3 perl571delta - what's new for perl v5.7.2
7 This document describes differences between the 5.7.1 release and the
10 (To view the differences between the 5.6.0 release and the 5.7.0
11 release, see L<perl570delta>. To view the differences between the
12 5.7.0 release and the 5.7.1 release, see L<perl571delta>.)
14 =head1 Security Vulnerability Closed
16 (This change was already made in 5.7.0 but bears repeating here.)
18 A security vulnerability affecting all Perl versions prior to 5.6.1
19 was found in August 2000. The vulnerability does not affect default
20 installations and as far as is known affects only the Linux platform.
22 You should upgrade your Perl to 5.6.1 as soon as possible. Patches
23 for earlier releases exist but using the patches require full
24 recompilation from the source code anyway, so 5.6.1 is your best
27 See http://www.cpan.org/src/5.0/sperl-2000-08-05/sperl-2000-08-05.txt
30 =head1 Incompatible Changes
32 =head2 64-bit platforms and malloc
34 If your pointers are 64 bits wide, the Perl malloc is no more being
35 used because it simply does not work with 8-byte pointers. Also,
36 usually the system malloc on such platforms are much better optimized
37 for such large memory models than the Perl malloc.
39 =head2 AIX Dynaloading
41 The AIX dynaloading now uses the native dlopen interface of AIX,
42 (given the AIX is recent enough) instead of the old emulated interface.
43 This will probably break backward compatibility with compiled modules.
45 =head2 Socket Extension Dynamic in VMS
47 The Socket extension is now dynamically loaded instead of being
48 statically built in. This may or may not be a problem with ancient
49 TCP/IP stacks of VMS: we do not know since we weren't able to test
50 Perl in such configurations.
52 =head2 Different Definition of the Unicode Character Classes \p{In...}
54 As suggested by the Unicode consortium, the Unicode character classes
55 now prefer I<scripts> as opposed to I<blocks> (as defined by Unicode);
56 in Perl, when the C<\p{In....}> and the C<\p{In....}> regular expression
57 constructs are used. This has changed the definition of some of those
60 The difference between scripts and blocks is that scripts are the
61 glyphs used by a language or a group of languages, while the blocks
62 are more artificial groupings of 256 characters based on the Unicode
65 In general this change results in more inclusive Unicode character
66 classes, but changes to the other direction also do take place:
67 for example while the script C<Latin> includes all the Latin
68 characters and their various diacritic-adorned versions, it
69 does not include the various punctuation or digits (since they
70 are not solely C<Latin>).
72 Changes in the character class semantics may have happened if a script
73 and a block happen to have the same name, for example C<Hebrew>.
74 In such cases the script wins and C<\p{InHebrew}> now means the script
75 definition of Hebrew. The block definition in still available,
76 though, by appending C<Block> to the name: C<\p{InHebrewBlock}> means
77 what C<\p{InHebrew}> meant in perl 5.6.0. For the full list
78 of affected character classes, see L<perlunicode/Blocks>.
82 The current user-visible implementation of pseudo-hashes (the weird
83 use of the first array element) is deprecated starting from Perl 5.8.0
84 and will be removed in Perl 5.10.0, and the feature will be
85 implemented differently. Not only is the current interface rather
86 ugly, but the current implementation slows down normal array and hash
87 use quite noticeably. The 'fields' pragma interface will remain
90 The syntaxes C<@a->[...]> and C<@h->{...}> have now been deprecated.
92 The suidperl is also considered to be too much a risk to continue
93 maintaining and the suidperl code is likely to be removed in a future
96 =head1 Core Enhancements
98 In general a lot of fixing has happened in the area of Perl's
99 understanding of numbers, both integer and floating point. Since in
100 many systems the standard number parsing functions like C<strtoul()>
101 and C<atof()> seem to have bugs, Perl tries to work around their
102 deficiencies. This results hopefully in more accurate numbers.
