3 perldelta - what is new for perl v5.8.0
7 This document describes differences between the 5.6.0 release and
10 Many of the bug fixes in 5.8.0 were already seen in the 5.6.1
11 maintenance release since the two releases were kept closely
12 coordinated (while 5.8.0 was still called 5.7.something).
14 Changes that were integrated into the 5.6.1 release are marked C<[561]>.
15 Many of these changes have been further developed since 5.6.1 was released,
16 those are marked C<[561+]>.
18 You can see the list of changes in the 5.6.1 release (both from the
19 5.005_03 release and the 5.6.0 release) by reading L<perl561delta>.
21 =head1 Highlights In 5.8.0
27 Better Unicode support
31 New Thread Implementation
39 Better Numeric Accuracy
47 More Extensive Regression Testing
51 =head1 Incompatible Changes
53 =head2 Binary Incompatibility
55 B<Perl 5.8 is not binary compatible with earlier releases of Perl.>
57 B<You have to recompile your XS modules.>
59 (Pure Perl modules should continue to work.)
61 The major reason for the discontinuity is the new IO architecture
62 called PerlIO. PerlIO is the default configuration because without
63 it many new features of Perl 5.8 cannot be used. In other words:
64 you just have to recompile your modules containing XS code, sorry
67 In future releases of Perl, non-PerlIO aware XS modules may become
68 completely unsupported. This shouldn't be too difficult for module
69 authors, however: PerlIO has been designed as a drop-in replacement
70 (at the source code level) for the stdio interface.
72 Depending on your platform, there are also other reasons why
73 we decided to break binary compatibility, please read on.
75 =head2 64-bit platforms and malloc
77 If your pointers are 64 bits wide, the Perl malloc is no longer being
78 used because it does not work well with 8-byte pointers. Also,
79 usually the system mallocs on such platforms are much better optimized
80 for such large memory models than the Perl malloc. Some memory-hungry
81 Perl applications like the PDL don't work well with Perl's malloc.
82 Finally, other applications than Perl (such as mod_perl) tend to prefer
83 the system malloc. Such platforms include Alpha and 64-bit HPPA,
86 =head2 AIX Dynaloading
88 The AIX dynaloading now uses in AIX releases 4.3 and newer the native
89 dlopen interface of AIX instead of the old emulated interface. This
90 change will probably break backward compatibility with compiled
91 modules. The change was made to make Perl more compliant with other
92 applications like mod_perl which are using the AIX native interface.
94 =head2 Attributes for C<my> variables now handled at run-time.
96 The C<my EXPR : ATTRS> syntax now applies variable attributes at
97 run-time. (Subroutine and C<our> variables still get attributes applied
98 at compile-time.) See L<attributes> for additional details. In particular,
99 however, this allows variable attributes to be useful for C<tie> interfaces,
100 which was a deficiency of earlier releases. Note that the new semantics
101 doesn't work with the Attribute::Handlers module (as of version 0.76).
103 =head2 Socket Extension Dynamic in VMS
105 The Socket extension is now dynamically loaded instead of being
106 statically built in. This may or may not be a problem with ancient
107 TCP/IP stacks of VMS: we do not know since we weren't able to test
108 Perl in such configurations.
110 =head2 IEEE-format Floating Point Default on OpenVMS Alpha
112 Perl now uses IEEE format (T_FLOAT) as the default internal floating
113 point format on OpenVMS Alpha, potentially breaking binary compatibility
114 with external libraries or existing data. G_FLOAT is still available as
115 a configuration option. The default on VAX (D_FLOAT) has not changed.
117 =head2 New Unicode Properties
119 Unicode I<scripts> are now supported. Scripts are similar to (and superior
120 to) Unicode I<blocks>. The difference between scripts and blocks is that
121 scripts are the glyphs used by a language or a group of languages, while
122 the blocks are more artificial groupings of (mostly) 256 characters based
123 on the Unicode numbering.
125 In general, scripts are more inclusive, but not universally so. For
126 example, while the script C<Latin> includes all the Latin characters and
127 their various diacritic-adorned versions, it does not include the various
128 punctuation or digits (since they are not solely C<Latin>).
