>, elicits
564 C<Use of /c modifier is meaningless in s///>.
566 Use of C</g> with C<split> elicits <Use of /g modifier is meaningless
571 =head1 Modules and Pragmata
573 =head2 New Modules and Pragmata
579 C<Attribute::Handlers> allows a class to define attribute handlers.
582 use Attribute::Handlers;
583 sub Wolf :ATTR(SCALAR) { print "howl!\n" }
585 # later, in some package using or inheriting from MyPack...
587 my MyPack $Fluffy : Wolf; # the attribute handler Wolf will be called
589 Both variables and routines can have attribute handlers. Handlers can
590 be specific to type (SCALAR, ARRAY, HASH, or CODE), or specific to the
591 exact compilation phase (BEGIN, CHECK, INIT, or END).
595 B<B::Concise> is a new compiler backend for walking the Perl syntax
596 tree, printing concise info about ops, from Stephen McCamant. The
597 output is highly customisable. See L<B::Concise>.
601 The new bignum, bigint, and bigrat pragmas implement transparent
602 bignum support (using the Math::BigInt, Math::BigFloat, and
603 Math::BigRat backends), by Tels.
607 C<Class::ISA> for reporting the search path for a class's ISA tree,
608 by Sean Burke, has been added. See L<Class::ISA>.
612 C<Cwd> has now a split personality: if possible, an XS extension is
613 used, (this will hopefully be faster, more secure, and more robust)
614 but if not possible, the familiar Perl implementation is used.
618 C<Devel::PPPort>, originally from Kenneth Albanowski and now
619 maintained by Paul Marquess, has been added. It is primarily used
620 by C<h2xs> to enhance portability of XS modules between different
625 C<Digest>, frontend module for calculating digests (checksums), from
626 Gisle Aas, has been added. See L<Digest>.
630 C<Digest::MD5> for calculating MD5 digests (checksums) as defined in
631 RFC 1321, from Gisle Aas, has been added. See L<Digest::MD5>.
633 use Digest::MD5 'md5_hex';
635 $digest = md5_hex("Thirsty Camel");
637 print $digest, "\n"; # 01d19d9d2045e005c3f1b80e8b164de1
639 NOTE: the C<MD5> backward compatibility module is deliberately not
640 included since its further use is discouraged.
644 C<Encode>, by Nick Ing-Simmons and Dan Kogai, provides a mechanism to
645 translate between different character encodings. Support for Unicode,
646 ISO-8859-1, and ASCII are compiled in to the module. Several other
647 encodings (like the rest of the ISO-8859, CP*/Win*, Mac, KOI8-R, three
648 variants EBCDIC, Chinese, Japanese, and Korean encodings) are included
649 and can be loaded at runtime. (For space considerations, the largest
650 Chinese encodings have been separated into their own CPAN module,
651 Encode::HanExtra, which Encode will use if available). See L<Encode>.
653 Any encoding supported by Encode module is also available to the
654 ":encoding()" layer if PerlIO is used.
658 C<Hash::Util> is the interface to the new I<restricted hashes>
659 feature. A restricted hash is restricted to a certain set of keys,
660 no keys outside the set can be added. Also individual keys can be
661 restricted so that the key cannot be deleted and the value cannot be
662 changed. (Implemented by Jeffrey Friedl, Nick Ing-Simmons, and
667 C<I18N::Langinfo> can be use to query locale information.
668 See L<I18N::Langinfo>.
672 C<I18N::LangTags> has functions for dealing with RFC3066-style
673 language tags, by Sean Burke. See L<I18N::LangTags>.
677 C<ExtUtils::Constant> is a new tool for extension writers for
678 generating XS code to import C header constants, by Nicholas Clark.
679 See L<ExtUtils::Constant>.
683 C<Filter::Simple> is an easy-to-use frontend to Filter::Util::Call,
684 from Damian Conway. See L<Filter::Simple>.
690 use Filter::Simple sub {
691 while (my ($from, $to) = splice @_, 0, 2) {
700 use MyFilter qr/red/ => 'green';
702 print "red\n"; # this code is filtered, will print "green\n"
703 print "bored\n"; # this code is filtered, will print "bogreen\n"
707 print "red\n"; # this code is not filtered, will print "red\n"
711 C<File::Temp> allows one to create temporary files and directories in
712 an easy, portable, and secure way, by Tim Jenness. See L<File::Temp>.
716 C<Filter::Util::Call> provides you with the framework to write
717 I<Source Filters> in Perl, from Paul Marquess. For most uses the
718 frontend Filter::Simple is to be preferred. See L<Filter::Util::Call>.
722 C<if> is a new pragma for conditional inclusion of modules, from
727 L<libnet> is a collection of perl5 modules related to network
728 programming, from Graham Barr. See L<Net::FTP>, L<Net::NNTP>,
729 L<Net::Ping>, L<Net::POP3>, L<Net::SMTP>, and L<Net::Time>.
731 Perl installation leaves libnet unconfigured, use F<libnetcfg> to configure.
735 C<List::Util> is a selection of general-utility list subroutines, like
736 sum(), min(), first(), and shuffle(), by Graham Barr. See L<List::Util>.
740 C<Locale::Constants>, C<Locale::Country>, C<Locale::Currency>, and
741 C<Locale::Language>, from Neil Bowers, have been added. They provide the
742 codes for various locale standards, such as "fr" for France, "usd" for
743 US Dollar, and "jp" for Japanese.
747 $country = code2country('jp'); # $country gets 'Japan'
748 $code = country2code('Norway'); # $code gets 'no'
750 See L<Locale::Constants>, L<Locale::Country>, L<Locale::Currency>,
751 and L<Locale::Language>.
755 C<Locale::Maketext> is localization framework from Sean Burke. See
756 L<Locale::Maketext>, and L<Locale::Maketext::TPJ13>. The latter is an
757 article about software localization, originally published in The Perl
758 Journal #13, republished here with kind permission.
762 Math::BigRat for big rational numbers, to accompany Math::BigInt and
763 Math::BigFloat, from Tels.
767 C<Memoize> can make your functions faster by trading space for time,
768 from Mark-Jason Dominus. See L<Memoize>.
772 C<MIME::Base64> allows you to encode data in base64, from Gisle Aas,
773 as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail
778 $encoded = encode_base64('Aladdin:open sesame');
779 $decoded = decode_base64($encoded);
781 print $encoded, "\n"; # "QWxhZGRpbjpvcGVuIHNlc2FtZQ=="
787 C<MIME::QuotedPrint> allows you to encode data in quoted-printable
788 encoding, as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail
789 Extensions)>, from Gisle Aas.
791 use MIME::QuotedPrint;
793 $encoded = encode_qp("Smiley in Unicode: \x{263a}");
794 $decoded = decode_qp($encoded);
796 print $encoded, "\n"; # "Smiley in Unicode: =263A"
798 MIME::QuotedPrint has been enhanced to provide the basic methods
799 necessary to use it with PerlIO::Via as in :
801 use MIME::QuotedPrint;
802 open($fh,">Via(MIME::QuotedPrint)",$path);
804 See L<MIME::QuotedPrint>.
808 C<NEXT> is pseudo-class for method redispatch, from Damian Conway.
813 C<open> is a new pragma for setting the default I/O disciplines
818 C<PerlIO::Scalar> provides the implementation of IO to "in memory"
819 Perl scalars as discussed above, from Nick Ing-Simmons. It also
820 serves as an example of a loadable PerlIO layer. Other future
821 possibilities include PerlIO::Array and PerlIO::Code.
822 See L<PerlIO::Scalar>.
826 C<PerlIO::Via> acts as a PerlIO layer and wraps PerlIO layer
827 functionality provided by a class (typically implemented in perl
828 code), from Nick Ing-Simmons.
830 use MIME::QuotedPrint;
831 open($fh,">Via(MIME::QuotedPrint)",$path);
833 This will automatically convert everything output to C<$fh>
834 to Quoted-Printable. See L<PerlIO::Via>.
