3 perldelta - what is new for perl v5.8.0
7 This document describes differences between the 5.6.0 release and the
10 =head1 Incompatible Changes
12 =head2 64-bit platforms and malloc
14 If your pointers are 64 bits wide, the Perl malloc is no more being
15 used because it simply does not work with 8-byte pointers. Also,
16 usually the system mallocs on such platforms are much better optimized
17 for such large memory models than the Perl malloc. Such platforms
18 include 64-bit Alpha, MIPS, HPPA, PPC, and Sparc.
20 =head2 AIX Dynaloading
22 The AIX dynaloading now uses in AIX releases 4.3 and newer the native
23 dlopen interface of AIX instead of the old emulated interface. This
24 change will probably break backward compatibility with compiled
25 modules. The change was made to make Perl more compliant with other
26 applications like modperl which are using the AIX native interface.
28 =head2 Socket Extension Dynamic in VMS
30 The Socket extension is now dynamically loaded instead of being
31 statically built in. This may or may not be a problem with ancient
32 TCP/IP stacks of VMS: we do not know since we weren't able to test
33 Perl in such configurations.
35 =head2 Different Definition of the Unicode Character Classes \p{In...}
37 As suggested by the Unicode consortium, the Unicode character classes
38 now prefer I<scripts> as opposed to I<blocks> (as defined by Unicode);
39 in Perl, when the C<\p{In....}> and the C<\p{In....}> regular expression
40 constructs are used. This has changed the definition of some of those
43 The difference between scripts and blocks is that scripts are the
44 glyphs used by a language or a group of languages, while the blocks
45 are more artificial groupings of 256 characters based on the Unicode
48 In general this change results in more inclusive Unicode character
49 classes, but changes to the other direction also do take place:
50 for example while the script C<Latin> includes all the Latin
51 characters and their various diacritic-adorned versions, it
52 does not include the various punctuation or digits (since they
53 are not solely C<Latin>).
55 Changes in the character class semantics may have happened if a script
56 and a block happen to have the same name, for example C<Hebrew>.
57 In such cases the script wins and C<\p{InHebrew}> now means the script
58 definition of Hebrew. The block definition in still available,
59 though, by appending C<Block> to the name: C<\p{InHebrewBlock}> means
60 what C<\p{InHebrew}> meant in perl 5.6.0. For the full list
61 of affected character classes, see L<perlunicode/Blocks>.
63 =head2 Perl Parser Stress Tested
65 The Perl parser has been stress tested using both random input and
66 Markov chain input and the few found crashes and lockups have been
71 The current user-visible implementation of pseudo-hashes (the weird
72 use of the first array element) is deprecated starting from Perl 5.8.0
73 and will be removed in Perl 5.10.0, and the feature will be
74 implemented differently. Not only is the current interface rather
75 ugly, but the current implementation slows down normal array and hash
76 use quite noticeably. The C<fields> pragma interface will remain
79 The syntaxes C<@a->[...]> and C<@h->{...}> have now been deprecated.
81 After years of trying the suidperl is considered to be too complex to
82 ever be considered truly secure. The suidperl functionality is likely
83 to be removed in a future release.
85 The C<package;> syntax (C<package> without an argument has been
86 deprecated. Its semantics were never that clear and its
87 implementation even less so. If you have used that feature to
88 disallow all but fully qualified variables, C<use strict;> instead.
94 The semantics of bless(REF, REF) were unclear and until someone proves
95 it to make some sense, it is forbidden.
99 A reference to a reference now stringify as "REF(0x81485ec)" instead
100 of "SCALAR(0x81485ec)" in order to be more consistent with the return
105 The very dusty examples in the eg/ directory have been removed.
106 Suggestions for new shiny examples welcome but the main issue is that
107 the examples need to be documented, tested and (most importantly)
112 The obsolete chat2 library that should never have been allowed
113 to escape the laboratory has been decommissioned.
117 The unimplemented POSIX regex features [[.cc.]] and [[=c=]] are still
118 recognised but now cause fatal errors. The previous behaviour of
119 ignoring them by default and warning if requested was unacceptable
120 since it, in a way, falsely promised that the features could be used.
124 The (bogus) escape sequences \8 and \9 now give an optional warning
125 ("Unrecognized escape passed through"). There is no need to \-escape
130 lstat(FILEHANDLE) now gives a warning because the operation makes no sense.
131 In future releases this may become a fatal error.
135 The long deprecated uppercase aliases for the string comparison
136 operators (EQ, NE, LT, LE, GE, GT) have now been removed.
140 The regular expression captured submatches ($1, $2, ...) are now
141 more consistently unset if the match fails, instead of leaving false
142 data lying around in them.
146 The tr///C and tr///U features have been removed and will not return;
147 the interface was a mistake. Sorry about that. For similar
148 functionality, see pack('U0', ...) and pack('C0', ...).
152 Although "you shouldn't do that", it was possible to write code that
153 depends on Perl's hashed key order (Data::Dumper does this). The new
154 algorithm "One-at-a-Time" produces a different hashed key order.
155 More details are in L</"Performance Enhancements">.
159 The list of filenames from glob() (or <...>) is now by default sorted
160 alphabetically to be csh-compliant. (bsd_glob() does still sort platform
161 natively, ASCII or EBCDIC, unless GLOB_ALPHASORT is specified.)
165 =head1 Core Enhancements
167 =head2 AUTOLOAD Is Now Lvaluable
169 AUTOLOAD is now lvaluable, meaning that you can add the :lvalue attribute
170 to AUTOLOAD subroutines and you can assign to the AUTOLOAD return value.
172 =head2 PerlIO is Now The Default
178 IO is now by default done via PerlIO rather than system's "stdio".
179 PerlIO allows "layers" to be "pushed" onto a file handle to alter the
180 handle's behaviour. Layers can be specified at open time via 3-arg
183 open($fh,'>:crlf :utf8', $path) || ...
185 or on already opened handles via extended C<binmode>:
187 binmode($fh,':encoding(iso-8859-7)');
189 The built-in layers are: unix (low level read/write), stdio (as in
190 previous Perls), perlio (re-implementation of stdio buffering in a
191 portable manner), crlf (does CRLF <=> "\n" translation as on Win32,
192 but available on any platform). A mmap layer may be available if
193 platform supports it (mostly UNIXes).
195 Layers to be applied by default may be specified via the 'open' pragma.
197 See L</"Installation and Configuration Improvements"> for the effects
198 of PerlIO on your architecture name.
202 File handles can be marked as accepting Perl's internal encoding of Unicode
203 (UTF-8 or UTF-EBCDIC depending on platform) by a pseudo layer ":utf8" :
205 open($fh,">:utf8","Uni.txt");
207 Note for EBCDIC users: the pseudo layer ":utf8" is erroneously named
208 for you since it's not UTF-8 what you will be getting but instead
209 UTF-EBCDIC. See L<perlunicode>, L<utf8>, and
210 http://www.unicode.org/unicode/reports/tr16/ for more information.
211 In future releases this naming may change.
215 File handles can translate character encodings from/to Perl's internal
216 Unicode form on read/write via the ":encoding()" layer.
220 File handles can be opened to "in memory" files held in Perl scalars via:
222 open($fh,'>', \$variable) || ...
226 Anonymous temporary files are available without need to
227 'use FileHandle' or other module via
229 open($fh,"+>", undef) || ...
231 That is a literal undef, not an undefined value.
235 The list form of C<open> is now implemented for pipes (at least on UNIX):
237 open($fh,"-|", 'cat', '/etc/motd')
239 creates a pipe, and runs the equivalent of exec('cat', '/etc/motd') in
244 The following builtin functions are now overridable: chop(), chomp(),
245 each(), keys(), pop(), push(), shift(), splice(), unshift().