130 A number of other properties are now supported, including C<\p{L&}>,
131 C<\p{Any}> C<\p{Assigned}>, C<\p{Unassigned}>, C<\p{Blank}> [561] and
132 C<\p{SpacePerl}> [561] (along with their C<\P{...}> versions, of course).
133 See L<perlunicode> for details, and more additions.
135 The C<In> or C<Is> prefix to names used with the C<\p{...}> and C<\P{...}>
136 are now almost always optional. The only exception is that a C<In> prefix
137 is required to signify a Unicode block when a block name conflicts with a
138 script name. For example, C<\p{Tibetan}> refers to the script, while
139 C<\p{InTibetan}> refers to the block. When there is no name conflict, you
140 can omit the C<In> from the block name (e.g. C<\p{BraillePatterns}>), but
141 to be safe, it's probably best to always use the C<In>).
143 =head2 REF(...) Instead Of SCALAR(...)
145 A reference to a reference now stringifies as "REF(0x81485ec)" instead
146 of "SCALAR(0x81485ec)" in order to be more consistent with the return
149 =head2 pack/unpack D/F recycled
151 The undocumented pack/unpack template letters D/F have been recycled
152 for better use: now they stand for long double (if supported by the
153 platform) and NV (Perl internal floating point type). (They used
154 to be aliases for d/f, but you never knew that.)
162 The semantics of bless(REF, REF) were unclear and until someone proves
163 it to make some sense, it is forbidden.
167 The obsolete chat2 library that should never have been allowed
168 to escape the laboratory has been decommissioned.
172 The builtin dump() function has probably outlived most of its
173 usefulness. The core-dumping functionality will remain in future
174 available as an explicit call to C<CORE::dump()>, but in future
175 releases the behaviour of an unqualified C<dump()> call may change.
179 The very dusty examples in the eg/ directory have been removed.
180 Suggestions for new shiny examples welcome but the main issue is that
181 the examples need to be documented, tested and (most importantly)
186 The (bogus) escape sequences \8 and \9 now give an optional warning
187 ("Unrecognized escape passed through"). There is no need to \-escape
192 The list of filenames from glob() (or <...>) is now by default sorted
193 alphabetically to be csh-compliant (which is what happened before
194 in most UNIX platforms). (bsd_glob() does still sort platform
195 natively, ASCII or EBCDIC, unless GLOB_ALPHASORT is specified.) [561]
199 Spurious syntax errors generated in certain situations, when glob()
200 caused File::Glob to be loaded for the first time, have been fixed. [561]
204 Although "you shouldn't do that", it was possible to write code that
205 depends on Perl's hashed key order (Data::Dumper does this). The new
206 algorithm "One-at-a-Time" produces a different hashed key order.
207 More details are in L</"Performance Enhancements">.
211 lstat(FILEHANDLE) now gives a warning because the operation makes no sense.
212 In future releases this may become a fatal error.
216 The C<package;> syntax (C<package> without an argument) has been
217 deprecated. Its semantics were never that clear and its
218 implementation even less so. If you have used that feature to
219 disallow all but fully qualified variables, C<use strict;> instead.
223 The unimplemented POSIX regex features [[.cc.]] and [[=c=]] are still
224 recognised but now cause fatal errors. The previous behaviour of
225 ignoring them by default and warning if requested was unacceptable
226 since it, in a way, falsely promised that the features could be used.
230 In future releases, non-PerlIO aware XS modules may become completely
231 unsupported. Since PerlIO is a drop-in replacement for stdio at the
232 source code level, this shouldn't be that drastic a change.
236 The current user-visible implementation of pseudo-hashes (the weird
237 use of the first array element) is deprecated starting from Perl 5.8.0
238 and will be removed in Perl 5.10.0, and the feature will be
239 implemented differently. Not only is the current interface rather
240 ugly, but the current implementation slows down normal array and hash
241 use quite noticeably. The C<fields> pragma interface will remain
242 available. The I<restricted hashes> interface is expected to
243 be the replacement interface (see L<Hash::Util>).
247 The syntaxes C<< @a->[...] >> and C<< %h->{...} >> have now been deprecated.
251 After years of trying, suidperl is considered to be too complex to
252 ever be considered truly secure. The suidperl functionality is likely
253 to be removed in a future release.