838 C<Pod::ParseLink>, by Russ Allbery, has been added,
839 to parse LZ<><> links in pods as described in the new
844 C<Pod::Text::Overstrike>, by Joe Smith, has been added.
845 It converts POD data to formatted overstrike text.
846 See L<Pod::Text::Overstrike>.
850 C<Scalar::Util> is a selection of general-utility scalar subroutines,
851 like blessed(), reftype(), and tainted(). See L<Scalar::Util>.
855 C<sort> is a new pragma for controlling the behaviour of sort().
859 C<Storable> gives persistence to Perl data structures by allowing the
860 storage and retrieval of Perl data to and from files in a fast and
861 compact binary format, from Raphael Manfredi. See L<Storable>.
865 C<Switch>, from Damian Conway, has been added. Just by saying
869 you have C<switch> and C<case> available in Perl.
875 case 1 { print "number 1" }
876 case "a" { print "string a" }
877 case [1..10,42] { print "number in list" }
878 case (@array) { print "number in list" }
879 case /\w+/ { print "pattern" }
880 case qr/\w+/ { print "pattern" }
881 case (%hash) { print "entry in hash" }
882 case (\%hash) { print "entry in hash" }
883 case (\&sub) { print "arg to subroutine" }
884 else { print "previous case not true" }
891 C<Test::More> is yet another framework for writing test scripts,
892 more extensive than Test::Simple, by Michael Schwern. See L<Test::More>.
896 C<Test::Simple> has basic utilities for writing tests, by Michael
897 Schwern. See L<Test::Simple>.
901 C<Text::Balanced> has been added, for extracting delimited text
902 sequences from strings, from Damian Conway.
904 use Text::Balanced 'extract_delimited';
906 ($a, $b) = extract_delimited("'never say never', he never said", "'", '');
908 $a will be "'never say never'", $b will be ', he never said'.
910 In addition to extract_delimited() there are also extract_bracketed(),
911 extract_quotelike(), extract_codeblock(), extract_variable(),
912 extract_tagged(), extract_multiple(), gen_delimited_pat(), and
913 gen_extract_tagged(). With these you can implement rather advanced
914 parsing algorithms. See L<Text::Balanced>.
918 C<threads> is an interface to interpreter threads, by Arthur Bergman.
919 Interpreter threads (ithreads) is the new thread model introduced in
920 Perl 5.6 but only available as an internal interface for extension
921 writers (and for Win32 Perl for C<fork()> emulation). See L<threads>.
925 C<threads::shared> allows data sharing for interpreter threads, from
926 Arthur Bergman. In the ithreads model any data sharing between
927 threads must be explicit, as opposed to the old 5.005 thread model
928 where data sharing was implicit. See L<threads::shared>.
932 C<Tie::File>, by Mark-Jason Dominus, associates a Perl array with the
937 C<Tie::Memoize>, by Ilya Zakharevich, provides on-demand loaded hashes.
941 C<Tie::RefHash::Nestable>, by Edward Avis, allows storing hash
942 references (unlike the standard Tie::RefHash) The module is contained
943 within Tie::RefHash, see L<Tie::RefHash>.
947 C<Time::HiRes> provides high resolution timing (ualarm, usleep,
948 and gettimeofday), from Douglas E. Wegscheid. See L<Time::HiRes>.
952 C<Unicode::UCD> offers a querying interface to the Unicode Character
953 Database. See L<Unicode::UCD>.
957 C<Unicode::Collate> implements the UCA (Unicode Collation Algorithm)
958 for sorting Unicode strings, by SADAHIRO Tomoyuki. See L<Unicode::Collate>.
962 C<Unicode::Normalize> implements the various Unicode normalization
963 forms, by SADAHIRO Tomoyuki. See L<Unicode::Normalize>.
967 C<XS::Typemap>, by Tim Jenness, is a test extension that exercises XS
968 typemaps. Nothing gets installed but for extension writers the code
973 =head2 Updated And Improved Modules and Pragmata
979 The following independently supported modules have been updated to the
980 newest versions from CPAN: CGI, CPAN, DB_File, File::Spec, File::Temp,
981 Getopt::Long, Math::BigFloat, Math::BigInt, the podlators bundle
982 (Pod::Man, Pod::Text), Pod::LaTeX, Pod::Parser, Storable,
983 Term::ANSIColor, Test, Text-Tabs+Wrap.
987 The attributes::reftype() now works on tied arguments.
991 AutoLoader can now be disabled with C<no AutoLoader;>.
995 B::Deparse has been significantly enhanced. It now can deparse almost
996 all of the standard test suite (so that the tests still succeed).
997 There is a make target "test.deparse" for trying this out.
1001 Class::Struct can now define the classes in compile time.
1005 Class::Struct now assigns the array/hash element if the accessor
1006 is called with an array/hash element as the B<sole> argument.
1010 Data::Dumper has now an option to sort hashes.
1014 Data::Dumper has now an option to dump code references
1019 DB_File now supports newer Berkeley DB versions, among
1024 The English module can now be used without the infamous performance
1027 use English '-no_match_vars';
1029 (Assuming, of course, that one doesn't need the troublesome variables
1030 C<$`>, C<$&>, or C<$'>.) Also, introduced C<@LAST_MATCH_START> and
1031 C<@LAST_MATCH_END> English aliases for C<@-> and C<@+>.
1035 Fcntl, Socket, and Sys::Syslog have been rewritten to use the
1036 new-style constant dispatch section (see L<ExtUtils::Constant>).
1037 This means that they will be more robust and hopefully faster.
1041 File::Find now chdir()s correctly when chasing symbolic links.
1045 File::Find now has pre- and post-processing callbacks. It also
1046 correctly changes directories when chasing symbolic links. Callbacks
1047 (naughtily) exiting with "next;" instead of "return;" now work.
1051 File::Find is now (again) reentrant. It also has been made
1056 The warnings issued by File::Find now belong to their own category.
1057 You can enable/disable them with C<use/no warnings 'File::Find';>.
1061 File::Glob::glob() renamed to File::Glob::bsd_glob() to avoid
1062 prototype mismatch with CORE::glob().
1066 File::Glob now supports C<GLOB_LIMIT> constant to limit the size of
1067 the returned list of filenames.
1071 Devel::Peek now has an interface for the Perl memory statistics
1072 (this works only if you are using perl's malloc, and if you have
1073 compiled with debugging).
1077 IPC::Open3 now allows the use of numeric file descriptors.
1081 IO::Socket has now atmark() method, which returns true if the socket
1082 is positioned at the out-of-band mark. The method is also exportable
1083 as a sockatmark() function.
1087 IO::Socket::INET has support for ReusePort option (if your platform
1088 supports it). The Reuse option now has an alias, ReuseAddr. For clarity
1089 you may want to prefer ReuseAddr.
1093 IO::Socket::INET now supports C<LocalPort> of zero (usually meaning
1094 that the operating system will make one up.)
1098 use lib now works identically to @INC. Removing directories
1099 with 'no lib' now works.
1103 ExtUtils::MakeMaker now uses File::Spec internally, which hopefully
1104 leads into better portability.
1108 Math::BigFloat and Math::BigInt have undergone a full rewrite.
1109 They are now magnitudes faster, and they support various
1110 bignum libraries such as GMP and PARI as their backends.
1114 Math::Complex handles inf, NaN etc., better.
1118 Net::Ping has been muchly enhanced. Multihoming is now supported.
1119 There is now "external" protocol which uses Net::Ping::External module
1120 which runs external ping(1) and parses the output. A version of
1121 Net::Ping::External is available in CPAN.
1125 POSIX::sigaction() is now much more flexible and robust.
1126 You can now install coderef handlers, 'DEFAULT', and 'IGNORE'
1127 handlers, installing new handlers was not atomic.
1131 In Safe the C<%INC> now localised in a Safe compartment so that
1136 In SDBM_File on dosish platforms, some keys went missing because of
1137 lack of support for files with "holes". A workaround for the problem
1142 In Search::Dict one can now have a pre-processing hook for the
1143 lines being searched.