249 Formats now support zero-padded decimal fields.
253 Perl now tries internally to use integer values in numeric conversions
254 and basic arithmetics (+ - * /) if the arguments are integers, and
255 tries also to keep the results stored internally as integers.
256 This change leads into often slightly faster and always less lossy
257 arithmetics. (Previously Perl always preferred floating point numbers
262 The printf() and sprintf() now support parameter reordering using the
263 C<%\d+\$> and C<*\d+\$> syntaxes. For example
265 print "%2\$s %1\$s\n", "foo", "bar";
267 will print "bar foo\n"; This feature helps in writing
268 internationalised software.
272 Unicode in general should be now much more usable. Unicode can be
273 used in hash keys, Unicode in regular expressions should work now,
274 Unicode in tr/// should work now (though tr/// seems to be a
275 particularly tricky to get right, so you have been warned)
279 The Unicode Character Database coming with Perl has been upgraded
280 to Unicode 3.1. For more information, see http://www.unicode.org/,
281 and http://www.unicode.org/unicode/reports/tr27/
283 For developers interested in enhancing Perl's Unicode capabilities:
284 almost all the UCD files are included with the Perl distribution in
285 the lib/unicode subdirectory. The most notable omission, for space
286 considerations, is the Unihan database.
290 The Unicode character classes \p{Blank} and \p{SpacePerl} have been
291 added. "Blank" is like C isblank(), that is, it contains only
292 "horizontal whitespace" (the space character is, the newline isn't),
293 and the "SpacePerl" is the Unicode equivalent of C<\s> (\p{Space}
294 isn't, since that includes the vertical tabulator character, whereas
299 =head2 Signals Are Now Safe
301 Perl used to be fragile in that signals arriving at inopportune moments
302 could corrupt Perl's internal state.
304 =head2 Understanding of Numbers
306 In general a lot of fixing has happened in the area of Perl's
307 understanding of numbers, both integer and floating point. Since in
308 many systems the standard number parsing functions like C<strtoul()>
309 and C<atof()> seem to have bugs, Perl tries to work around their
310 deficiencies. This results hopefully in more accurate numbers.
316 C<perl -d:Module=arg,arg,arg> now works (previously one couldn't pass
317 in multiple arguments.)
321 END blocks are now run even if you exit/die in a BEGIN block.
322 Internally, the execution of END blocks is now controlled by
323 PL_exit_flags & PERL_EXIT_DESTRUCT_END. This enables the new
324 behaviour for Perl embedders. This will default in 5.10. See
329 Lexicals I: lexicals outside an eval "" weren't resolved
330 correctly inside a subroutine definition inside the eval "" if they
331 were not already referenced in the top level of the eval""ed code.
335 Lexicals II: lexicals leaked at file scope into subroutines that
336 were declared before the lexicals.
340 Lvalue subroutines can now return C<undef> in list context.
344 A new special regular expression variable has been introduced:
345 C<$^N>, which contains the most-recently closed group (submatch).
349 C<no Module;> now works even if there is no "sub unimport" in the Module.
353 The numerical comparison operators return C<undef> if either operand
354 is a NaN. Previously the behaviour was unspecified.
358 C<pack('U0a*', ...)> can now be used to force a string to UTF8.
362 my __PACKAGE__ $obj now works.
366 prototype(\&) is now available.
370 Right-hand side magic (GMAGIC) could in many cases such as string
371 concatenation be invoked too many times.
375 The rules for allowing underscores (underbars) in numeric constants
376 have been relaxed and simplified: now you can have an underscore
377 simply B<between digits>.
381 An UNTIE method is now available.
385 L<utime> now supports C<utime undef, undef, @files> to change the
386 file timestamps to the current time.
390 C<eval "v200"> now works.
394 =head1 Modules and Pragmata
402 C<Attribute::Handlers> allows a class to define attribute handlers.
405 use Attribute::Handlers;
406 sub Wolf :ATTR(SCALAR) { print "howl!\n" }
408 # later, in some package using or inheriting from MyPack...
410 my MyPack $Fluffy : Wolf; # the attribute handler Wolf will be called
412 Both variables and routines can have attribute handlers. Handlers can
413 be specific to type (SCALAR, ARRAY, HASH, or CODE), or specific to the
414 exact compilation phase (BEGIN, CHECK, INIT, or END).
418 B<B::Concise> is a new compiler backend for walking the Perl syntax
419 tree, printing concise info about ops, from Stephen McCamant. The
420 output is highly customisable. See L<B::Concise>.
424 C<Class::ISA> for reporting the search path for a class's ISA tree,
425 by Sean Burke, has been added. See L<Class::ISA>.
429 C<Cwd> has now a split personality: if possible, an XS extension is
430 used, (this will hopefully be faster, more secure, and more robust)
431 but if not possible, the familiar Perl implementation is used.
435 C<Digest>, frontend module for calculating digests (checksums), from
436 Gisle Aas, has been added. See L<Digest>.
440 C<Digest::MD5> for calculating MD5 digests (checksums) as defined in
441 RFC 1321, from Gisle Aas, has been added. See L<Digest::MD5>.
443 use Digest::MD5 'md5_hex';
445 $digest = md5_hex("Thirsty Camel");
447 print $digest, "\n"; # 01d19d9d2045e005c3f1b80e8b164de1
449 NOTE: the C<MD5> backward compatibility module is deliberately not
450 included since its use is discouraged.
454 C<Encode>, by Nick Ing-Simmons, provides a mechanism to translate
455 between different character encodings. Support for Unicode,
456 ISO-8859-*, ASCII, CP*, KOI8-R, and three variants of EBCDIC are
457 compiled in to the module. Several other encodings (like Japanese,
458 Chinese, and MacIntosh encodings) are included and will be loaded at
459 runtime. See L<Encode>.
461 Any encoding supported by Encode module is also available to the
462 ":encoding()" layer if PerlIO is used.
466 C<I18N::Langinfo> can be use to query locale information.
467 See L<I18N::Langinfo>.
471 C<I18N::LangTags> has functions for dealing with RFC3066-style
472 language tags, by Sean Burke. See <I18N::LangTags>.
476 C<ExtUtils::Constant> is a new tool for extension writers for
477 generating XS code to import C header constants, by Nicholas Clark.
478 See L<ExtUtils::Constant>.
482 C<Filter::Simple> is an easy-to-use frontend to Filter::Util::Call,
483 from Damian Conway. See L<Filter::Simple>.
489 use Filter::Simple sub {
490 while (my ($from, $to) = splice @_, 0, 2) {
499 use MyFilter qr/red/ => 'green';
501 print "red\n"; # this code is filtered, will print "green\n"
502 print "bored\n"; # this code is filtered, will print "bogreen\n"
506 print "red\n"; # this code is not filtered, will print "red\n"
510 C<File::Temp> allows one to create temporary files and directories in
511 an easy, portable, and secure way, by Tim Jenness. See L<File::Temp>.
515 C<Filter::Util::Call> provides you with the framework to write
516 I<Source Filters> in Perl, from Paul Marquess. For most uses the
517 frontend Filter::Simple is to be preferred. See L<Filter::Util::Call>.
521 L<libnet> is a collection of perl5 modules related to network
522 programming, from Graham Barr. See L<Net::FTP>, L<Net::NNTP>,
523 L<Net::Ping>, L<Net::POP3>, L<Net::SMTP>, and L<Net::Time>.
525 Perl installation leaves libnet unconfigured, use F<libnetcfg> to configure.
529 C<List::Util> is a selection of general-utility list subroutines, like
530 sum(), min(), first(), and shuffle(), by Graham barr. See L<List::Util>.