257 The 5.005 threads model (module C<Thread>) is deprecated and expected
258 to be removed in Perl 5.10. Multithreaded code should be migrated to
259 the new ithreads model (see L<threads>, L<threads::shared> and
264 The long deprecated uppercase aliases for the string comparison
265 operators (EQ, NE, LT, LE, GE, GT) have now been removed.
269 The tr///C and tr///U features have been removed and will not return;
270 the interface was a mistake. Sorry about that. For similar
271 functionality, see pack('U0', ...) and pack('C0', ...). [561]
275 Earlier Perls treated "sub foo (@bar)" as equivalent to "sub foo (@)".
276 The prototypes are now checked better at compile-time for invalid
277 syntax. An optional warning is generated ("Illegal character in
278 prototype...") but this may be upgraded to a fatal error in a future
283 The C<exec LIST> and C<system LIST> operations will produce fatal
284 errors on tainted data in some future release.
288 The existing behaviour when localising tied arrays and hashes is wrong,
289 and will be changed in a future release, so do not rely on the existing
290 behaviour. See L<"Localising Tied Arrays and Hashes Is Broken">.
294 =head1 Core Enhancements
296 =head2 PerlIO is Now The Default
302 IO is now by default done via PerlIO rather than system's "stdio".
303 PerlIO allows "layers" to be "pushed" onto a file handle to alter the
304 handle's behaviour. Layers can be specified at open time via 3-arg
307 open($fh,'>:crlf :utf8', $path) || ...
309 or on already opened handles via extended C<binmode>:
311 binmode($fh,':encoding(iso-8859-7)');
313 The built-in layers are: unix (low level read/write), stdio (as in
314 previous Perls), perlio (re-implementation of stdio buffering in a
315 portable manner), crlf (does CRLF <=> "\n" translation as on Win32,
316 but available on any platform). A mmap layer may be available if
317 platform supports it (mostly UNIXes).
319 Layers to be applied by default may be specified via the 'open' pragma.
321 See L</"Installation and Configuration Improvements"> for the effects
322 of PerlIO on your architecture name.
326 File handles can be marked as accepting Perl's internal encoding of Unicode
327 (UTF-8 or UTF-EBCDIC depending on platform) by a pseudo layer ":utf8" :
329 open($fh,">:utf8","Uni.txt");
331 Note for EBCDIC users: the pseudo layer ":utf8" is erroneously named
332 for you since it's not UTF-8 what you will be getting but instead
333 UTF-EBCDIC. See L<perlunicode>, L<utf8>, and
334 http://www.unicode.org/unicode/reports/tr16/ for more information.
335 In future releases this naming may change. See L<perluniintro>
336 for more information about UTF-8.
340 If your environment variables (LC_ALL, LC_CTYPE, LANG, LANGUAGE) look
341 like you want to use UTF-8 (any of the the variables match C</utf-?8/i>),
342 your STDIN, STDOUT, STDERR handles and the default open discipline
343 (see L<open>) are marked as UTF-8.
345 Note that after this Perl really does assume that everything is UTF-8:
346 for example if some input handle is not, Perl will probably very soon
347 complain about the input data like this "Malformed UTF-8 ..." since
348 any old eight-bit data is not legal UTF-8.
350 Note for code authors: if you want to enable your users to use UTF-8
351 as their default encoding but in your code still have eight-bit I/O streams
352 (such as images or zip files), you need to explicitly open() or binmode()
353 with C<:bytes> (see L<perlfunc/open> and L<perlfunc/binmode>), or you
354 can just use C<binmode(FH)> (nice for pre-5.8.0 backward compatibility).
358 File handles can translate character encodings from/to Perl's internal
359 Unicode form on read/write via the ":encoding()" layer.
363 File handles can be opened to "in memory" files held in Perl scalars via:
365 open($fh,'>', \$variable) || ...
369 Anonymous temporary files are available without need to
370 'use FileHandle' or other module via
372 open($fh,"+>", undef) || ...