1147 The Shell module now has an OO interface.
1151 In Sys::Syslog there is now a failover mechanism that will go
1152 through alternative connection mechanisms until the message
1153 is successfully logged.
1157 The Test module has been significantly enhanced.
1161 Time::Local::timelocal() does not handle fractional seconds anymore.
1162 The rationale is that neither does localtime(), and timelocal() and
1163 localtime() are supposed to be inverses of each other.
1167 The vars pragma now supports declaring fully qualified variables.
1168 (Something that C<our()> does not and will not support.)
1172 The C<utf8::> name space (as in the pragma) provides various
1173 Perl-callable functions to provide low level access to Perl's
1174 internal Unicode representation. At the moment only length()
1175 has been implemented.
1179 =head1 Utility Changes
1185 Emacs perl mode (emacs/cperl-mode.el) has been updated to version
1190 F<emacs/e2ctags.pl> is now much faster.
1194 C<enc2xs> is a tool for people adding their own encodings to the
1199 C<h2ph> now supports C trigraphs.
1203 C<h2xs> now produces a template README.
1207 C<h2xs> now uses C<Devel::PPort> for better portability between
1208 different versions of Perl.
1212 C<h2xs> uses the new L<ExtUtils::Constant> module which will affect
1213 newly created extensions that define constants. Since the new code is
1214 more correct (if you have two constants where the first one is a
1215 prefix of the second one, the first constant B<never> gets defined),
1216 less lossy (it uses integers for integer constant, as opposed to the
1217 old code that used floating point numbers even for integer constants),
1218 and slightly faster, you might want to consider regenerating your
1219 extension code (the new scheme makes regenerating easy).
1220 L<h2xs> now also supports C trigraphs.
1224 C<libnetcfg> has been added to configure the libnet.
1228 C<perlbug> is now much more robust. It also sends the bug report to
1229 perl.org, not perl.com.
1233 C<perlcc> has been rewritten and its user interface (that is,
1234 command line) is much more like that of the UNIX C compiler, cc.
1235 (The perlbc tools has been removed. Use C<perlcc -B> instead.)
1239 C<perlivp> is a new Installation Verification Procedure utility
1240 for running any time after installing Perl.
1244 C<piconv> is an implementation of the character conversion utility
1245 C<iconv>, demonstrating the new Encode module.
1249 C<pod2html> now allows specifying a cache directory.
1253 C<s2p> has been completely rewritten in Perl. (It is in fact a full
1254 implementation of sed in Perl: you can use the sed functionality by
1255 using the C<psed> utility.)
1259 C<xsubpp> now understands POD documentation embedded in the *.xs files.
1263 C<xsubpp> now supports OUT keyword.
1267 =head1 New Documentation
1273 perl56delta details the changes between the 5.005 release and the
1278 perlclib documents the internal replacements for standard C library
1279 functions. (Interesting only for extension writers and Perl core
1284 perldebtut is a Perl debugging tutorial.
1288 perlebcdic contains considerations for running Perl on EBCDIC platforms.
1292 perlintro is a gentle introduction to Perl.
1296 perliol documents the internals of PerlIO with layers.
1300 perlmodstyle is a style guide for writing modules.
1304 perlnewmod tells about writing and submitting a new module.
1308 perlpacktut is a pack() tutorial.
1312 perlpod has been rewritten to be clearer and to record the best
1313 practices gathered over the years.
1317 perlpodspec is a more formal specification of the pod format,
1318 mainly of interest for writers of pod applications, not to
1319 people writing in pod.
1323 perlretut is a regular expression tutorial.
1327 perlrequick is a regular expressions quick-start guide.
1328 Yes, much quicker than perlretut.
1332 perltodo has been updated.
1336 perltootc has been renamed as perltooc (to not to conflict
1337 with perltoot in filesystems restricted to "8.3" names)
1341 perluniintro is an introduction to using Unicode in Perl.
1342 (perlunicode is more of a detailed reference and background
1347 perlutil explains the command line utilities packaged with the Perl
1352 The following platform-specific documents are available before
1353 the installation as README.I<platform>, and after the installation
1356 perlaix perlamiga perlapollo perlbeos perlbs2000
1357 perlce perlcygwin perldgux perldos perlepoc perlhpux
1358 perlhurd perlmachten perlmacos perlmint perlmpeix
1359 perlnetware perlos2 perlos390 perlplan9 perlqnx perlsolaris
1360 perltru64 perluts perlvmesa perlvms perlvos perlwin32
1366 The documentation for the POSIX-BC platform is called "BS2000", to avoid
1367 confusion with the Perl POSIX module.
1371 The documentation for the WinCE platform is called perlce (README.ce
1372 in the source code kit), to avoid confusion with the perlwin32
1373 documentation on 8.3-restricted filesystems.
1377 =head1 Performance Enhancements
1383 map() could get pathologically slow when the result list it generates
1384 is larger than the source list. The performance has been improved for
1389 sort() has been changed to use primarily mergesort internally as
1390 opposed to the earlier quicksort. For very small lists this may
1391 result in slightly slower sorting times, but in general the speedup
1392 should be at least 20%. Additional bonuses are that the worst case
1393 behaviour of sort() is now better (in computer science terms it now
1394 runs in time O(N log N), as opposed to quicksort's Theta(N**2)
1395 worst-case run time behaviour), and that sort() is now stable
1396 (meaning that elements with identical keys will stay ordered as they
1397 were before the sort). See the C<sort> pragma for information.
1399 The story in more detail: suppose you want to serve yourself a little
1402 @digits = ( 3,1,4,1,5,9 );
1404 A numerical sort of the digits will yield (1,1,3,4,5,9), as expected.
1405 Which C<1> comes first is hard to know, since one C<1> looks pretty
1406 much like any other. You can regard this as totally trivial,
1407 or somewhat profound. However, if you just want to sort the even
1408 digits ahead of the odd ones, then what will
1410 sort { ($a % 2) <=> ($b % 2) } @digits;
1412 yield? The only even digit, C<4>, will come first. But how about
1413 the odd numbers, which all compare equal? With the quicksort algorithm
1414 used to implement Perl 5.6 and earlier, the order of ties is left up
1415 to the sort. So, as you add more and more digits of Pi, the order
1416 in which the sorted even and odd digits appear will change.
1417 and, for sufficiently large slices of Pi, the quicksort algorithm
1418 in Perl 5.8 won't return the same results even if reinvoked with the
1419 same input. The justification for this rests with quicksort's
1420 worst case behavior. If you run
1422 sort { $a <=> $b } ( 1 .. $N , 1 .. $N );
1424 (something you might approximate if you wanted to merge two sorted
1425 arrays using sort), doubling $N doesn't just double the quicksort time,
1426 it I<quadruples> it. Quicksort has a worst case run time that can
1427 grow like N**2, so-called I<quadratic> behaviour, and it can happen
1428 on patterns that may well arise in normal use. You won't notice this
1429 for small arrays, but you I<will> notice it with larger arrays,
1430 and you may not live long enough for the sort to complete on arrays
1431 of a million elements. So the 5.8 quicksort scrambles large arrays
1432 before sorting them, as a statistical defence against quadratic behaviour.
1433 But that means if you sort the same large array twice, ties may be
1434 broken in different ways.
1436 Because of the unpredictability of tie-breaking order, and the quadratic
1437 worst-case behaviour, quicksort was I<almost> replaced completely with
1438 a stable mergesort. I<Stable> means that ties are broken to preserve
1439 the original order of appearance in the input array. So
1441 sort { ($a % 2) <=> ($b % 2) } (3,1,4,1,5,9);
1443 will yield (4,3,1,1,5,9), guaranteed. The even and odd numbers
1444 appear in the output in the same order they appeared in the input.
1445 Mergesort has worst case O(NlogN) behaviour, the best value
1446 attainable. And, ironically, this mergesort does particularly
1447 well where quicksort goes quadratic: mergesort sorts (1..$N, 1..$N)
1448 in O(N) time. But quicksort was rescued at the last moment because
1449 it is faster than mergesort on certain inputs and platforms.