108 The rules for allowing underscores (underbars) in numeric constants
109 have been relaxed and simplified: now you can have an underscore
114 GMAGIC (right-hand side magic) could in many cases such as string
115 concatenation be invoked too many times.
119 Lexicals I: lexicals outside an eval "" weren't resolved
120 correctly inside a subroutine definition inside the eval "" if they
121 were not already referenced in the top level of the eval""ed code.
125 Lexicals II: lexicals leaked at file scope into subroutines that
126 were declared before the lexicals.
130 Lvalue subroutines can now return C<undef> in list context.
134 The C<op_clear> and C<op_null> are now exported.
138 A new special regular expression variable has been introduced:
139 C<$^N>, which contains the most-recently closed group (submatch).
143 L<utime> now supports C<utime undef, undef, @files> to change the
144 file timestamps to the current time.
148 The Perl parser has been stress tested using both random input and
153 C<eval "v200"> now works.
157 VMS now works under PerlIO.
161 =head1 Modules and Pragmata
163 =head2 New Modules and Distributions
169 L<Attribute::Handlers> - Simpler definition of attribute handlers
173 L<ExtUtils::Constant> - generate XS code to import C header constants
177 L<I18N::LangTags> - functions for dealing with RFC3066-style language tags
181 L<libnet> - a collection of perl5 modules related to network programming
183 Perl installation leaves libnet unconfigured, use F<libnetcfg> to configure.
187 L<List::Util> - selection of general-utility list subroutines
191 L<Locale::Maketext> - framework for localization
195 L<Memoize> - Make your functions faster by trading space for time
199 L<NEXT> - pseudo-class for method redispatch
203 L<Scalar::Util> - selection of general-utility scalar subroutines
207 L<Test::More> - yet another framework for writing test scripts
211 L<Test::Simple> - Basic utilities for writing tests
215 L<Time::HiRes> - high resolution ualarm, usleep, and gettimeofday
219 L<Time::Piece> - Object Oriented time objects
221 (Previously known as L<Time::Object>.)
225 L<Time::Seconds> - a simple API to convert seconds to other date values
229 L<Unicode::UCD> - Unicode Character Database
233 =head2 Updated And Improved Modules and Pragmata
239 L<B::Deparse> module has been significantly enhanced. It now
240 can deparse almost all of the standard test suite (so that the
241 tests still succeed). There is a make target "test.deparse"
246 L<Class::Struct> now assigns the array/hash element if the accessor
247 is called with an array/hash element as the B<sole> argument.
251 L<Cwd> extension is now (even) faster.
255 L<DB_File> extension has been updated to version 1.77.
259 L<Fcntl>, L<Socket>, and L<Sys::Syslog> have been rewritten to use the
260 new-style constant dispatch section (see L<ExtUtils::Constant>).
264 L<File::Find> is now (again) reentrant. It also has been made
269 L<File::Glob> now supports C<GLOB_LIMIT> constant to limit the
270 size of the returned list of filenames.
274 L<IO::Socket::INET> now supports C<LocalPort> of zero (usually meaning
275 that the operating system will make one up.)
279 The L<vars> pragma now supports declaring fully qualified variables.
280 (Something that C<our()> does not and will not support.)
284 =head1 Utility Changes
290 The F<emacs/e2ctags.pl> is now much faster.
294 L<h2ph> now supports C trigraphs.
298 L<h2xs> uses the new L<ExtUtils::Constant> module which will affect
299 newly created extensions that define constants. Since the new code is
300 more correct (if you have two constants where the first one is a
301 prefix of the second one, the first constant B<never> gets defined),
302 less lossy (it uses integers for integer constant, as opposed to the
303 old code that used floating point numbers even for integer constants),
304 and slightly faster, you might want to consider regenerating your
305 extension code (the new scheme makes regenerating easy).