374 That is a literal undef, not an undefined value.
378 The list form of C<open> is now implemented for pipes (at least on UNIX):
380 open($fh,"-|", 'cat', '/etc/motd')
382 creates a pipe, and runs the equivalent of exec('cat', '/etc/motd') in
387 If your locale environment variables (LANGUAGE, LC_ALL, LC_CTYPE, LANG)
388 contain the strings 'UTF-8' or 'UTF8' (case-insensitive matching),
389 the default encoding of your STDIN, STDOUT, and STDERR, and of
390 B<any subsequent file open>, is UTF-8.
394 =head2 Restricted Hashes
396 A restricted hash is restricted to a certain set of keys, no keys
397 outside the set can be added. Also individual keys can be restricted
398 so that the key cannot be deleted and the value cannot be changed.
399 No new syntax is involved: the Hash::Util module is the interface.
403 Perl used to be fragile in that signals arriving at inopportune moments
404 could corrupt Perl's internal state. Now Perl postpones handling of
405 signals until it's safe (between opcodes).
407 This change may have surprising side effects because signals no longer
408 interrupt Perl instantly. Perl will now first finish whatever it was
409 doing, like finishing an internal operation (like sort()) or an
410 external operation (like an I/O operation), and only then look at any
411 arrived signals (and before starting the next operation). No more corrupt
412 internal state since the current operation is always finished first,
413 but the signal may take more time to get heard. Note that breaking
414 out from potentially blocking operations should still work, though.
416 =head2 Unicode Overhaul
418 Unicode in general should be now much more usable than in Perl 5.6.0
419 (or even in 5.6.1). Unicode can be used in hash keys, Unicode in
420 regular expressions should work now, Unicode in tr/// should work now,
421 Unicode in I/O should work now. See L<perluniintro> for introduction
422 and L<perlunicode> for details.
428 The Unicode Character Database coming with Perl has been upgraded
429 to Unicode 3.2.0. For more information, see http://www.unicode.org/ .
430 [561+] (5.6.1 has UCD 3.0.1.)
434 For developers interested in enhancing Perl's Unicode capabilities:
435 almost all the UCD files are included with the Perl distribution in
436 the F<lib/unicore> subdirectory. The most notable omission, for space
437 considerations, is the Unihan database.
441 The properties \p{Blank} and \p{SpacePerl} have been added. "Blank" is like
442 C isblank(), that is, it contains only "horizontal whitespace" (the space
443 character is, the newline isn't), and the "SpacePerl" is the Unicode
444 equivalent of C<\s> (\p{Space} isn't, since that includes the vertical
445 tabulator character, whereas C<\s> doesn't.)
447 See "New Unicode Properties" earlier in this document for additional
448 information on changes with Unicode properties.
452 =head2 Understanding of Numbers
454 In general a lot of fixing has happened in the area of Perl's
455 understanding of numbers, both integer and floating point. Since in
456 many systems the standard number parsing functions like C<strtoul()>
457 and C<atof()> seem to have bugs, Perl tries to work around their
458 deficiencies. This results hopefully in more accurate numbers.
460 Perl now tries internally to use integer values in numeric conversions
461 and basic arithmetics (+ - * /) if the arguments are integers, and
462 tries also to keep the results stored internally as integers.
463 This change leads to often slightly faster and always less lossy
464 arithmetics. (Previously Perl always preferred floating point numbers
467 =head2 Arrays now always interpolate into double-quoted strings [561]
469 In double-quoted strings, arrays now interpolate, no matter what. The
470 behavior in earlier versions of perl 5 was that arrays would interpolate
471 into strings if the array had been mentioned before the string was
472 compiled, and otherwise Perl would raise a fatal compile-time error.
473 In versions 5.000 through 5.003, the error was
475 Literal @example now requires backslash
477 In versions 5.004_01 through 5.6.0, the error was
479 In string, @example now must be written as \@example
481 The idea here was to get people into the habit of writing
482 C<"fred\@example.com"> when they wanted a literal C<@> sign, just as
483 they have always written C<"Give me back my \$5"> when they wanted a
486 Starting with 5.6.1, when Perl now sees an C<@> sign in a
487 double-quoted string, it I<always> attempts to interpolate an array,
488 regardless of whether or not the array has been used or declared
489 already. The fatal error has been downgraded to an optional warning:
491 Possible unintended interpolation of @example in string
493 This warns you that C<"fred@example.com"> is going to turn into
494 C<fred.com> if you don't backslash the C<@>.
495 See http://www.plover.com/~mjd/perl/at-error.html for more details
496 about the history here.