534 C<Locale::Constants>, C<Locale::Country>, C<Locale::Currency>, and
535 C<Locale::Language>, from Neil Bowers, have been added. They provide the
536 codes for various locale standards, such as "fr" for France, "usd" for
537 US Dollar, and "jp" for Japanese.
541 $country = code2country('jp'); # $country gets 'Japan'
542 $code = country2code('Norway'); # $code gets 'no'
544 See L<Locale::Constants>, L<Locale::Country>, L<Locale::Currency>,
545 and L<Locale::Language>.
549 C<Locale::Maketext> is localization framework from Sean Burke. See
550 L<Locale::Maketext>, and L<Locale::Maketext::TPJ13>. The latter is an
551 article about software localization, originally published in The Perl
552 Journal #13, republished here with kind permission.
556 C<Memoize> can make your functions faster by trading space for time,
557 from Mark-Jason Dominus. See L<Memoize>.
561 C<MIME::Base64> allows you to encode data in base64, from Gisle Aas,
562 as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail
567 $encoded = encode_base64('Aladdin:open sesame');
568 $decoded = decode_base64($encoded);
570 print $encoded, "\n"; # "QWxhZGRpbjpvcGVuIHNlc2FtZQ=="
576 C<MIME::QuotedPrint> allows you to encode data in quoted-printable
577 encoding, as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail
578 Extensions)>, from Gisle Aas.
580 use MIME::QuotedPrint;
582 $encoded = encode_qp("Smiley in Unicode: \x{263a}");
583 $decoded = decode_qp($encoded);
585 print $encoded, "\n"; # "Smiley in Unicode: =263A"
587 MIME::QuotedPrint has been enhanced to provide the basic methods
588 necessary to use it with PerlIO::Via as in :
590 use MIME::QuotedPrint;
591 open($fh,">Via(MIME::QuotedPrint)",$path)
593 See L<MIME::QuotedPrint>.
597 C<NEXT> is pseudo-class for method redispatch, from Damian Conway.
602 C<PerlIO::Scalar> provides the implementation of IO to "in memory"
603 Perl scalars as discussed above, from Nick Ing-Simmons. It also
604 serves as an example of a loadable PerlIO layer. Other future
605 possibilities include PerlIO::Array and PerlIO::Code.
606 See L<PerlIO::Scalar>.
610 C<PerlIO::Via> acts as a PerlIO layer and wraps PerlIO layer
611 functionality provided by a class (typically implemented in perl
612 code), from Nick Ing-Simmons.
614 use MIME::QuotedPrint;
615 open($fh,">Via(MIME::QuotedPrint)",$path)
617 This will automatically convert everything output to C<$fh>
618 to Quoted-Printable. See L<PerlIO::Via>.
622 C<Pod::Text::Overstrike>, by Joe Smith, has been added.
623 It converts POD data to formatted overstrike text.
624 See L<Pod::Text::Overstrike>.
628 C<Scalar::Util> is a selection of general-utility scalar subroutines,
629 like blessed(), reftype(), and tainted(). See L<Scalar::Util>.
633 C<Storable> gives persistence to Perl data structures by allowing the
634 storage and retrieval of Perl data to and from files in a fast and
635 compact binary format, from Raphael Manfredi. See L<Storable>.
639 C<Switch>, from Damian Conway, has been added. Just by saying
643 you have C<switch> and C<case> available in Perl.
649 case 1 { print "number 1" }
650 case "a" { print "string a" }
651 case [1..10,42] { print "number in list" }
652 case (@array) { print "number in list" }
653 case /\w+/ { print "pattern" }
654 case qr/\w+/ { print "pattern" }
655 case (%hash) { print "entry in hash" }
656 case (\%hash) { print "entry in hash" }
657 case (\&sub) { print "arg to subroutine" }
658 else { print "previous case not true" }
665 C<Test::More> is yet another framework for writing test scripts,
666 more extensive than Test::Simple, by Michael Schwern. See L<Test::More>.
670 C<Test::Simple> has the- basic utilities for writing tests, by Michael
671 Schwern. See L<Test::Simple>.
675 C<Text::Balanced> has been added, for extracting delimited text
676 sequences from strings, from Damian Conway.
678 use Text::Balanced 'extract_delimited';
680 ($a, $b) = extract_delimited("'never say never', he never said", "'", '');
682 $a will be "'never say never'", $b will be ', he never said'.
684 In addition to extract_delimited() there are also extract_bracketed(),
685 extract_quotelike(), extract_codeblock(), extract_variable(),
686 extract_tagged(), extract_multiple(), gen_delimited_pat(), and
687 gen_extract_tagged(). With these you can implement rather advanced
688 parsing algorithms. See L<Text::Balanced>.
692 C<threads> is an interface interpreter threads, by Arthur Bergman.
693 Interpreter threads (ithreads) is the new thread model introduced in
694 Perl 5.6 but then available only as an internal interface for
695 extension writers. See L<threads>.
699 C<threads::shared> allows data sharing for interpreter threads, from
700 Arthur Bergman. In the ithreads model any data sharing between
701 threads must be explicit, as opposed to the old 5.005 thread model
702 where data sharing was implicit. See L<threads::shared>.
706 C<Tie::RefHash::Nestable>, by Edward Avis, allows storing hash
707 references (unlike the standard Tie::RefHash) The module is contained
708 within Tie::RefHash, see L<Tie::RefHash>.
712 C<Time::HiRes> provides high resolution timing (ualarm, usleep,
713 and gettimeofday), from Douglas E. Wegscheid. See L<Time::HiRes>.
717 C<Unicode::UCD> offers a querying interface to the Unicode Character
718 Database. See L<Unicode::UCD>.
722 C<Unicode::Collate> implements the UCA (Unicode Collation Algorithm)
723 for sorting Unicode strings, by SADAHIRO Tomoyuki. See L<Unicode::Collate>.
727 C<Unicode::Normalize> implements the various Unicode normalization
728 forms, by SADAHIRO Tomoyuki. See L<Unicode::Normalize>.
732 C<XS::Typemap>, by Tim Jenness, is a test extension that exercises XS
733 typemaps. Nothing gets installed but for extension writers the code
738 =head2 Updated And Improved Modules and Pragmata
744 The following independently supported modules have been updated to the
745 newest versions from CPAN: CGI, CPAN, DB_File, File::Spec, File::Temp,
746 Getopt::Long, Math::BigFloat, Math::BigInt, the podlators bundle
747 (Pod::Man, Pod::Text), Pod::LaTeX, Pod::Parser, Storable,
748 Term::ANSIColor, Test, Text-Tabs+Wrap.
752 The attributes::reftype() now works on tied arguments.
756 AutoLoader can now be disabled with C<no AutoLoader;>,
760 Data::Dumper has now an option to sort hashes.
764 The English module can now be used without the infamous performance
767 use English '-no_performance_hit';
769 (Assuming, of course, that one doesn't need the troublesome variables
770 C<$`>, C<$&>, or C<$'>.) Also, introduced C<@LAST_MATCH_START> and
771 C<@LAST_MATCH_END> English aliases for C<@-> and C<@+>.
775 File::Find now has pre- and post-processing callbacks. It also
776 correctly changes directories when chasing symbolic links. Callbacks
777 (naughtily) exiting with "next;" instead of "return;" now work.
781 File::Glob::glob() renamed to File::Glob::bsd_glob() to avoid
782 prototype mismatch with CORE::glob().
786 IPC::Open3 now allows the use of numeric file descriptors.
790 use lib now works identically to @INC. Removing directories
791 with 'no lib' now works.
795 C<%INC> now localised in a Safe compartment so that use/require work.
799 The Shell module now has an OO interface.
803 B::Deparse has been significantly enhanced. It now can deparse almost
804 all of the standard test suite (so that the tests still succeed).