1450 For example, if you really I<don't> care about the order of even
1451 and odd digits, quicksort will run in O(N) time; it's very good
1452 at sorting many repetitions of a small number of distinct elements.
1453 The quicksort divide and conquer strategy works well on platforms
1454 with relatively small, very fast, caches. Eventually, the problem gets
1455 whittled down to one that fits in the cache, from which point it
1456 benefits from the increased memory speed.
1458 Quicksort was rescued by implementing a sort pragma to control aspects
1459 of the sort. The B<stable> subpragma forces stable behaviour,
1460 regardless of algorithm. The B<_quicksort> and B<_mergesort>
1461 subpragmas are heavy-handed ways to select the underlying implementation.
1462 The leading C<_> is a reminder that these subpragmas may not survive
1463 beyond 5.8. More appropriate mechanisms for selecting the implementation
1464 exist, but they wouldn't have arrived in time to save quicksort.
1468 Hashes now use Bob Jenkins "One-at-a-Time" hashing key algorithm
1469 ( http://burtleburtle.net/bob/hash/doobs.html ). This algorithm is
1470 reasonably fast while producing a much better spread of values than
1471 the old hashing algorithm (originally by Chris Torek, later tweaked by
1472 Ilya Zakharevich). Hash values output from the algorithm on a hash of
1473 all 3-char printable ASCII keys comes much closer to passing the
1474 DIEHARD random number generation tests. According to perlbench, this
1475 change has not affected the overall speed of Perl.
1479 unshift() should now be noticeably faster.
1483 =head1 Installation and Configuration Improvements
1485 =head2 Generic Improvements
1491 INSTALL now explains how you can configure Perl to use 64-bit
1492 integers even on non-64-bit platforms.
1496 Policy.sh policy change: if you are reusing a Policy.sh file
1497 (see INSTALL) and you use Configure -Dprefix=/foo/bar and in the old
1498 Policy $prefix eq $siteprefix and $prefix eq $vendorprefix, all of
1499 them will now be changed to the new prefix, /foo/bar. (Previously
1500 only $prefix changed.) If you do not like this new behaviour,
1501 specify prefix, siteprefix, and vendorprefix explicitly.
1505 A new optional location for Perl libraries, otherlibdirs, is available.
1506 It can be used for example for vendor add-ons without disturbing Perl's
1507 own library directories.
1511 In many platforms the vendor-supplied 'cc' is too stripped-down to
1512 build Perl (basically, 'cc' doesn't do ANSI C). If this seems
1513 to be the case and 'cc' does not seem to be the GNU C compiler
1514 'gcc', an automatic attempt is made to find and use 'gcc' instead.
1518 gcc needs to closely track the operating system release to avoid
1519 build problems. If Configure finds that gcc was built for a different
1520 operating system release than is running, it now gives a clearly visible
1521 warning that there may be trouble ahead.
1525 If binary compatibility with the 5.005 release is not wanted, Configure
1526 no longer suggests including the 5.005 modules in @INC.
1530 Configure C<-S> can now run non-interactively.
1534 Configure support for pdp11-style memory models has been removed due
1539 configure.gnu now works with options with whitespace in them.
1543 installperl now outputs everything to STDERR.
1547 $Config{byteorder} is now computed dynamically (this is more robust
1548 with "fat binaries" where an executable image contains binaries for
1549 more than one binary platform.)
1553 Because PerlIO is now the default on most platforms, "-perlio" doesn't
1554 get appended to the $Config{archname} (also known as $^O) anymore.
1555 Instead, if you explicitly choose not to use perlio (Configure command
1556 line option -Uuseperlio), you will get "-stdio" appended.
1560 Another change related to the architecture name is that "-64all"
1561 (-Duse64bitall, or "maximally 64-bit") is appended only if your
1562 pointers are 64 bits wide. (To be exact, the use64bitall is ignored.)
1566 In AFS installations one can configure the root of the AFS to be
1567 somewhere else than the default F</afs> by using the Configure
1568 parameter C<-Dafsroot=/some/where/else>.
1572 APPLLIB_EXP, a less-know configuration-time definition, has been
1573 documented. It can be used to prepend site-specific directories
1574 to Perl's default search path (@INC), see INSTALL for information.
1578 The version of Berkeley DB used when the Perl (and, presumably, the
1579 DB_File extension) was built is now available as
1580 C<@Config{qw(db_version_major db_version_minor db_version_patch)}>
1581 from Perl and as C<DB_VERSION_MAJOR_CFG DB_VERSION_MINOR_CFG
1582 DB_VERSION_PATCH_CFG> from C.
1586 Building Berkeley DB3 for compatibility modes for DB, NDBM, and ODBM
1587 has been documented in INSTALL.
1591 If you have CPAN access (either network or a local copy such as a
1592 CD-ROM) you can during specify extra modules to Configure to build and
1593 install with Perl using the -Dextras=... option. See INSTALL for
1598 In addition to config.over a new override file, config.arch, is
1599 available. That is supposed to be used by hints file writers for
1600 architecture-wide changes (as opposed to config.over which is for
1605 If your file system supports symbolic links you can build Perl outside
1606 of the source directory by
1608 mkdir /tmp/perl/build/directory
1609 cd /tmp/perl/build/directory
1610 sh /path/to/perl/source/Configure -Dmksymlinks ...
1612 This will create in /tmp/perl/build/directory a tree of symbolic links
1613 pointing to files in /path/to/perl/source. The original files are left
1614 unaffected. After Configure has finished you can just say
1618 and Perl will be built and tested, all in /tmp/perl/build/directory.
1622 For Perl developers several new make targets for profiling
1623 and debugging have been added, see L<perlhack>.
1629 Use of the F<gprof> tool to profile Perl has been documented in
1630 L<perlhack>. There is a make target called "perl.gprof" for
1631 generating a gprofiled Perl executable.
1635 If you have GCC 3, there is a make target called "perl.gcov" for
1636 creating a gcoved Perl executable for coverage analysis. See
1641 If you are on IRIX or Tru64 platforms, new profiling/debugging options
1642 have been added, see L<perlhack> for more information about pixie and
1649 Guidelines of how to construct minimal Perl installations have
1650 been added to INSTALL.
1654 The Thread extension is now not built at all under ithreads
1655 (C<Configure -Duseithreads>) because it wouldn't work anyway (the
1656 Thread extension requires being Configured with C<-Duse5005threads>).
1658 But note that the Thread.pm interface is now shared by both
1663 The Gconvert macro ($Config{d_Gconvert}) used by perl for stringifying
1664 floating-point numbers is now more picky about using sprintf %.*g
1665 rules for the conversion. Some platforms that used to use gcvt may
1666 now resort to the slower sprintf.
1670 =head2 New Or Improved Platforms
1672 For the list of platforms known to support Perl,
1673 see L<perlport/"Supported Platforms">.
1679 AIX dynamic loading should be now better supported.
1683 AIX should now work better with gcc, threads, and 64-bitness. Also the
1684 long doubles support in AIX should be better now. See L<perlaix>.
1688 After a long pause, AmigaOS has been verified to be happy with Perl.
1692 AtheOS ( http://www.atheos.cx/ ) is a new platform.
1696 BeOS has been reclaimed.
1700 DG/UX platform now supports the 5.005-style threads. See L<perldgux>.
1704 DYNIX/ptx platform (a.k.a. dynixptx) is supported at or near osvers 4.5.2.
1708 EBCDIC platforms (z/OS, also known as OS/390, POSIX-BC, and VM/ESA)
1709 have been regained. Many test suite tests still fail and the
1710 co-existence of Unicode and EBCDIC isn't quite settled, but the
1711 situation is much better than with Perl 5.6. See L<perlos390>,
1712 L<perlbs2000> (for POSIX-BC), and L<perlvmesa> for more information.
1716 Building perl with -Duseithreads or -Duse5005threads now works under
1717 HP-UX 10.20 (previously it only worked under 10.30 or later). You will
1718 need a thread library package installed. See README.hpux.