309 L<libnetcfg> has been added to configure the libnet.
313 The F<Pod::Html> (and thusly L<pod2html>) now allows specifying
318 =head1 New Documentation
324 L<Locale::Maketext::TPJ13> is an article about software localization,
325 originally published in The Perl Journal #13, republished here with
330 More README.$PLATFORM files have been converted into pod, which also
331 means that they also be installed as perl$PLATFORM documentation
332 files. The new files are L<perlapollo>, L<perlbeos>, L<perldgux>,
333 L<perlhurd>, L<perlmint>, L<perlnetware>, L<perlplan9>, L<perlqnx>,
338 The F<Todo> and F<Todo-5.6> files have been merged into L<perltodo>.
342 Use of the F<gprof> tool to profile Perl has been documented in
343 L<perlhack>. There is a make target "perl.gprof" for generating a
344 gprofiled Perl executable.
348 =head1 Installation and Configuration Improvements
350 =head2 New Or Improved Platforms
356 AIX should now work better with gcc. Also longdouble support in AIX
357 should be better now. See L<perlaix>.
361 AtheOS (http://www.atheos.cx/) is a new platform.
365 DG/UX platform now supports the 5.005-style threads. See L<perldgux>.
369 Several MacOS (Classic) portability patches have been applied. We
370 hope to get a fully working port by 5.8.0. (The remaining problems
371 relate to the changed IO model of Perl.) See L<perlmacos>.
375 MacOS X (or Darwin) should now be able to build Perl even on HFS+
376 filesystems. (The case-insensitivity confused the Perl build process.)
380 NetWare from Novell is now supported. See L<perlnetware>.
384 The Amdahl UTS UNIX mainframe platform is now supported.
388 =head2 Generic Improvements
394 In AFS installations one can configure the root of the AFS to be
395 somewhere else than the default F</afs> by using the Configure
396 parameter C<-Dafsroot=/some/where/else>.
400 The version of Berkeley DB used when the Perl (and, presumably, the
401 DB_File extension) was built is now available as
402 C<@Config{qw(db_version_major db_version_minor db_version_patch)}>
403 from Perl and as C<DB_VERSION_MAJOR_CFG DB_VERSION_MINOR_CFG
404 DB_VERSION_PATCH_CFG> from C.
408 The Thread extension is now not built at all under ithreads
409 (C<Configure -Duseithreads>) because it wouldn't work anyway (the
410 Thread extension requires being Configured with C<-Duse5005threads>).
414 The C<B::Deparse> compiler backend has been so significantly improved
415 that almost the whole Perl test suite passes after being deparsed. A
416 make target has been added to help in further testing: C<make test.deparse>.
420 =head1 Selected Bug Fixes
426 The autouse pragma didn't work for Multi::Part::Function::Names.
430 The behaviour of non-decimal but numeric string constants such as
431 "0x23" was platform-dependent: in some platforms that was seen as 35,
432 in some as 0, in some as a floating point number (don't ask). This
433 was caused by Perl using the operating system libraries in a situation
434 where the result of the string to number conversion is undefined: now
435 Perl consistently handles such strings as zero in numeric contexts.
439 L<dprofpp> -R didn't work.
443 PERL5OPT with embedded spaces didn't work.
447 L<Sys::Syslog> ignored the C<LOG_AUTH> constant.
451 =head2 Platform Specific Changes and Fixes
457 Some versions of glibc have a broken modfl(). This affects builds
458 with C<-Duselongdouble>. This version of Perl detects this brokenness
459 and has a workaround for it. The glibc release 2.2.2 is known to have
460 fixed the modfl() bug.
464 =head1 New or Changed Diagnostics
470 In the regular expression diagnostics the C<E<lt>E<lt> HERE> marker
471 introduced in 5.7.0 has been changed to be C<E<lt>-- HERE> since too
472 many people found the C<E<lt>E<lt>> to be too similar to here-document
477 If you try to L<perlfunc/pack> a number less than 0 or larger than 255
478 using the C<"C"> format you will get an optional warning. Similarly
479 for the C<"c"> format and a number less than -128 or more than 127.