805 There is a make target "test.deparse" for trying this out.
809 Class::Struct can now define the classes in compile time.
813 Class::Struct now assigns the array/hash element if the accessor
814 is called with an array/hash element as the B<sole> argument.
818 Fcntl, Socket, and Sys::Syslog have been rewritten to use the
819 new-style constant dispatch section (see L<ExtUtils::Constant>).
823 File::Find is now (again) reentrant. It also has been made
828 File::Glob now supports C<GLOB_LIMIT> constant to limit the size of
829 the returned list of filenames.
833 Devel::Peek now has an interface for the Perl memory statistics
834 (this works only if you are using perl's malloc, and if you have
835 compiled with debugging).
839 IO::Socket has now atmark() method, which returns true if the socket
840 is positioned at the out-of-band mark. The method is also exportable
841 as a sockatmark() function.
845 IO::Socket::INET has support for ReusePort option (if your platform
846 supports it). The Reuse option now has an alias, ReuseAddr. For clarity
847 you may want to prefer ReuseAddr.
851 IO::Socket::INET now supports C<LocalPort> of zero (usually meaning
852 that the operating system will make one up.)
856 Math::BigFloat and Math::BigInt have undergone much fixing,
857 they are now magnitudes faster, and they support various
858 bignum libraries such as GMP and PARI as their backends.
862 Net::Ping has been enhanced. There is now "external" protocol which
863 uses Net::Ping::External module which runs external ping(1) and parses
864 the output. An alpha version of Net::Ping::External is available in
865 CPAN and in 5.7.2 the Net::Ping::External may be integrated to Perl.
869 The C<open> pragma allows layers other than ":raw" and ":crlf" when
874 POSIX::sigaction() is now much more flexible and robust.
875 You can now install coderef handlers, 'DEFAULT', and 'IGNORE'
876 handlers, installing new handlers was not atomic.
880 The Test module has been significantly enhanced.
884 The C<vars> pragma now supports declaring fully qualified variables.
885 (Something that C<our()> does not and will not support.)
889 The utf8:: name space (as in the pragma) provides various
890 Perl-callable functions to provide low level access to Perl's
891 internal Unicode representation. At the moment only length()
892 has been implemented.
896 =head1 Utility Changes
902 Emacs perl mode (emacs/cperl-mode.el) has been updated to version
907 F<emacs/e2ctags.pl> is now much faster.
911 h2xs now produces a template README.
915 L<h2ph> now supports C trigraphs.
919 L<h2xs> uses the new L<ExtUtils::Constant> module which will affect
920 newly created extensions that define constants. Since the new code is
921 more correct (if you have two constants where the first one is a
922 prefix of the second one, the first constant B<never> gets defined),
923 less lossy (it uses integers for integer constant, as opposed to the
924 old code that used floating point numbers even for integer constants),
925 and slightly faster, you might want to consider regenerating your
926 extension code (the new scheme makes regenerating easy).
927 L<h2xs> now also supports C trigraphs.
931 L<libnetcfg> has been added to configure the libnet.
935 perlbug is now much more robust. It also sends the bug report to
936 perl.org, not perl.com.
940 perlcc has been rewritten and its user interface (that is,
941 command line) is much more like that of the UNIX C compiler, cc.
945 perlivp is a new utility for doing Installation Verification
946 Procedure after installing Perl.
950 pod2html now allows specifying a cache directory.
954 s2p has been completely rewritten in Perl. (It is in fact a full
955 implementation of sed in Perl.)
959 xsubpp now understands POD documentation embedded in the *.xs files.
963 xsubpp now supports OUT keyword.
967 =head1 New Documentation
973 perl56delta details the changes between the 5.005 release and the
978 perlclib documents the internal replacements for standard C library
979 functions. (Interesting only for extension writers and Perl core
984 perldebtut is a Perl debugging tutorial.
988 perlebcdic contains considerations for running Perl on EBCDIC platforms.
989 Note that unfortunately EBCDIC platforms that used to supported back in
990 Perl 5.005 are still unsupported by Perl 5.7.0; the plan, however, is to
991 bring them back to the fold.
995 perlintro is a gentle introduction to Perl.
999 perliol documents the internals of PerlIO with layers.
1003 perlmodstyle is a style guide for writing modules.
1007 perlnewmod tells about writing and submitting a new module.
1011 perlpod has been rewritten to be clearer and to record the best
1012 practices gathered over the years.
1016 perlpodstyle is a more formal specification of the pod format,
1017 mainly of interest for writers of pod applications, not to
1018 people writing in pod.
1022 perlposix-bc explains using Perl on the POSIX-BC platform
1023 (an EBCDIC mainframe platform).
1027 perlretut is a regular expression tutorial.
1031 perlrequick is a regular expressions quick-start guide.
1032 Yes, much quicker than perlretut.
1036 perltodo has been updated.
1040 perltootc has been renamed as perltooc (to not to conflict
1041 with perltoot in filesystems restricted to "8.3" names)
1045 perluniintro is an introduction to using Unicode in Perl
1046 (perlunicode is more of a reference)
1050 perlutil explains the command line utilities packaged with the Perl
1055 The following platform-specific documents are available before
1056 the installation as README.I<platform>, and after the installation
1059 perlaix perlamiga perlapollo perlbeos perlbs2000
1060 perlce perlcygwin perldgux perldos perlepoc perlhpux
1061 perlhurd perlmachten perlmacos perlmint perlmpeix
1062 perlnetware perlos2 perlos390 perlplan9 perlqnx perlsolaris
1063 perltru64 perluts perlvmesa perlvms perlvos perlwin32
1069 The documentation for the POSIX-BC platform is called "BS2000", to avoid
1070 confusion with the Perl POSIX module.
1074 The documentation for the WinCE platform is called "CE", to avoid
1075 confusion with the perlwin32 documentation on 8.3-restricted filesystems.
1079 =head1 Performance Enhancements
1085 map() that changes the size of the list should now work faster.
1089 sort() has been changed to use mergesort internally as opposed to the
1090 earlier quicksort. For very small lists this may result in slightly
1091 slower sorting times, but in general the speedup should be at least
1092 20%. Additional bonuses are that the worst case behaviour of sort()
1093 is now better (in computer science terms it now runs in time O(N log N),
1094 as opposed to quicksort's Theta(N**2) worst-case run time behaviour),
1095 and that sort() is now stable (meaning that elements with identical
1096 keys will stay ordered as they were before the sort).
1100 Hashes now use Bob Jenkins "One-at-a-Time" hashing key algorithm
1101 (http://burtleburtle.net/bob/hash/doobs.html). This algorithm is
1102 reasonably fast while producing a much better spread of values than
1103 the old hashing algorithm (originally by Chris Torek, later tweaked by
1104 Ilya Zakharevich). Hash values output from the algorithm on a hash of
1105 all 3-char printable ASCII keys comes much closer to passing the
1106 DIEHARD random number generation tests. According to perlbench, this
1107 change has not affected the overall speed of Perl.
1111 unshift() should now be noticeably faster.
1115 =head1 Installation and Configuration Improvements
1117 =head2 Generic Improvements
1123 INSTALL now explains how you can configure Perl to use 64-bit
1124 integers even on non-64-bit platforms.
1128 Policy.sh policy change: if you are reusing a Policy.sh file
1129 (see INSTALL) and you use Configure -Dprefix=/foo/bar and in the old
1130 Policy $prefix eq $siteprefix and $prefix eq $vendorprefix, all of
1131 them will now be changed to the new prefix, /foo/bar. (Previously
1132 only $prefix changed.) If you do not like this new behaviour,
1133 specify prefix, siteprefix, and vendorprefix explicitly.