498 =head2 Miscellaneous Changes
504 AUTOLOAD is now lvaluable, meaning that you can add the :lvalue attribute
505 to AUTOLOAD subroutines and you can assign to the AUTOLOAD return value.
509 The $Config{byteorder} (and corresponding BYTEORDER in config.h) was
510 previously wrong in platforms if sizeof(long) was 4, but sizeof(IV)
511 was 8. The byteorder was only sizeof(long) bytes long (1234 or 4321),
512 but now it is correctly sizeof(IV) bytes long, (12345678 or 87654321).
513 (This problem didn't affect Windows platforms.)
515 Also, $Config{byteorder} is now computed dynamically--this is more
516 robust with "fat binaries" where an executable image contains binaries
517 for more than one binary platform, and when cross-compiling.
521 C<perl -d:Module=arg,arg,arg> now works (previously one couldn't pass
522 in multiple arguments.)
526 C<do> followed by a bareword now ensures that this bareword isn't
527 a keyword (to avoid a bug where C<do q(foo.pl)> tried to call a
528 subroutine called C<q>). This means that for example instead of
529 C<do format()> you must write C<do &format()>.
533 The builtin dump() now gives an optional warning
534 C<dump() better written as CORE::dump()>,
535 meaning that by default C<dump(...)> is resolved as the builtin
536 dump() which dumps core and aborts, not as (possibly) user-defined
537 C<sub dump>. To call the latter, qualify the call as C<&dump(...)>.
538 (The whole dump() feature is to considered deprecated, and possibly
539 removed/changed in future releases.)
543 chomp() and chop() are now overridable. Note, however, that their
544 prototype (as given by C<prototype("CORE::chomp")> is undefined,
545 because it cannot be expressed and therefore one cannot really write
546 replacements to override these builtins.
550 END blocks are now run even if you exit/die in a BEGIN block.
551 Internally, the execution of END blocks is now controlled by
552 PL_exit_flags & PERL_EXIT_DESTRUCT_END. This enables the new
553 behaviour for Perl embedders. This will default in 5.10. See
558 Formats now support zero-padded decimal fields.
562 Lvalue subroutines can now return C<undef> in list context. However,
563 the lvalue subroutine feature still remains experimental. [561+]
567 A lost warning "Can't declare ... dereference in my" has been
568 restored (Perl had it earlier but it became lost in later releases.)
572 A new special regular expression variable has been introduced:
573 C<$^N>, which contains the most-recently closed group (submatch).
577 C<no Module;> does not produce an error even if Module does not have an
578 unimport() method. This parallels the behavior of C<use> vis-a-vis
583 The numerical comparison operators return C<undef> if either operand
584 is a NaN. Previously the behaviour was unspecified.
588 C<our> can now have an experimental optional attribute C<unique> that
589 affects how global variables are shared among multiple interpreters,
594 The following builtin functions are now overridable: each(), keys(),
595 pop(), push(), shift(), splice(), unshift(). [561]
599 C<pack() / unpack()> can now group template letters with C<()> and then
600 apply repetition/count modifiers on the groups.
604 C<pack() / unpack()> can now process the Perl internal numeric types:
605 IVs, UVs, NVs-- and also long doubles, if supported by the platform.
606 The template letters are C<j>, C<J>, C<F>, and C<D>.
610 C<pack('U0a*', ...)> can now be used to force a string to UTF8.
614 my __PACKAGE__ $obj now works. [561]
618 POSIX::sleep() now returns the number of I<unslept> seconds
619 (as the POSIX standard says), as opposed to CORE::sleep() which
620 returns the number of slept seconds.
624 The printf() and sprintf() now support parameter reordering using the
625 C<%\d+\$> and C<*\d+\$> syntaxes. For example
627 print "%2\$s %1\$s\n", "foo", "bar";
629 will print "bar foo\n". This feature helps in writing
630 internationalised software, and in general when the order
631 of the parameters can vary.