483 Certain regex modifiers such as C<(?o)> make sense only if applied to
484 the entire regex. You will an optional warning if you try to do otherwise.
488 Using arrays or hashes as references (e.g. C<%foo->{bar}> has been
489 deprecated for a while. Now you will get an optional warning.
493 =head1 Source Code Enhancements
495 =head2 MAGIC constants
497 The MAGIC constants (e.g. C<'P'>) have been macrofied
498 (e.g. C<PERL_MAGIC_TIED>) for better source code readability
501 =head2 Better commented code
503 F<perly.c>, F<sv.c>, and F<sv.h> have now been extensively commented.
505 =head2 Regex pre-/post-compilation items matched up
507 The regex compiler now maintains a structure that identifies nodes in
508 the compiled bytecode with the corresponding syntactic features of the
509 original regex expression. The information is attached to the new
510 C<offsets> member of the C<struct regexp>. See L<perldebguts> for more
511 complete information.
515 The C code has been made much more C<gcc -Wall> clean. Some warning
516 messages still remain, though, so if you are compiling with gcc you
517 will see some warnings about dubious practices. The warnings are
522 Several new tests have been added, especially for the F<lib> subsection.
524 The tests are now reported in a different order than in earlier Perls.
525 (This happens because the test scripts from under t/lib have been moved
526 to be closer to the library/extension they are testing.)
528 =head1 Known Problems
530 Note that unlike other sections in this document (which describe
531 changes since 5.7.0) this section is cumulative containing known
532 problems for all the 5.7 releases.
540 If Perl is configured to use long doubles the op/int subtests 13 and
541 14 and the ext/POSIX subtest 14 may fail.
545 If Perl is configured to use threads the op/magic subtest 28 may fail.
549 vac 5.0.0.0 May Produce Buggy Code For Perl
551 The AIX C compiler vac version 5.0.0.0 may produce buggy code,
552 resulting in few random tests failing, but when the failing tests
553 are run by hand, they succeed. We suggest upgrading to at least
554 vac version 5.0.1.0, that has been known to compile Perl correctly.
555 "lslpp -L|grep vac.C" will tell you the vac version.
559 =head2 Amiga Perl Invoking Mystery
561 One cannot call Perl using the C<volume:> syntax, that is, C<perl -v>
562 works, but for example C<bin:perl -v> doesn't. The exact reason is
563 known but the current suspect is the F<ixemul> library.
565 =head2 lib/ftmp-security tests warn 'system possibly insecure'
567 Don't panic. Read INSTALL 'make test' section instead.
569 =head2 Cygwin intermittent failures of lib/Memoize/t/expire_file 11 and 12
571 The subtests 11 and 12 sometimes fail and sometimes work.
573 =head2 HP-UX lib/io_multihomed Fails When LP64-Configur
575 The lib/io_multihomed test may hang in HP-UX if Perl has been
576 configured to be 64-bit. Because other 64-bit platforms do not hang in
577 this test, HP-UX is suspect. All other tests pass in 64-bit HP-UX. The
578 test attempts to create and connect to "multihomed" sockets (sockets
579 which have multiple IP addresses).
581 =head2 HP-UX lib/posix Subtest 9 Fails When LP64-Configured
583 If perl is configured with -Duse64bitall, the successful result of the
584 subtest 10 of lib/posix may arrive before the successful result of the
585 subtest 9, which confuses the test harness so much that it thinks the
588 =head2 Linux With Sfio Fails op/misc Test 48
594 OS/390 has rather many test failures but the situation is actually
595 better than it was in 5.6.0, it's just that so many new modules and
596 tests have been added.