1137 A new optional location for Perl libraries, otherlibdirs, is available.
1138 It can be used for example for vendor add-ons without disturbing Perl's
1139 own library directories.
1143 In many platforms the vendor-supplied 'cc' is too stripped-down to
1144 build Perl (basically, 'cc' doesn't do ANSI C). If this seems
1145 to be the case and 'cc' does not seem to be the GNU C compiler
1146 'gcc', an automatic attempt is made to find and use 'gcc' instead.
1150 gcc needs to closely track the operating system release to avoid
1151 build problems. If Configure finds that gcc was built for a different
1152 operating system release than is running, it now gives a clearly visible
1153 warning that there may be trouble ahead.
1157 If binary compatibility with the 5.005 release is not wanted, Configure
1158 no longer suggests including the 5.005 modules in @INC.
1162 Configure C<-S> can now run non-interactively.
1166 configure.gnu now works with options with whitespace in them.
1170 installperl now outputs everything to STDERR.
1174 $Config{byteorder} is now computed dynamically (this is more robust
1175 with "fat binaries" where an executable image contains binaries for
1176 more than one binary platform.)
1180 Because PerlIO is now the default on most platforms, "-perlio" doesn't
1181 get appended to the $Config{archname} (also known as $^O) anymore.
1182 Instead, if you explicitly choose not to use perlio (Configure command
1183 line option -Uuseperlio), you will get "-stdio" appended.
1187 Another change related to the architecture name is that "-64all"
1188 (-Duse64bitall, or "maximally 64-bit") is appended only if your
1189 pointers are 64 bits wide. (To be exact, the use64bitall is ignored.)
1193 In AFS installations one can configure the root of the AFS to be
1194 somewhere else than the default F</afs> by using the Configure
1195 parameter C<-Dafsroot=/some/where/else>.
1199 APPLLIB_EXP, a less-know configuration-time definition, has been
1200 documented. It can be used to prepend site-specific directories
1201 to Perl's default search path (@INC), see INSTALL for information.
1205 The version of Berkeley DB used when the Perl (and, presumably, the
1206 DB_File extension) was built is now available as
1207 C<@Config{qw(db_version_major db_version_minor db_version_patch)}>
1208 from Perl and as C<DB_VERSION_MAJOR_CFG DB_VERSION_MINOR_CFG
1209 DB_VERSION_PATCH_CFG> from C.
1213 Building Berkeley DB3 for compatibility modes for DB, NDBM, and ODBM
1214 has been documented in INSTALL.
1218 If you have CPAN access (either network or a local copy such as a
1219 CD-ROM) you can during specify extra modules to Configure to build and
1220 install with Perl using the -Dextras=... option. See INSTALL for
1225 In addition to config.over a new override file, config.arch, is
1226 available. That is supposed to be used by hints file writers for
1227 architecture-wide changes (as opposed to config.over which is for
1232 For Perl developers several new make targets for profiling
1233 and debugging have been added, see L<perlhack>.
1239 Use of the F<gprof> tool to profile Perl has been documented in
1240 L<perlhack>. There is a make target called "perl.gprof" for
1241 generating a gprofiled Perl executable.
1245 If you have GCC 3, there is a make target called "perl.gcov" for
1246 creating a gcoved Perl executable for coverage analysis. See
1251 If you are on IRIX or Tru64 platforms, new profiling/debugging options
1252 have been added, see L<perlhack> for more information about pixie and
1259 Guidelines of how to construct minimal Perl installations have
1260 been added to INSTALL.
1264 The Thread extension is now not built at all under ithreads
1265 (C<Configure -Duseithreads>) because it wouldn't work anyway (the
1266 Thread extension requires being Configured with C<-Duse5005threads>).
1268 But note that the Thread.pm interface is now shared by both
1273 =head2 New Or Improved Platforms
1275 For the list of platforms known to support Perl,
1276 see L<perlport/"Supported Platforms">.
1282 AIX dynamic loading should be now better supported.
1286 AIX should now work better with gcc, threads, and 64-bitness. Also the
1287 long doubles support in AIX should be better now. See L<perlaix>.
1291 After a long pause, AmigaOS has been verified to be happy with Perl.
1295 AtheOS (http://www.atheos.cx/) is a new platform.
1299 DG/UX platform now supports the 5.005-style threads. See L<perldgux>.
1303 DYNIX/ptx platform (a.k.a. dynixptx) is supported at or near osvers 4.5.2.
1307 EBCDIC platforms (z/OS, also known as OS/390, POSIX-BC, and VM/ESA)
1308 have been regained. Many test suite tests still fail and the
1309 co-existence of Unicode and EBCDIC isn't quite settled, but the
1310 situation is much better than with Perl 5.6. See L<perlos390>,
1311 L<perlbs2000> (for POSIX-BC), and L<perlvmesa> for more information.
1315 Building perl with -Duseithreads or -Duse5005threads now works under
1316 HP-UX 10.20 (previously it only worked under 10.30 or later). You will
1317 need a thread library package installed. See README.hpux.
1321 MacOS Classic (MacPerl has of course been available since
1322 perl 5.004 but now the source code bases of standard Perl
1323 and MacPerl have been synchronised)
1327 MacOS X (or Darwin) should now be able to build Perl even on HFS+
1328 filesystems. (The case-insensitivity confused the Perl build process.)
1332 NCR MP-RAS is now supported.
1336 NetWare from Novell is now supported. See L<perlnetware>.
1340 NonStop-UX is now supported.
1344 Amdahl UTS UNIX mainframe platform is now supported.
1348 WinCE is now supported. See L<perlce>.
1352 z/OS (formerly known as OS/390, formerly known as MVS OE) has now
1353 support for dynamic loading. This is not selected by default,
1354 however, you must specify -Dusedl in the arguments of Configure.
1358 =head1 Selected Bug Fixes
1360 Numerous memory leaks and uninitialized memory accesses have been hunted down.
1361 Most importantly anonymous subs used to leak quite a bit.
1367 Several debugger fixes: exit code now reflects the script exit code,
1368 condition C<"0"> now treated correctly, the C<d> command now checks
1369 line number, the C<$.> no longer gets corrupted, all debugger output now
1370 goes correctly to the socket if RemotePort is set.
1374 C<*foo{FORMAT}> now works.
1378 Lexical warnings now propagating correctly between scopes.
1382 Line renumbering with eval and C<#line> now works.
1386 Fixed numerous memory leaks, especially in eval "".
1390 Modulus of unsigned numbers now works (4063328477 % 65535 used to
1391 return 27406, instead of 27047).
1395 Some "not a number" warnings introduced in 5.6.0 eliminated to be
1396 more compatible with 5.005. Infinity is now recognised as a number.
1400 our() variables will not cause "will not stay shared" warnings.
1404 pack "Z" now correctly terminates the string with "\0".
1408 Fix password routines which in some shadow password platforms
1409 (e.g. HP-UX) caused getpwent() to return every other entry.
1413 printf() no longer resets the numeric locale to "C".
1417 C<q(a\\b)> now parses correctly as C<'a\\b'>.
1421 Printing quads (64-bit integers) with printf/sprintf now works
1422 without the q L ll prefixes (assuming you are on a quad-capable platform).
1426 Regular expressions on references and overloaded scalars now work.
1430 scalar() now forces scalar context even when used in void context.
1434 sort() arguments are now compiled in the right wantarray context
1435 (they were accidentally using the context of the sort() itself).
1439 Changed the POSIX character class C<[[:space:]]> to include the (very
1440 rare) vertical tab character. Added a new POSIX-ish character class
1441 C<[[:blank:]]> which stands for horizontal whitespace (currently,
1442 the space and the tab).