1722 MacOS Classic (MacPerl has of course been available since
1723 perl 5.004 but now the source code bases of standard Perl
1724 and MacPerl have been synchronised)
1728 MacOS X (or Darwin) should now be able to build Perl even on HFS+
1729 filesystems. (The case-insensitivity confused the Perl build process.)
1733 NCR MP-RAS is now supported.
1737 All the NetBSD specific patches (except for the installation
1738 specific ones) have been merged back to the main distribution.
1742 NetWare from Novell is now supported. See L<perlnetware>.
1746 NonStop-UX is now supported.
1750 NEC SUPER-UX is now supported.
1754 All the OpenBSD specific patches (except for the installation
1755 specific ones) have been merged back to the main distribution.
1759 Perl has been tested with the GNU pth userlevel thread package
1760 ( http://www.gnu.org/software/pth/pth.html ) . All but one thread
1761 test worked, and that one failure was because of test results arriving
1762 in unexpected order.
1766 Amdahl UTS UNIX mainframe platform is now supported.
1770 WinCE is now supported. See L<perlce>.
1774 z/OS (formerly known as OS/390, formerly known as MVS OE) has now
1775 support for dynamic loading. This is not selected by default,
1776 however, you must specify -Dusedl in the arguments of Configure.
1780 =head1 Selected Bug Fixes
1782 Numerous memory leaks and uninitialized memory accesses have been
1783 hunted down. Most importantly anonymous subs used to leak quite
1790 The autouse pragma didn't work for Multi::Part::Function::Names.
1794 caller() could cause core dumps in certain situations. Carp was sometimes
1795 affected by this problem.
1799 chop(@list) in list context returned the characters chopped in
1800 reverse order. This has been reversed to be in the right order.
1804 Configure no longer includes the DBM libraries (dbm, gdbm, db, ndbm)
1805 when building the Perl binary. The only exception to this is SunOS 4.x,
1810 The behaviour of non-decimal but numeric string constants such as
1811 "0x23" was platform-dependent: in some platforms that was seen as 35,
1812 in some as 0, in some as a floating point number (don't ask). This
1813 was caused by Perl using the operating system libraries in a situation
1814 where the result of the string to number conversion is undefined: now
1815 Perl consistently handles such strings as zero in numeric contexts.
1819 The order of DESTROYs has been made more predictable.
1823 Several debugger fixes: exit code now reflects the script exit code,
1824 condition C<"0"> now treated correctly, the C<d> command now checks
1825 line number, the C<$.> no longer gets corrupted, all debugger output
1826 now goes correctly to the socket if RemotePort is set.
1830 Perl 5.6.0 could emit spurious warnings about redefinition of dl_error()
1831 when statically building extensions into perl. This has been corrected.
1835 L<dprofpp> -R didn't work.
1839 C<*foo{FORMAT}> now works.
1842 Infinity is now recognized as a number.
1846 UNIVERSAL::isa no longer caches methods incorrectly. (This broke
1847 the Tk extension with 5.6.0.)
1851 Lexicals I: lexicals outside an eval "" weren't resolved
1852 correctly inside a subroutine definition inside the eval "" if they
1853 were not already referenced in the top level of the eval""ed code.
1857 Lexicals II: lexicals leaked at file scope into subroutines that
1858 were declared before the lexicals.
1862 Lexical warnings now propagating correctly between scopes
1863 and into C<eval "...">.
1867 C<use warnings qw(FATAL all)> did not work as intended. This has been
1872 warnings::enabled() now reports the state of $^W correctly if the caller
1873 isn't using lexical warnings.
1877 Line renumbering with eval and C<#line> now works.
1881 Fixed numerous memory leaks, especially in eval "".
1885 mkdir() now ignores trailing slashes in the directory name,
1886 as mandated by POSIX.
1890 Some versions of glibc have a broken modfl(). This affects builds
1891 with C<-Duselongdouble>. This version of Perl detects this brokenness
1892 and has a workaround for it. The glibc release 2.2.2 is known to have
1893 fixed the modfl() bug.
1897 Modulus of unsigned numbers now works (4063328477 % 65535 used to
1898 return 27406, instead of 27047).
1902 Some "not a number" warnings introduced in 5.6.0 eliminated to be
1903 more compatible with 5.005. Infinity is now recognised as a number.
1907 Numeric conversions did not recognize changes in the string value
1908 properly in certain circumstances.
1912 Attributes (like :shared) didn't work with our().
1916 our() variables will not cause "will not stay shared" warnings.
1920 "our" variables of the same name declared in two sibling blocks
1921 resulted in bogus warnings about "redeclaration" of the variables.
1922 The problem has been corrected.
1926 pack "Z" now correctly terminates the string with "\0".
1930 Fix password routines which in some shadow password platforms
1931 (e.g. HP-UX) caused getpwent() to return every other entry.
1935 The PERL5OPT environment variable (for passing command line arguments
1936 to Perl) didn't work for more than a single group of options.
1940 PERL5OPT with embedded spaces didn't work.
1944 printf() no longer resets the numeric locale to "C".
1948 C<qw(a\\b)> now parses correctly as C<'a\\b'>.
1952 pos() did not return the correct value within s///ge in earlier
1953 versions. This is now handled correctly.
1957 Printing quads (64-bit integers) with printf/sprintf now works
1958 without the q L ll prefixes (assuming you are on a quad-capable platform).
1962 Regular expressions on references and overloaded scalars now work.
1966 Right-hand side magic (GMAGIC) could in many cases such as string
1967 concatenation be invoked too many times.
1971 scalar() now forces scalar context even when used in void context.
1975 SOCKS support is now much more robust.
1979 sort() arguments are now compiled in the right wantarray context
1980 (they were accidentally using the context of the sort() itself).
1981 The comparison block is now run in scalar context, and the arguments
1982 to be sorted are always provided list context.
1986 Changed the POSIX character class C<[[:space:]]> to include the (very
1987 rarely used) vertical tab character. Added a new POSIX-ish character
1988 class C<[[:blank:]]> which stands for horizontal whitespace
1989 (currently, the space and the tab).
1993 The tainting behaviour of sprintf() has been rationalized. It does
1994 not taint the result of floating point formats anymore, making the
1995 behaviour consistent with that of string interpolation.
1999 Some cases of inconsistent taint propagation (such as within hash
2000 values) have been fixed.
2004 The RE engine found in Perl 5.6.0 accidentally pessimised certain kinds
2005 of simple pattern matches. These are now handled better.
2009 Regular expression debug output (whether through C<use re 'debug'>
2010 or via C<-Dr>) now looks better.
2014 Multi-line matches like C<"a\nxb\n" =~ /(?!\A)x/m> were flawed. The
2019 Use of $& could trigger a core dump under some situations. This
2024 The regular expression captured submatches ($1, $2, ...) are now
2025 more consistently unset if the match fails, instead of leaving false
2026 data lying around in them.
2030 readline() on files opened in "slurp" mode could return an extra "" at
2031 the end in certain situations. This has been corrected.
2035 Autovivification of symbolic references of special variables described
2036 in L<perlvar> (as in C<${$num}>) was accidentally disabled. This works
2041 Sys::Syslog ignored the C<LOG_AUTH> constant.
2045 All but the first argument of the IO syswrite() method are now optional.
2049 $AUTOLOAD, sort(), lock(), and spawning subprocesses
2050 in multiple threads simultaneously are now thread-safe.
2054 Tie::ARRAY SPLICE method was broken.
2058 Allow read-only string on left hand side of non-modifying tr///.
2062 If C<STDERR> is tied, warnings caused by C<warn> and C<die> now
2063 correctly pass to it.
2067 Several Unicode fixes.
2073 BOMs (byte order marks) in the beginning of Perl files
2074 (scripts, modules) should now be transparently skipped.
2075 UTF-16 (UCS-2) encoded Perl files should now be read correctly.
2079 The character tables have been updated to Unicode 3.1.1.
2083 Comparing with utf8 data does not magically upgrade non-utf8 data
2084 into utf8. (This was a problem for example if you were mixing data
2085 from I/O and Unicode data: your output might have got magically encoded
2090 Generating illegal Unicode code points like U+FFFE, or the UTF-16
2091 surrogates, now also generates an optional warning.