635 The (\&) prototype now works properly. [561]
639 prototype(\[$@%&]) is now available to implicitly create references
640 (useful for example if you want to emulate the tie() interface).
644 A new command-line option, C<-t> is available. It is the
645 little brother of C<-T>: instead of dying on taint violations,
646 lexical warnings are given. B<This is only meant as a temporary
647 debugging aid while securing the code of old legacy applications.
648 This is not a substitute for -T.>
652 In other taint news, the C<exec LIST> and C<system LIST> have now been
653 considered too risky (think C<exec @ARGV>: it can start any program
654 with any arguments), and now the said forms cause a warning under
655 lexical warnings. You should carefully launder the arguments to
656 guarantee their validity. In future releases of Perl the forms will
657 become fatal errors so consider starting laundering now.
661 Tied hash interfaces are now required to have the EXISTS and DELETE
662 methods (either own or inherited).
666 If tr/// is just counting characters, it doesn't attempt to
671 untie() will now call an UNTIE() hook if it exists. See L<perltie>
676 L<utime> now supports C<utime undef, undef, @files> to change the
677 file timestamps to the current time.
681 The rules for allowing underscores (underbars) in numeric constants
682 have been relaxed and simplified: now you can have an underscore
683 simply B<between digits>.
687 Rather than relying on C's argv[0] (which may not contain a full pathname)
688 where possible $^X is now set by asking the operating system.
689 (eg by reading F</proc/self/exe> on Linux, F</proc/curproc/file> on FreeBSD)
693 A new variable, C<${^TAINT}>, indicates whether taint mode is enabled.
697 You can now override the readline() builtin, and this overrides also
698 the <FILEHANDLE> angle bracket operator.
702 The command-line options -s and -F are now recognized on the shebang
707 Use of the C</c> match modifier without an accompanying C</g> modifier
708 elicits a new warning: C<Use of /c modifier is meaningless without /g>.
710 Use of C</c> in substitutions, even with C</g>, elicits
711 C<Use of /c modifier is meaningless in s///>.
713 Use of C</g> with C<split> elicits C<Use of /g modifier is meaningless
718 Support for the C<CLONE> special subroutine had been added.
719 With ithreads, when a new thread is created, all Perl data is cloned,
720 however non-Perl data cannot be cloned automatically. In C<CLONE> you
721 can do whatever you need to do, like for example handle the cloning of
722 non-Perl data, if necessary. C<CLONE> will be executed once for every
723 package that has it defined or inherited. It will be called in the
724 context of the new thread, so all modifications are made in the new area.
730 =head1 Modules and Pragmata
732 =head2 New Modules and Pragmata
738 C<Attribute::Handlers> allows a class to define attribute handlers.
741 use Attribute::Handlers;
742 sub Wolf :ATTR(SCALAR) { print "howl!\n" }
744 # later, in some package using or inheriting from MyPack...
746 my MyPack $Fluffy : Wolf; # the attribute handler Wolf will be called
748 Both variables and routines can have attribute handlers. Handlers can
749 be specific to type (SCALAR, ARRAY, HASH, or CODE), or specific to the
750 exact compilation phase (BEGIN, CHECK, INIT, or END).
751 See L<Attribute::Handlers>.
755 C<B::Concise>, by Stephen McCamant, is a new compiler backend for
756 walking the Perl syntax tree, printing concise info about ops.
757 The output is highly customisable. See L<B::Concise>. [561+]
761 The new bignum, bigint, and bigrat pragmas, by Tels, implement
762 transparent bignum support (using the Math::BigInt, Math::BigFloat,
763 and Math::BigRat backends).
767 C<Class::ISA>, by Sean Burke, is a module for reporting the search
768 path for a class's ISA tree. See L<Class::ISA>.
772 C<Cwd> now has a split personality: if possible, an XS extension is
773 used, (this will hopefully be faster, more secure, and more robust)
774 but if not possible, the familiar Perl implementation is used.
778 C<Devel::PPPort>, originally by Kenneth Albanowski and now
779 maintained by Paul Marquess, has been added. It is primarily used
780 by C<h2xs> to enhance portability of XS modules between different
781 versions of Perl. See L<Devel::PPPort>.