598 Failed Test Stat Wstat Total Fail Failed List of Failed
599 -----------------------------------------------------------------------------
600 ../ext/B/Deparse.t 14 1 7.14% 14
601 ../ext/B/Showlex.t 1 1 100.00% 1
602 ../ext/Encode/Encode/Tcl.t 610 13 2.13% 592 594 596 598
604 ../ext/IO/lib/IO/t/io_unix.t 113 28928 5 3 60.00% 3-5
605 ../ext/POSIX/POSIX.t 29 1 3.45% 14
606 ../ext/Storable/t/lock.t 255 65280 5 3 60.00% 3-5
607 ../lib/locale.t 129 33024 117 19 16.24% 99-117
608 ../lib/warnings.t 434 1 0.23% 75
609 ../lib/ExtUtils.t 27 1 3.70% 25
610 ../lib/Math/BigInt/t/bigintpm.t 1190 1 0.08% 1145
611 ../lib/Unicode/UCD.t 81 48 59.26% 1-16 49-64 66-81
612 ../lib/User/pwent.t 9 1 11.11% 4
613 op/pat.t 660 6 0.91% 242-243 424-425
615 op/split.t 0 9 ?? ?? % ??
616 op/taint.t 174 3 1.72% 156 162 168
617 op/tr.t 70 3 4.29% 50 58-59
618 Failed 16/422 test scripts, 96.21% okay. 105/23251 subtests failed, 99.55% okay.
620 =head2 op/sprintf tests 129 and 130
622 The op/sprintf tests 129 and 130 are known to fail on some platforms.
623 Examples include any platform using sfio, and Compaq/Tandem's NonStop-UX.
624 The failing platforms do not comply with the ANSI C Standard, line
625 19ff on page 134 of ANSI X3.159 1989 to be exact. (They produce
626 something else than "1" and "-1" when formatting 0.6 and -0.6 using
627 the printf format "%.0f", most often they produce "0" and "-0".)
629 =head2 Failure of Thread tests
631 B<Note that support for 5.005-style threading remains experimental.>
633 The following tests are known to fail due to fundamental problems in
634 the 5.005 threading implementation. These are not new failures--Perl
635 5.005_0x has the same bugs, but didn't have these tests.
638 t/lib/thr5005.t 19-20
646 ext/POSIX/sigaction subtests 6 and 13 may fail.
650 lib/ExtUtils may spuriously claim that subtest 28 failed,
651 which is interesting since the test only has 27 tests.
655 Numerous numerical test failures
657 op/numconvert 209,210,217,218
659 ext/Time/HiRes/HiRes 9
660 lib/Math/BigInt/t/bigintpm 1145
663 These tests fail because of yet unresolved floating point inaccuracies.
669 Many floating point inaccuracies:
671 op/numconvert 511,657,658,659,665-667,831,991,1151
672 op/pack 10,22,149,156
673 op/sprintf 8,10,13,102-107,134-135,146-153,159-162
674 lib/Math/BigInt/bigintpm 1145,1183
675 lib/Math/Complex 250,257,514,521,722-724,
676 934,935,945,949,955,956,975,976
681 Rather many tests are failing in VMS but that actually more tests
682 succeed in VMS than they used to, it's just that there are many,
683 many more tests than there used to be.