2095 C<IsAlnum>, C<IsAlpha>, and C<IsWord> now match titlecase.
2099 Concatenation with the C<.> operator or via variable interpolation,
2100 C<eq>, C<substr>, C<reverse>, C<quotemeta>, the C<x> operator,
2101 substitution with C<s///>, single-quoted UTF8, should now work.
2105 The C<tr///> operator now works. Note that the C<tr///CU>
2106 functionality has been removed (but see pack('U0', ...)).
2110 C<eval "v200"> now works.
2114 Perl 5.6.0 parsed m/\x{ab}/ incorrectly, leading to spurious warnings.
2115 This has been corrected.
2119 Zero entries were missing from the Unicode classes like C<IsDigit>.
2125 Large unsigned numbers (those above 2**31) could sometimes lose their
2126 unsignedness, causing bogus results in arithmetic operations.
2130 =head2 Platform Specific Changes and Fixes
2138 Perl now works on post-4.0 BSD/OSes.
2144 Setting C<$0> now works (as much as possible; see L<perlvar> for details).
2150 Numerous updates; currently synchronised with Cygwin 1.3.10.
2154 Previously DYNIX/ptx had problems in its Configure probe for non-blocking I/O.
2160 EPOC update after Perl 5.6.0. See README.epoc.
2166 Perl now works on post-3.0 FreeBSDs.
2172 README.hpux updated; C<Configure -Duse64bitall> now works.
2178 Numerous compilation flag and hint enhancements; accidental mixing
2179 of 32-bit and 64-bit libraries (a doomed attempt) made much harder.
2189 Long doubles should now work (see INSTALL).
2193 Linux previously had problems related to sockaddrlen when using
2194 accept(), revcfrom() (in Perl: recv()), getpeername(), and getsockname().
2202 Compilation of the standard Perl distribution in MacOS Classic should
2203 now work if you have the Metrowerks development environment and
2204 the missing Mac-specific toolkit bits. Contact the macperl mailing
2211 MPE/iX update after Perl 5.6.0. See README.mpeix.
2217 Perl now works on NetBSD/sparc.
2223 Now works with usethreads (see INSTALL).
2229 64-bitness using the Sun Workshop compiler now works.
2233 Tru64 (aka Digital UNIX, aka DEC OSF/1)
2235 The operating system version letter now recorded in $Config{osvers}.
2236 Allow compiling with gcc (previously explicitly forbidden). Compiling
2237 with gcc still not recommended because buggy code results, even with
2244 Fixed various alignment problems that lead into core dumps either
2245 during build or later; no longer dies on math errors at runtime;
2246 now using full quad integers (64 bits), previously was using
2247 only 46 bit integers for speed.
2253 chdir() now works better despite a CRT bug; now works with MULTIPLICITY
2254 (see INSTALL); now works with Perl's malloc.
2256 The tainting of C<%ENV> elements via C<keys> or C<values> was previously
2257 unimplemented. It now works as documented.
2259 The C<waitpid> emulation has been improved. The worst bug (now fixed)
2260 was that a pid of -1 would cause a wildcard search of all processes on
2261 the system. The most significant enhancement is that we can now
2262 usually get the completion status of a terminated process.
2264 POSIX-style signals are now emulated much better on VMS versions prior
2267 The C<system> function and backticks operator have improved
2268 functionality and better error handling.
2270 File access tests now use current process privileges rather than the
2271 user's default privileges, which could sometimes result in a mismatch
2272 between reported access and actual access.
2282 accept() no longer leaks memory.
2286 Borland C++ v5.5 is now a supported compiler that can build Perl.
2287 However, the generated binaries continue to be incompatible with those
2288 generated by the other supported compilers (GCC and Visual C++).
2292 Better chdir() return value for a non-existent directory.
2296 Duping socket handles with open(F, ">&MYSOCK") now works under Windows 9x.
2300 New %ENV entries now propagate to subprocesses.
2304 Current directory entries in %ENV are now correctly propagated to child
2309 $ENV{LIB} now used to search for libs under Visual C.
2313 fork() emulation has been improved in various ways, but still continues
2314 to be experimental. See L<perlfork> for known bugs and caveats.
2318 A failed (pseudo)fork now returns undef and sets errno to EAGAIN.
2322 Win32::GetCwd() correctly returns C:\ instead of C: when at the drive root.
2323 Other bugs in chdir() and Cwd::cwd() have also been fixed.
2327 HTML files will be installed in c:\perl\html instead of c:\perl\lib\pod\html
2331 The makefiles now provide a single switch to bulk-enable all the features
2332 enabled in ActiveState ActivePerl (a popular Win32 binary distribution).
2336 Allow REG_EXPAND_SZ keys in the registry.
2340 Can now send() from all threads, not just the first one.
2344 Fake signal handling reenabled, bugs and all.
2348 %SIG has been enabled under USE_ITHREADS, but its use is completely
2349 unsupported under all configurations.
2353 Less stack reserved per thread so that more threads can run
2354 concurrently. (Still 16M per thread.)
2358 C<File::Spec->tmpdir()> now prefers C:/temp over /tmp
2359 (works better when perl is running as service).
2363 Better UNC path handling under ithreads.
2367 wait(), waitpid() and backticks now return the correct exit status under
2372 winsock handle leak fixed.
2376 The Perl parser has been stress tested using both random input and
2377 Markov chain input and the few found crashes and lockups have been
2384 =head1 New or Changed Diagnostics
2390 The lexical warnings category "deprecated" is no longer a sub-category
2391 of the "syntax" category. It is now a top-level category in its own
2396 All regular expression compilation error messages are now hopefully
2397 easier to understand both because the error message now comes before
2398 the failed regex and because the point of failure is now clearly
2399 marked by a C<E<lt>-- HERE> marker.
2403 The various "opened only for", "on closed", "never opened" warnings
2404 drop the C<main::> prefix for filehandles in the C<main> package,
2405 for example C<STDIN> instead of C<main::STDIN>.
2409 The "Unrecognized escape" warning has been extended to include C<\8>,
2410 C<\9>, and C<\_>. There is no need to escape any of the C<\w> characters.
2414 Two new debugging options have been added: if you have compiled your
2415 Perl with debugging, you can use the -DT and -DR options to trace
2416 tokenising and to add reference counts to displaying variables,
2421 The debugger (perl5db.pl) has been modified to present a more
2422 consistent commands interface, via (CommandSet=580). perl5db.t was
2423 also added to test the changes, and as a placeholder for further tests.
2429 The debugger has a new C<dumpDepth> option to control the maximum
2430 depth to which nested structures are dumped. The C<x> command has
2431 been extended so that C<x N EXPR> dumps out the value of I<EXPR> to a
2432 depth of at most I<N> levels.
2436 The debugger can now show lexical variables if you have the CPAN
2437 module PadWalker installed.
2441 If an attempt to use a (non-blessed) reference as an array index
2442 is made, a warning is given.
2446 C<push @a;> and C<unshift @a;> (with no values to push or unshift)
2447 now give a warning. This may be a problem for generated and evaled
2452 If you try to L<perlfunc/pack> a number less than 0 or larger than 255
2453 using the C<"C"> format you will get an optional warning. Similarly
2454 for the C<"c"> format and a number less than -128 or more than 127.
2458 Certain regex modifiers such as C<(?o)> make sense only if applied to
2459 the entire regex. You will get an optional warning if you try to do
2464 Using arrays or hashes as references (e.g. C<< %foo->{bar} >>
2465 has been deprecated for a while. Now you will get an optional warning.
2469 Using C<sort> in scalar context now issues an optional warning.
2470 This didn't do anything useful, as the sort was not performed.
2474 =head1 Changed Internals
2480 perlapi.pod (a companion to perlguts) now attempts to document the
2485 You can now build a really minimal perl called microperl.
2486 Building microperl does not require even running Configure;
2487 C<make -f Makefile.micro> should be enough. Beware: microperl makes
2488 many assumptions, some of which may be too bold; the resulting
2489 executable may crash or otherwise misbehave in wondrous ways.