1446 $AUTOLOAD, sort(), lock(), and spawning subprocesses
1447 in multiple threads simultaneously are now thread-safe.
1451 Allow read-only string on left hand side of non-modifying tr///.
1455 Several Unicode fixes (but still not perfect).
1461 BOMs (byte order marks) in the beginning of Perl files
1462 (scripts, modules) should now be transparently skipped.
1463 UTF-16 (UCS-2) encoded Perl files should now be read correctly.
1467 The character tables have been updated to Unicode 3.0.1.
1471 chr() for values greater than 127 now create utf8 when under use
1476 Comparing with utf8 data does not magically upgrade non-utf8 data into
1481 C<IsAlnum>, C<IsAlpha>, and C<IsWord> now match titlecase.
1485 Concatenation with the C<.> operator or via variable interpolation,
1486 C<eq>, C<substr>, C<reverse>, C<quotemeta>, the C<x> operator,
1487 substitution with C<s///>, single-quoted UTF8, should now work--in
1492 The C<tr///> operator now works I<slightly> better but is still rather
1493 broken. Note that the C<tr///CU> functionality has been removed (but
1494 see pack('U0', ...)).
1498 vec() now refuses to deal with characters >255.
1502 Zero entries were missing from the Unicode classes like C<IsDigit>.
1506 chop(@list) in list context returned the characters chopped in
1507 reverse order. This has been reversed to be in the right order.
1511 The order of DESTROYs has been made more predictable.
1515 mkdir() now ignores trailing slashes in the directory name,
1516 as mandated by POSIX.
1520 Attributes (like :shared) didn't work with our().
1524 The PERL5OPT environment variable (for passing command line arguments
1525 to Perl) didn't work for more than a single group of options.
1529 The tainting behaviour of sprintf() has been rationalized. It does
1530 not taint the result of floating point formats anymore, making the
1531 behaviour consistent with that of string interpolation.
1535 All but the first argument of the IO syswrite() method are now optional.
1539 Tie::ARRAY SPLICE method was broken.
1543 vec() now tries to work with characters <= 255 when possible, but it leaves
1544 higher character values in place. In that case, if vec() was used to modify
1545 the string, it is no longer considered to be utf8-encoded.
1549 The autouse pragma didn't work for Multi::Part::Function::Names.
1553 The behaviour of non-decimal but numeric string constants such as
1554 "0x23" was platform-dependent: in some platforms that was seen as 35,
1555 in some as 0, in some as a floating point number (don't ask). This
1556 was caused by Perl using the operating system libraries in a situation
1557 where the result of the string to number conversion is undefined: now
1558 Perl consistently handles such strings as zero in numeric contexts.
1562 L<dprofpp> -R didn't work.
1566 PERL5OPT with embedded spaces didn't work.
1570 L<Sys::Syslog> ignored the C<LOG_AUTH> constant.
1574 Some versions of glibc have a broken modfl(). This affects builds
1575 with C<-Duselongdouble>. This version of Perl detects this brokenness
1576 and has a workaround for it. The glibc release 2.2.2 is known to have
1577 fixed the modfl() bug.
1581 Linux previously had problems related to sockaddrlen when using
1582 accept(), revcfrom() (in Perl: recv()), getpeername(), and getsockname().
1586 Previously DYNIX/ptx had problems in its Configure probe for non-blocking I/O.
1596 Borland C++ v5.5 is now a supported compiler that can build Perl.
1597 However, the generated binaries continue to be incompatible with those
1598 generated by the other supported compilers (GCC and Visual C++).
1602 Win32::GetCwd() correctly returns C:\ instead of C: when at the drive root.
1603 Other bugs in chdir() and Cwd::cwd() have also been fixed.
1607 Duping socket handles with open(F, ">&MYSOCK") now works under Windows 9x.
1611 HTML files will be installed in c:\perl\html instead of c:\perl\lib\pod\html
1615 The makefiles now provide a single switch to bulk-enable all the features
1616 enabled in ActiveState ActivePerl (a popular Win32 binary distribution).
1624 UNIVERSAL::isa no longer caches methods incorrectly. (This broke
1625 the Tk extension with 5.6.0.)
1629 Configure no longer includes the DBM libraries (dbm, gdbm, db, ndbm)
1630 when building the Perl binary. The only exception to this is SunOS 4.x,
1635 SOCKS support is now much more robust.
1639 If your file system supports symbolic links you can build Perl outside
1640 of the source directory by
1642 mkdir /tmp/perl/build/directory
1643 cd /tmp/perl/build/directory
1644 sh /path/to/perl/source/Configure -Dmksymlinks ...
1646 This will create in /tmp/perl/build/directory a tree of symbolic links
1647 pointing to files in /path/to/perl/source. The original files are left
1648 unaffected. After Configure has finished you can just say
1652 and Perl will be built and tested, all in /tmp/perl/build/directory.
1656 =head2 Platform Specific Changes and Fixes
1664 Perl now works on post-4.0 BSD/OSes.
1670 Setting C<$0> now works (as much as possible; see perlvar for details).
1676 Numerous updates; currently synchronised with Cygwin 1.1.4.
1682 EPOC update after Perl 5.6.0. See README.epoc.
1688 Perl now works on post-3.0 FreeBSDs.
1694 README.hpux updated; C<Configure -Duse64bitall> now almost works.
1700 Numerous compilation flag and hint enhancements; accidental mixing
1701 of 32-bit and 64-bit libraries (a doomed attempt) made much harder.
1707 Long doubles should now work (see INSTALL).
1713 Compilation of the standard Perl distribution in MacOS Classic should
1714 now work if you have the Metrowerks development environment and
1715 the missing Mac-specific toolkit bits. Contact the macperl mailing
1722 MPE/iX update after Perl 5.6.0. See README.mpeix.
1728 Perl now works on NetBSD/sparc.
1734 Now works with usethreads (see INSTALL).
1740 64-bitness using the Sun Workshop compiler now works.
1744 Tru64 (aka Digital UNIX, aka DEC OSF/1)
1746 The operating system version letter now recorded in $Config{osvers}.
1747 Allow compiling with gcc (previously explicitly forbidden). Compiling
1748 with gcc still not recommended because buggy code results, even with
1755 Fixed various alignment problems that lead into core dumps either
1756 during build or later; no longer dies on math errors at runtime;
1757 now using full quad integers (64 bits), previously was using
1758 only 46 bit integers for speed.
1764 chdir() now works better despite a CRT bug; now works with MULTIPLICITY
1765 (see INSTALL); now works with Perl's malloc.
1775 accept() no longer leaks memory.
1779 Better chdir() return value for a non-existent directory.
1783 New %ENV entries now propagate to subprocesses.
1787 $ENV{LIB} now used to search for libs under Visual C.
1791 A failed (pseudo)fork now returns undef and sets errno to EAGAIN.
1795 Allow REG_EXPAND_SZ keys in the registry.
1799 Can now send() from all threads, not just the first one.
1803 Fake signal handling reenabled, bugs and all.
1807 Less stack reserved per thread so that more threads can run
1808 concurrently. (Still 16M per thread.)
1812 C<File::Spec->tmpdir()> now prefers C:/temp over /tmp
1813 (works better when perl is running as service).
1817 Better UNC path handling under ithreads.
1821 wait() and waitpid() now work much better.
1825 winsock handle leak fixed.
1831 =head1 New or Changed Diagnostics
1837 All regular expression compilation error messages are now hopefully
1838 easier to understand both because the error message now comes before
1839 the failed regex and because the point of failure is now clearly
1840 marked by a C<E<lt>-- HERE> marker.
1844 The various "opened only for", "on closed", "never opened" warnings
1845 drop the C<main::> prefix for filehandles in the C<main> package,
1846 for example C<STDIN> instead of <main::STDIN>.