785 C<Digest>, frontend module for calculating digests (checksums), from
786 Gisle Aas, has been added. See L<Digest>.
790 C<Digest::MD5> for calculating MD5 digests (checksums) as defined in
791 RFC 1321, from Gisle Aas, has been added. See L<Digest::MD5>.
793 use Digest::MD5 'md5_hex';
795 $digest = md5_hex("Thirsty Camel");
797 print $digest, "\n"; # 01d19d9d2045e005c3f1b80e8b164de1
799 NOTE: the C<MD5> backward compatibility module is deliberately not
800 included since its further use is discouraged.
804 C<Encode>, originally by Nick Ing-Simmons and now maintained by Dan
805 Kogai, provides a mechanism to translate between different character
806 encodings. Support for Unicode, ISO-8859-1, and ASCII are compiled in
807 to the module. Several other encodings (like the rest of the
808 ISO-8859, CP*/Win*, Mac, KOI8-R, three variants EBCDIC, Chinese,
809 Japanese, and Korean encodings) are included and can be loaded at
810 runtime. (For space considerations, the largest Chinese encodings
811 have been separated into their own CPAN module, Encode::HanExtra,
812 which Encode will use if available). See L<Encode>.
814 Any encoding supported by Encode module is also available to the
815 ":encoding()" layer if PerlIO is used.
819 C<Hash::Util> is the interface to the new I<restricted hashes>
820 feature. (Implemented by Jeffrey Friedl, Nick Ing-Simmons, and
821 Michael Schwern.) See L<Hash::Util>.
825 C<I18N::Langinfo> can be used to query locale information.
826 See L<I18N::Langinfo>.
830 C<I18N::LangTags>, by Sean Burke, has functions for dealing with
831 RFC3066-style language tags. See L<I18N::LangTags>.
835 C<ExtUtils::Constant>, by Nicholas Clark, is a new tool for extension
836 writers for generating XS code to import C header constants.
837 See L<ExtUtils::Constant>.
841 C<Filter::Simple>, by Damian Conway, is an easy-to-use frontend to
842 Filter::Util::Call. See L<Filter::Simple>.
848 use Filter::Simple sub {
849 while (my ($from, $to) = splice @_, 0, 2) {
858 use MyFilter qr/red/ => 'green';
860 print "red\n"; # this code is filtered, will print "green\n"
861 print "bored\n"; # this code is filtered, will print "bogreen\n"
865 print "red\n"; # this code is not filtered, will print "red\n"
869 C<File::Temp>, by Tim Jenness, allows one to create temporary files
870 and directories in an easy, portable, and secure way. See L<File::Temp>.
875 C<Filter::Util::Call>, by Paul Marquess, provides you with the
876 framework to write I<source filters> in Perl. For most uses, the
877 frontend Filter::Simple is to be preferred. See L<Filter::Util::Call>.
881 C<if>, by Ilya Zakharevich, is a new pragma for conditional inclusion
886 L<libnet>, by Graham Barr, is a collection of perl5 modules related
887 to network programming. See L<Net::FTP>, L<Net::NNTP>, L<Net::Ping>
888 (not part of libnet, but related), L<Net::POP3>, L<Net::SMTP>,
891 Perl installation leaves libnet unconfigured; use F<libnetcfg>
896 C<List::Util>, by Graham Barr, is a selection of general-utility
897 list subroutines, such as sum(), min(), first(), and shuffle().
902 C<Locale::Constants>, C<Locale::Country>, C<Locale::Currency>
903 C<Locale::Language>, and L<Locale::Script>, by Neil Bowers, have
904 been added. They provide the codes for various locale standards, such
905 as "fr" for France, "usd" for US Dollar, and "ja" for Japanese.
909 $country = code2country('jp'); # $country gets 'Japan'
910 $code = country2code('Norway'); # $code gets 'no'
912 See L<Locale::Constants>, L<Locale::Country>, L<Locale::Currency>,
913 and L<Locale::Language>.