2490 For careful hackers only.
2494 Added rsignal(), whichsig(), do_join(), op_clear, op_null,
2495 ptr_table_clear(), ptr_table_free(), sv_setref_uv(), and several UTF-8
2496 interfaces to the publicised API. For the full list of the available
2497 APIs see L<perlapi>.
2501 Made possible to propagate customised exceptions via croak()ing.
2505 Now xsubs can have attributes just like subs. (Well, at least the
2506 built-in attributes.)
2510 dTHR and djSP have been obsoleted; the former removed (because it's
2511 a no-op) and the latter replaced with dSP.
2515 PERL_OBJECT has been completely removed.
2519 The MAGIC constants (e.g. C<'P'>) have been macrofied
2520 (e.g. C<PERL_MAGIC_TIED>) for better source code readability
2521 and maintainability.
2525 The regex compiler now maintains a structure that identifies nodes in
2526 the compiled bytecode with the corresponding syntactic features of the
2527 original regex expression. The information is attached to the new
2528 C<offsets> member of the C<struct regexp>. See L<perldebguts> for more
2529 complete information.
2533 The C code has been made much more C<gcc -Wall> clean. Some warning
2534 messages still remain in some platforms, so if you are compiling with
2535 gcc you may see some warnings about dubious practices. The warnings
2536 are being worked on.
2540 F<perly.c>, F<sv.c>, and F<sv.h> have now been extensively commented.
2544 Documentation on how to use the Perl source repository has been added
2545 to F<Porting/repository.pod>.
2549 There are now several profiling make targets.
2553 =head1 Security Vulnerability Closed
2555 (This change was already made in 5.7.0 but bears repeating here.)
2557 A potential security vulnerability in the optional suidperl component
2558 of Perl was identified in August 2000. suidperl is neither built nor
2559 installed by default. As of November 2001 the only known vulnerable
2560 platform is Linux, most likely all Linux distributions. CERT and
2561 various vendors and distributors have been alerted about the vulnerability.
2562 See http://www.cpan.org/src/5.0/sperl-2000-08-05/sperl-2000-08-05.txt
2563 for more information.
2565 The problem was caused by Perl trying to report a suspected security
2566 exploit attempt using an external program, /bin/mail. On Linux
2567 platforms the /bin/mail program had an undocumented feature which
2568 when combined with suidperl gave access to a root shell, resulting in
2569 a serious compromise instead of reporting the exploit attempt. If you
2570 don't have /bin/mail, or if you have 'safe setuid scripts', or if
2571 suidperl is not installed, you are safe.
2573 The exploit attempt reporting feature has been completely removed from
2574 Perl 5.8.0 (and the maintenance release 5.6.1, and it was removed also
2575 from all the Perl 5.7 releases), so that particular vulnerability
2576 isn't there anymore. However, further security vulnerabilities are,
2577 unfortunately, always possible. The suidperl functionality is most
2578 probably going to be removed in Perl 5.10. In any case, suidperl
2579 should only be used by security experts who know exactly what they are
2580 doing and why they are using suidperl instead of some other solution
2581 such as sudo ( see http://www.courtesan.com/sudo/ ).
2585 Several new tests have been added, especially for the F<lib>
2586 subsection. There are now about 56 000 individual tests (spread over
2587 about 620 test scripts), in the regression suite (5.6.1 has about
2588 11700 tests, in 258 test scripts) Many of the new tests are introduced
2589 by the new modules, but still in general Perl is now more thoroughly
2592 Because of the large number of tests, running the regression suite
2593 will take considerably longer time than it used to: expect the suite
2594 to take up to 4-5 times longer to run than in perl 5.6. In a really
2595 fast machine you can hope to finish the suite in about 6-8 minutes
2598 The tests are now reported in a different order than in earlier Perls.
2599 (This happens because the test scripts from under t/lib have been moved
2600 to be closer to the library/extension they are testing.)
2602 =head1 Known Problems
2610 In AIX 4.2 Perl extensions that use C++ functions that use statics
2611 may have problems in that the statics are not getting initialized.
2612 In newer AIX releases this has been solved by linking Perl with
2613 the libC_r library, but unfortunately in AIX 4.2 the said library
2614 has an obscure bug where the various functions related to time
2615 (such as time() and gettimeofday()) return broken values, and
2616 therefore in AIX 4.2 Perl is not linked against the libC_r.
2620 vac 5.0.0.0 May Produce Buggy Code For Perl
2622 The AIX C compiler vac version 5.0.0.0 may produce buggy code,
2623 resulting in few random tests failing, but when the failing tests
2624 are run by hand, they succeed. We suggest upgrading to at least
2625 vac version 5.0.1.0, that has been known to compile Perl correctly.
2626 "lslpp -L|grep vac.C" will tell you the vac version. See README.aix.
2630 If building threaded Perl, you may get compilation warning from pp_sys.c:
2632 "pp_sys.c", line 4651.39: 1506-280 (W) Function argument assignment between types "unsigned char*" and "const void*" is not allowed.
2634 This is harmless; it is caused by the getnetbyaddr() and getnetbyaddr_r()
2635 having slightly different types for their first argument.
2639 =head2 Amiga Perl Invoking Mystery
2641 One cannot call Perl using the C<volume:> syntax, that is, C<perl -v>
2642 works, but for example C<bin:perl -v> doesn't. The exact reason isn't
2643 known but the current suspect is the F<ixemul> library.
2645 =head2 FreeBSD Failing locale Test 117 For ISO8859-15 Locales
2647 The ISO8859-15 locales may fail the locale test 117 in FreeBSD.
2648 This is caused by the characters \xFF (y with diaeresis) and \xBE
2649 (Y with diaeresis) not behaving correctly when being matched
2652 =head2 mod_perl 1.26 Doesn't Build With Threaded Perl
2654 Use mod_perl 1.27 or higher.
2656 =head2 lib/ftmp-security tests warn 'system possibly insecure'
2658 Don't panic. Read INSTALL 'make test' section instead.
2660 =head2 HP-UX lib/posix Subtest 9 Fails When LP64-Configured
2662 If perl is configured with -Duse64bitall, the successful result of the
2663 subtest 10 of lib/posix may arrive before the successful result of the
2664 subtest 9, which confuses the test harness so much that it thinks the
2667 =head2 Linux With Sfio Fails op/misc Test 48
2673 Please remember to set your environment variable LC_ALL to "C"
2674 (setenv LC_ALL C) before running "make test" to avoid a lot of
2675 warnings about the broken locales of Mac OS X.
2677 The following tests are known to fail:
2679 Failed Test Stat Wstat Total Fail Failed List of Failed
2680 -------------------------------------------------------------------------
2681 ../ext/DB_File/t/db-btree.t 0 11 ?? ?? % ??
2682 ../ext/DB_File/t/db-recno.t 149 3 2.01% 61 63 65
2683 ../ext/POSIX/t/posix.t 31 1 3.23% 10
2685 If you are building on a UFS partition, you will also probably see
2686 t/op/stat.t subtest #9 fail. This is caused by Darwin's UFS not
2687 supporting inode change time.
2689 =head2 op/sprintf tests 91, 129, and 130
2691 The op/sprintf tests 91, 129, and 130 are known to fail on some platforms.
2692 Examples include any platform using sfio, and Compaq/Tandem's NonStop-UX.
2694 The test 91 is known to fail at QNX6 (nto), because C<sprintf '%e',0>
2695 incorrectly produces C<0.000000e+0> instead of C<0.000000e+00>.
2697 For the tests 129 and 130 the failing platforms do not comply with
2698 the ANSI C Standard, line 19ff on page 134 of ANSI X3.159 1989 to
2699 be exact. (They produce something other than "1" and "-1" when
2700 formatting 0.6 and -0.6 using the printf format "%.0f", most often
2701 they produce "0" and "-0".)