1850 The "Unrecognized escape" warning has been extended to include C<\8>,
1851 C<\9>, and C<\_>. There is no need to escape any of the C<\w> characters.
1855 Two new debugging options have been added: if you have compiled your
1856 Perl with debugging, you can use the -DT and -DR options to trace
1857 tokenising and to add reference counts to displaying variables,
1862 If an attempt to use a (non-blessed) reference as an array index
1863 is made, a warning is given.
1867 C<push @a;> and C<unshift @a;> (with no values to push or unshift)
1868 now give a warning. This may be a problem for generated and evaled
1873 If you try to L<perlfunc/pack> a number less than 0 or larger than 255
1874 using the C<"C"> format you will get an optional warning. Similarly
1875 for the C<"c"> format and a number less than -128 or more than 127.
1879 Certain regex modifiers such as C<(?o)> make sense only if applied to
1880 the entire regex. You will an optional warning if you try to do otherwise.
1884 Using arrays or hashes as references (e.g. C<%foo->{bar}> has been
1885 deprecated for a while. Now you will get an optional warning.
1889 =head1 Changed Internals
1895 perlapi.pod (a companion to perlguts) now attempts to document the
1900 You can now build a really minimal perl called microperl.
1901 Building microperl does not require even running Configure;
1902 C<make -f Makefile.micro> should be enough. Beware: microperl makes
1903 many assumptions, some of which may be too bold; the resulting
1904 executable may crash or otherwise misbehave in wondrous ways.
1905 For careful hackers only.
1909 Added rsignal(), whichsig(), do_join() to the publicised API.
1913 Made possible to propagate customised exceptions via croak()ing.
1917 Added is_utf8_char(), is_utf8_string(), bytes_to_utf8(), and utf8_to_bytes().
1921 Now xsubs can have attributes just like subs.
1925 Some new APIs: ptr_table_clear(), ptr_table_free(), sv_setref_uv().
1926 For the full list of the available APIs see L<perlapi>.
1930 dTHR and djSP have been obsoleted; the former removed (because it's
1931 a no-op) and the latter replaced with dSP.
1935 PERL_OBJECT has been completely removed.
1939 The MAGIC constants (e.g. C<'P'>) have been macrofied
1940 (e.g. C<PERL_MAGIC_TIED>) for better source code readability
1941 and maintainability.
1945 The regex compiler now maintains a structure that identifies nodes in
1946 the compiled bytecode with the corresponding syntactic features of the
1947 original regex expression. The information is attached to the new
1948 C<offsets> member of the C<struct regexp>. See L<perldebguts> for more
1949 complete information.
1953 The C code has been made much more C<gcc -Wall> clean. Some warning
1954 messages still remain in some platforms, so if you are compiling with
1955 gcc you may see some warnings about dubious practices. The warnings
1956 are being worked on.
1960 F<perly.c>, F<sv.c>, and F<sv.h> have now been extensively commented.
1964 Documentation on how to use the Perl source repository has been added
1965 to F<Porting/repository.pod>.
1969 There are now several profiling make targets
1973 The C<op_clear> and C<op_null> are now exported.
1977 =head1 Security Vulnerability Closed
1979 (This change was already made in 5.7.0 but bears repeating here.)
1981 A potential security vulnerability in the optional suidperl component
1982 of Perl was identified in August 2000. suidperl is neither built nor
1983 installed by default. As of November 2001 the only known vulnerable
1984 platform is Linux, most likely all Linux distributions. CERT and
1985 various vendors and distributors have been alerted about the vulnerability.
1986 See http://www.cpan.org/src/5.0/sperl-2000-08-05/sperl-2000-08-05.txt
1987 for more information.
1989 The problem was caused by Perl trying to report a suspected security
1990 exploit attempt using an external program, /bin/mail. On Linux
1991 platforms the /bin/mail program had an undocumented feature which
1992 when combined with suidperl gave access to a root shell, resulting in
1993 a serious compromise instead of reporting the exploit attempt. If you
1994 don't have /bin/mail, or if you have 'safe setuid scripts', or if
1995 suidperl is not installed, you are safe.
1997 The exploit attempt reporting feature has been completely removed from
1998 Perl 5.8.0 (and the maintenance release 5.6.1, and it was removed also
1999 from all the Perl 5.7 releases), so that particular vulnerability
2000 isn't there anymore. However, further security vulnerabilities are,
2001 unfortunately, always possible. The suidperl functionality is most
2002 probably going to be removed in Perl 5.10. In any case, suidperl
2003 should only be used by security experts who know exactly what they are
2004 doing and why they are using suidperl instead of some other solution
2005 such as sudo (see http://www.courtesan.com/sudo/).
2009 Several new tests have been added, especially for the F<lib> subsection.
2011 The tests are now reported in a different order than in earlier Perls.
2012 (This happens because the test scripts from under t/lib have been moved
2013 to be closer to the library/extension they are testing.)
2015 =head1 Known Problems
2017 Note that unlike other sections in this document (which describe
2018 changes since 5.7.0) this section is cumulative containing known
2019 problems for all the 5.7 releases.
2027 In AIX 4.2 Perl extensions that use C++ functions that use statics
2028 may have problems in that the statics are not getting initialized.
2029 In newer AIX releases this has been solved by linking Perl with
2030 the libC_r library, but unfortunately in AIX 4.2 the said library
2031 has an obscure bug where the various functions related to time
2032 (such as time() and gettimeofday()) return broken values, and
2033 therefore in AIX 4.2 Perl is not linked against the libC_r.
2037 vac 5.0.0.0 May Produce Buggy Code For Perl
2039 The AIX C compiler vac version 5.0.0.0 may produce buggy code,
2040 resulting in few random tests failing, but when the failing tests
2041 are run by hand, they succeed. We suggest upgrading to at least
2042 vac version 5.0.1.0, that has been known to compile Perl correctly.
2043 "lslpp -L|grep vac.C" will tell you the vac version.
2047 =head2 Amiga Perl Invoking Mystery
2049 One cannot call Perl using the C<volume:> syntax, that is, C<perl -v>
2050 works, but for example C<bin:perl -v> doesn't. The exact reason is
2051 known but the current suspect is the F<ixemul> library.
2053 =head2 lib/ftmp-security tests warn 'system possibly insecure'
2055 Don't panic. Read INSTALL 'make test' section instead.
2057 =head2 Cygwin intermittent failures of lib/Memoize/t/expire_file 11 and 12
2059 The subtests 11 and 12 sometimes fail and sometimes work.
2061 =head2 HP-UX lib/io_multihomed Fails When LP64-Configured
2063 The lib/io_multihomed test may hang in HP-UX if Perl has been
2064 configured to be 64-bit. Because other 64-bit platforms do not hang in
2065 this test, HP-UX is suspect. All other tests pass in 64-bit HP-UX. The
2066 test attempts to create and connect to "multihomed" sockets (sockets
2067 which have multiple IP addresses).
2069 =head2 HP-UX lib/posix Subtest 9 Fails When LP64-Configured
2071 If perl is configured with -Duse64bitall, the successful result of the
2072 subtest 10 of lib/posix may arrive before the successful result of the
2073 subtest 9, which confuses the test harness so much that it thinks the
2076 =head2 Linux With Sfio Fails op/misc Test 48
2082 OS/390 has rather many test failures but the situation is actually
2083 better than it was in 5.6.0, it's just that so many new modules and
2084 tests have been added.