685 Here are the known failures from some compiler/platform combinations.
687 DEC C V5.3-006 on OpenVMS VAX V6.2
689 [-.ext.list.util.t]tainted..............FAILED on test 3
690 [-.ext.posix]sigaction..................FAILED on test 7
691 [-.ext.time.hires]hires.................FAILED on test 14
692 [-.lib.file.find]taint..................FAILED on test 17
693 [-.lib.math.bigint.t]bigintpm...........FAILED on test 1183
694 [-.lib.test.simple.t]exit...............FAILED on test 1
695 [.lib]vmsish............................FAILED on test 13
696 [.op]sprintf............................FAILED on test 12
697 Failed 8/399 tests, 91.23% okay.
699 DEC C V6.0-001 on OpenVMS Alpha V7.2-1
701 [-.ext.list.util.t]tainted..............FAILED on test 3
702 [-.lib.file.find]taint..................FAILED on test 17
703 [-.lib.test.simple.t]exit...............FAILED on test 1
704 [.lib]vmsish............................FAILED on test 13
705 Failed 4/399 tests, 92.48% okay.
707 Compac C V6.4-005 on OpenVMS Alpha 7.2.1
709 [-.ext.b]showlex........................FAILED on test 1
710 [-.ext.list.util.t]tainted..............FAILED on test 3
711 [-.lib.file.find]taint..................FAILED on test 17
712 [-.lib.test.simple.t]exit...............FAILED on test 1
713 [.lib]vmsish............................FAILED on test 13
714 [.op]misc...............................FAILED on test 49
715 Failed 6/401 tests, 92.77% okay.
717 =head2 Localising a Tied Variable Leaks Memory
720 tie my %tie_hash => 'Tie::StdHash';
724 local($tie_hash{Foo}) = 1; # leaks
726 Code like the above is known to leak memory every time the local()
729 =head2 Self-tying of Arrays and Hashes Is Forbidden
731 Self-tying of arrays and hashes is broken in rather deep and
732 hard-to-fix ways. As a stop-gap measure to avoid people from getting
733 frustrated at the mysterious results (core dumps, most often) it is
734 for now forbidden (you will get a fatal error even from an attempt).
736 =head2 Variable Attributes are not Currently Usable for Tieing
738 This limitation will hopefully be fixed in future. (Subroutine
739 attributes work fine for tieing, see L<Attribute::Handlers>).
741 =head2 Building Extensions Can Fail Because Of Largefiles
743 Some extensions like mod_perl are known to have issues with
744 `largefiles', a change brought by Perl 5.6.0 in which file offsets
745 default to 64 bits wide, where supported. Modules may fail to compile
746 at all or compile and work incorrectly. Currently there is no good
747 solution for the problem, but Configure now provides appropriate
748 non-largefile ccflags, ldflags, libswanted, and libs in the %Config
749 hash (e.g., $Config{ccflags_nolargefiles}) so the extensions that are
750 having problems can try configuring themselves without the
751 largefileness. This is admittedly not a clean solution, and the
752 solution may not even work at all. One potential failure is whether
753 one can (or, if one can, whether it's a good idea) link together at
754 all binaries with different ideas about file offsets, all this is
757 =head2 The Compiler Suite Is Still Experimental
759 The compiler suite is slowly getting better but is nowhere near
762 =head2 The Long Double Support is Still Experimental
764 The ability to configure Perl's numbers to use "long doubles",
765 floating point numbers of hopefully better accuracy, is still
766 experimental. The implementations of long doubles are not yet
767 widespread and the existing implementations are not quite mature
768 or standardised, therefore trying to support them is a rare
769 and moving target. The gain of more precision may also be offset
770 by slowdown in computations (more bits to move around, and the
771 operations are more likely to be executed by less optimised
774 =head1 Reporting Bugs
776 If you find what you think is a bug, you might check the articles
777 recently posted to the comp.lang.perl.misc newsgroup and the perl
778 bug database at http://bugs.perl.org. There may also be
779 information at http://www.perl.com/perl/, the Perl Home Page.
781 If you believe you have an unreported bug, please run the B<perlbug>
782 program included with your release. Be sure to trim your bug down
783 to a tiny but sufficient test case. Your bug report, along with the
784 output of C<perl -V>, will be sent off to perlbug@perl.org to be
785 analysed by the Perl porting team.
789 The F<Changes> file for exhaustive details on what changed.
791 The F<INSTALL> file for how to build Perl.
793 The F<README> file for general stuff.
795 The F<Artistic> and F<Copying> files for copyright information.
799 Written by Jarkko Hietaniemi <F<jhi@iki.fi>>, with many contributions
800 from The Perl Porters and Perl Users submitting feedback and patches.
802 Send omissions or corrections to <F<perlbug@perl.org>>.