2703 =head2 Failure of Thread (5.005-style) tests
2705 B<Note that support for 5.005-style threading remains experimental
2706 and practically unsupported.>
2708 The following tests are known to fail due to fundamental problems in
2709 the 5.005 threading implementation. These are not new failures--Perl
2710 5.005_0x has the same bugs, but didn't have these tests.
2712 ../ext/List/Util/t/first.t 255 65280 7 4 57.14% 2 5-7
2713 ../lib/English.t 2 512 54 2 3.70% 2-3
2714 ../lib/Filter/Simple/t/data.t 6 3 50.00% 1-3
2715 ../lib/Filter/Simple/t/filter_onl 9 3 33.33% 1-2 5
2716 ../lib/autouse.t 10 1 10.00% 4
2717 op/flip.t 15 1 6.67% 15
2719 These failures are unlikely to get fixed as the 5.005-style
2720 threads are considered fundamentally broken.
2724 Failed Test Stat Wstat Total Fail Failed List of Failed
2725 -------------------------------------------------------------------------
2726 ../ext/Socket/socketpair.t 1 256 45 1 2.22% 12
2727 ../lib/Math/Trig.t 26 1 3.85% 25
2728 ../lib/warnings.t 460 1 0.22% 425
2729 io/fs.t 36 1 2.78% 31
2730 op/numconvert.t 1440 13 0.90% 208 509-510
2731 657-658 665-666 829-830 989-990 1149-1150
2733 =head2 UNICOS and UNICOS/mk
2735 The io/fs test #31 is failing because in UNICOS and UNICOS/mk the Perl
2736 truncate() cannot be used to grow the size of filehandles, only to
2737 reduce the size. The workaround is to truncate files instead of
2746 During Configure the test
2748 Guessing which symbols your C compiler and preprocessor define...
2750 will probably fail with error messages like
2752 CC-20 cc: ERROR File = try.c, Line = 3
2753 The identifier "bad" is undefined.
2755 bad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K
2758 CC-65 cc: ERROR File = try.c, Line = 3
2759 A semicolon is expected at this point.
2761 This is caused by a bug in awk utility of UNICOS/mk. You can ignore
2762 the error, but it does cause a slight problem: you cannot fully
2763 benefit from the h2ph utility (see L<h2ph>) that can be used to
2764 convert C headers to Perl libraries, mainly used to be able to access
2765 from Perl the constants defined using C preprocessor, cpp. Because of
2766 the above error parts of the converted headers will be invisible.
2767 Luckily, these days the need for h2ph is rare.
2771 If building Perl with the interpreter threads (ithreads), the
2772 getgrent(), getgrnam(), and getgrgid() functions cannot return the
2773 list of the group members due to a bug in the multithreaded support of
2774 UNICOS/mk. What this means that in list context the functions will
2775 return only three values, not four.
2781 There are a few known test failures, see L<perluts>.
2785 There should be no reported test failures with a default configuration,
2786 though there are a number of tests marked TODO that point to areas
2787 needing further debugging and/or porting work.
2791 In multi-CPU boxes there are some problems with the I/O buffering:
2792 some output may appear twice. The following Win32 failures are known
2795 Failed Test Stat Wstat Total Fail Failed List of Failed
2796 -------------------------------------------------------------------------
2797 ..\ext/threads/t/end.t 6 4 66.67% 3-6
2799 =head2 Localising a Tied Variable Leaks Memory
2802 tie my %tie_hash => 'Tie::StdHash';
2806 local($tie_hash{Foo}) = 1; # leaks
2808 Code like the above is known to leak memory every time the local()
2811 =head2 z/OS (OS/390)
2813 z/OS has rather many test failures but the situation is actually
2814 better than it was in 5.6.0, it's just that so many new modules and
2815 tests have been added.
2817 Failed Test Stat Wstat Total Fail Failed List of Failed
2818 -------------------------------------------------------------------------
2819 ../ext/Data/Dumper/t/dumper.t 321 2 0.62% 311 314
2820 ../ext/IO/lib/IO/t/io_unix.t 5 4 80.00% 2-5
2821 ../lib/utf8.t 94 13 13.83% 27 30-31 43 46 73
2824 ../lib/Benchmark.t 1 256 159 1 0.63% 75
2825 ../lib/ExtUtils/t/Constant.t 27 19 70.37% 5-23
2826 ../lib/ExtUtils/t/Embed.t 9 9 100.00% 1-9
2827 op/pat.t 864 9 1.04% 242-243 665 776
2829 op/sprintf.t 224 3 1.34% 98 100 136
2830 op/tr.t 97 5 5.15% 63 71-74
2831 uni/fold.t 767 8 1.04% 25-26 62 169 196
2833 60 tests and 384 subtests skipped.
2835 =head2 Localising Tied Arrays and Hashes Is Broken
2839 doesn't work as one would expect: the old value is restored
2842 =head2 Self-tying of Arrays and Hashes Is Forbidden
2844 Self-tying of arrays and hashes is broken in rather deep and
2845 hard-to-fix ways. As a stop-gap measure to avoid people from getting
2846 frustrated at the mysterious results (core dumps, most often) it is
2847 for now forbidden (you will get a fatal error even from an attempt).
2849 =head2 Building Extensions Can Fail Because Of Largefiles
2851 Some extensions like mod_perl are known to have issues with
2852 `largefiles', a change brought by Perl 5.6.0 in which file offsets
2853 default to 64 bits wide, where supported. Modules may fail to compile
2854 at all or compile and work incorrectly. Currently there is no good
2855 solution for the problem, but Configure now provides appropriate
2856 non-largefile ccflags, ldflags, libswanted, and libs in the %Config
2857 hash (e.g., $Config{ccflags_nolargefiles}) so the extensions that are
2858 having problems can try configuring themselves without the
2859 largefileness. This is admittedly not a clean solution, and the
2860 solution may not even work at all. One potential failure is whether
2861 one can (or, if one can, whether it's a good idea) link together at
2862 all binaries with different ideas about file offsets, all this is
2865 =head2 Unicode Support on EBCDIC Still Spotty
2867 Though mostly working, Unicode support still has problem spots on
2868 EBCDIC platforms. One such known spot are the C<\p{}> and C<\P{}>
2869 regular expression constructs for code points less than 256: the
2870 pP are testing for Unicode code points, not knowing about EBCDIC.
2872 =head2 The Compiler Suite Is Still Experimental
2874 The compiler suite is slowly getting better but it continues to be
2875 highly experimental. Use in production environments is discouraged.
2877 =head2 The Long Double Support Is Still Experimental
2879 The ability to configure Perl's numbers to use "long doubles",
2880 floating point numbers of hopefully better accuracy, is still
2881 experimental. The implementations of long doubles are not yet
2882 widespread and the existing implementations are not quite mature
2883 or standardised, therefore trying to support them is a rare
2884 and moving target. The gain of more precision may also be offset
2885 by slowdown in computations (more bits to move around, and the
2886 operations are more likely to be executed by less optimised
2889 =head2 Seen In Perl 5.7 But Gone Now
2891 C<Time::Piece> (previously known as C<Time::Object>) was removed
2892 because it was felt that it didn't have enough value in it to be a
2893 core module. It is still a useful module, though, and is available
2896 =head1 Reporting Bugs
2898 If you find what you think is a bug, you might check the articles
2899 recently posted to the comp.lang.perl.misc newsgroup and the perl
2900 bug database at http://bugs.perl.org/ There may also be
2901 information at http://www.perl.com/ , the Perl Home Page.
2903 If you believe you have an unreported bug, please run the B<perlbug>
2904 program included with your release. Be sure to trim your bug down
2905 to a tiny but sufficient test case. Your bug report, along with the
2906 output of C<perl -V>, will be sent off to perlbug@perl.org to be
2907 analysed by the Perl porting team.
2911 The F<Changes> file for exhaustive details on what changed.
2913 The F<INSTALL> file for how to build Perl.
2915 The F<README> file for general stuff.
2917 The F<Artistic> and F<Copying> files for copyright information.
2921 Written by Jarkko Hietaniemi <F<jhi@iki.fi>>.