2086 Failed Test Stat Wstat Total Fail Failed List of Failed
2087 -----------------------------------------------------------------------------
2088 ../ext/B/Deparse.t 14 1 7.14% 14
2089 ../ext/B/Showlex.t 1 1 100.00% 1
2090 ../ext/Encode/Encode/Tcl.t 610 13 2.13% 592 594 596 598
2092 ../ext/IO/lib/IO/t/io_unix.t 113 28928 5 3 60.00% 3-5
2093 ../ext/POSIX/POSIX.t 29 1 3.45% 14
2094 ../ext/Storable/t/lock.t 255 65280 5 3 60.00% 3-5
2095 ../lib/locale.t 129 33024 117 19 16.24% 99-117
2096 ../lib/warnings.t 434 1 0.23% 75
2097 ../lib/ExtUtils.t 27 1 3.70% 25
2098 ../lib/Math/BigInt/t/bigintpm.t 1190 1 0.08% 1145
2099 ../lib/Unicode/UCD.t 81 48 59.26% 1-16 49-64 66-81
2100 ../lib/User/pwent.t 9 1 11.11% 4
2101 op/pat.t 660 6 0.91% 242-243 424-425
2103 op/split.t 0 9 ?? ?? % ??
2104 op/taint.t 174 3 1.72% 156 162 168
2105 op/tr.t 70 3 4.29% 50 58-59
2106 Failed 16/422 test scripts, 96.21% okay. 105/23251 subtests failed, 99.55% okay.
2108 =head2 op/sprintf tests 129 and 130
2110 The op/sprintf tests 129 and 130 are known to fail on some platforms.
2111 Examples include any platform using sfio, and Compaq/Tandem's NonStop-UX.
2112 The failing platforms do not comply with the ANSI C Standard, line
2113 19ff on page 134 of ANSI X3.159 1989 to be exact. (They produce
2114 something other than "1" and "-1" when formatting 0.6 and -0.6 using
2115 the printf format "%.0f", most often they produce "0" and "-0".)
2117 =head2 Failure of Thread tests
2119 B<Note that support for 5.005-style threading remains experimental.>
2121 The following tests are known to fail due to fundamental problems in
2122 the 5.005 threading implementation. These are not new failures--Perl
2123 5.005_0x has the same bugs, but didn't have these tests.
2126 t/lib/thr5005.t 19-20
2134 ext/POSIX/sigaction subtests 6 and 13 may fail.
2138 lib/ExtUtils may spuriously claim that subtest 28 failed,
2139 which is interesting since the test only has 27 tests.
2143 Numerous numerical test failures
2145 op/numconvert 209,210,217,218
2147 ext/Time/HiRes/HiRes 9
2148 lib/Math/BigInt/t/bigintpm 1145
2151 These tests fail because of yet unresolved floating point inaccuracies.
2157 There are a few known test failures, see L<perluts>.
2161 Rather many tests are failing in VMS but that actually more tests
2162 succeed in VMS than they used to, it's just that there are many,
2163 many more tests than there used to be.
2165 Here are the known failures from some compiler/platform combinations.
2167 DEC C V5.3-006 on OpenVMS VAX V6.2
2169 [-.ext.list.util.t]tainted..............FAILED on test 3
2170 [-.ext.posix]sigaction..................FAILED on test 7
2171 [-.ext.time.hires]hires.................FAILED on test 14
2172 [-.lib.file.find]taint..................FAILED on test 17
2173 [-.lib.math.bigint.t]bigintpm...........FAILED on test 1183
2174 [-.lib.test.simple.t]exit...............FAILED on test 1
2175 [.lib]vmsish............................FAILED on test 13
2176 [.op]sprintf............................FAILED on test 12
2177 Failed 8/399 tests, 91.23% okay.
2179 DEC C V6.0-001 on OpenVMS Alpha V7.2-1 and
2180 Compaq C V6.2-008 on OpenVMS Alpha V7.1
2182 [-.ext.list.util.t]tainted..............FAILED on test 3
2183 [-.lib.file.find]taint..................FAILED on test 17
2184 [-.lib.test.simple.t]exit...............FAILED on test 1
2185 [.lib]vmsish............................FAILED on test 13
2186 Failed 4/399 tests, 92.48% okay.
2188 Compaq C V6.4-005 on OpenVMS Alpha 7.2.1
2190 [-.ext.b]showlex........................FAILED on test 1
2191 [-.ext.list.util.t]tainted..............FAILED on test 3
2192 [-.lib.file.find]taint..................FAILED on test 17
2193 [-.lib.test.simple.t]exit...............FAILED on test 1
2194 [.lib]vmsish............................FAILED on test 13
2195 [.op]misc...............................FAILED on test 49
2196 Failed 6/401 tests, 92.77% okay.
2200 In multi-CPU boxes there are some problems with the I/O buffering:
2201 some output may appear twice.
2203 =head2 Localising a Tied Variable Leaks Memory
2206 tie my %tie_hash => 'Tie::StdHash';
2210 local($tie_hash{Foo}) = 1; # leaks
2212 Code like the above is known to leak memory every time the local()
2215 =head2 Self-tying of Arrays and Hashes Is Forbidden
2217 Self-tying of arrays and hashes is broken in rather deep and
2218 hard-to-fix ways. As a stop-gap measure to avoid people from getting
2219 frustrated at the mysterious results (core dumps, most often) it is
2220 for now forbidden (you will get a fatal error even from an attempt).
2222 =head2 Variable Attributes are not Currently Usable for Tieing
2224 This limitation will hopefully be fixed in future. (Subroutine
2225 attributes work fine for tieing, see L<Attribute::Handlers>).
2227 =head2 Building Extensions Can Fail Because Of Largefiles
2229 Some extensions like mod_perl are known to have issues with
2230 `largefiles', a change brought by Perl 5.6.0 in which file offsets
2231 default to 64 bits wide, where supported. Modules may fail to compile
2232 at all or compile and work incorrectly. Currently there is no good
2233 solution for the problem, but Configure now provides appropriate
2234 non-largefile ccflags, ldflags, libswanted, and libs in the %Config
2235 hash (e.g., $Config{ccflags_nolargefiles}) so the extensions that are
2236 having problems can try configuring themselves without the
2237 largefileness. This is admittedly not a clean solution, and the
2238 solution may not even work at all. One potential failure is whether
2239 one can (or, if one can, whether it's a good idea) link together at
2240 all binaries with different ideas about file offsets, all this is
2243 =head2 The Compiler Suite Is Still Experimental
2245 The compiler suite is slowly getting better but is nowhere near
2248 =head2 The Long Double Support is Still Experimental
2250 The ability to configure Perl's numbers to use "long doubles",
2251 floating point numbers of hopefully better accuracy, is still
2252 experimental. The implementations of long doubles are not yet
2253 widespread and the existing implementations are not quite mature
2254 or standardised, therefore trying to support them is a rare
2255 and moving target. The gain of more precision may also be offset
2256 by slowdown in computations (more bits to move around, and the
2257 operations are more likely to be executed by less optimised
2260 =head1 Reporting Bugs
2262 If you find what you think is a bug, you might check the articles
2263 recently posted to the comp.lang.perl.misc newsgroup and the perl
2264 bug database at http://bugs.perl.org. There may also be
2265 information at http://www.perl.com/perl/, the Perl Home Page.
2267 If you believe you have an unreported bug, please run the B<perlbug>
2268 program included with your release. Be sure to trim your bug down
2269 to a tiny but sufficient test case. Your bug report, along with the
2270 output of C<perl -V>, will be sent off to perlbug@perl.org to be
2271 analysed by the Perl porting team.
2275 The F<Changes> file for exhaustive details on what changed.
2277 The F<INSTALL> file for how to build Perl.
2279 The F<README> file for general stuff.
2281 The F<Artistic> and F<Copying> files for copyright information.
2285 Written by Jarkko Hietaniemi <F<jhi@iki.fi>>.