917 C<Locale::Maketext>, by Sean Burke, is a localization framework. See
918 L<Locale::Maketext>, and L<Locale::Maketext::TPJ13>. The latter is an
919 article about software localization, originally published in The Perl
920 Journal #13, and republished here with kind permission.
924 C<Math::BigRat> for big rational numbers, to accompany Math::BigInt and
925 Math::BigFloat, from Tels. See L<Math::BigRat>.
929 C<Memoize> can make your functions faster by trading space for time,
930 from Mark-Jason Dominus. See L<Memoize>.
934 C<MIME::Base64>, by Gisle Aas, allows you to encode data in base64,
935 as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail
940 $encoded = encode_base64('Aladdin:open sesame');
941 $decoded = decode_base64($encoded);
943 print $encoded, "\n"; # "QWxhZGRpbjpvcGVuIHNlc2FtZQ=="
949 C<MIME::QuotedPrint>, by Gisle Aas, allows you to encode data
950 in quoted-printable encoding, as defined in RFC 2045 - I<MIME
951 (Multipurpose Internet Mail Extensions)>.
953 use MIME::QuotedPrint;
955 $encoded = encode_qp("Smiley in Unicode: \x{263a}");
956 $decoded = decode_qp($encoded);
958 print $encoded, "\n"; # "Smiley in Unicode: =263A"
960 MIME::QuotedPrint has been enhanced to provide the basic methods
961 necessary to use it with PerlIO::Via as in :
963 use MIME::QuotedPrint;
964 open($fh,">Via(MIME::QuotedPrint)",$path);
966 See L<MIME::QuotedPrint>.
970 C<NEXT>, by Damian Conway, is a pseudo-class for method redispatch.
975 C<open> is a new pragma for setting the default I/O disciplines
980 C<PerlIO::Scalar>, by Nick Ing-Simmons, provides the implementation
981 of IO to "in memory" Perl scalars as discussed above. It also serves
982 as an example of a loadable PerlIO layer. Other future possibilities
983 include PerlIO::Array and PerlIO::Code. See L<PerlIO::Scalar>.
987 C<PerlIO::Via>, by Nick Ing-Simmons, acts as a PerlIO layer and wraps
988 PerlIO layer functionality provided by a class (typically implemented
991 use MIME::QuotedPrint;
992 open($fh,">Via(MIME::QuotedPrint)",$path);
994 This will automatically convert everything output to C<$fh>
995 to Quoted-Printable. See L<PerlIO::Via>.
999 C<Pod::ParseLink>, by Russ Allbery, has been added,
1000 to parse LZ<><> links in pods as described in the new
1005 C<Pod::Text::Overstrike>, by Joe Smith, has been added.
1006 It converts POD data to formatted overstrike text.
1007 See L<Pod::Text::Overstrike>. [561+]
1011 C<Scalar::Util> is a selection of general-utility scalar subroutines,
1012 such as blessed(), reftype(), and tainted(). See L<Scalar::Util>.
1016 C<sort> is a new pragma for controlling the behaviour of sort().
1020 C<Storable> gives persistence to Perl data structures by allowing the
1021 storage and retrieval of Perl data to and from files in a fast and
1022 compact binary format. Because in effect Storable does serialisation
1023 of Perl data structues, with it you can also clone deep, hierarchical
1024 datastructures. Storable was originally created by Raphael Manfredi,
1025 but it is now maintained by Abhijit Menon-Sen. Storable has been
1026 enhanced to understand the two new hash features, Unicode keys and
1027 restricted hashes. See L<Storable>.
1031 C<Switch>, by Damian Conway, has been added. Just by saying
1035 you have C<switch> and C<case> available in Perl.
1041 case 1 { print "number 1" }
1042 case "a" { print "string a" }
1043 case [1..10,42] { print "number in list" }
1044 case (@array) { print "number in list" }
1045 case /\w+/ { print "pattern" }
1046 case qr/\w+/ { print "pattern" }
1047 case (%hash) { print "entry in hash" }
1048 case (\%hash) { print "entry in hash" }
1049 case (\&sub) { print "arg to subroutine" }
1050 else { print "previous case not true" }
1057 C<Test::More>, by Michael Schwern, is yet another framework for writing
1058 test scripts, more extensive than Test::Simple. See L<Test::More>.
1062 C<Test::Simple>, by Michael Schwern, has basic utilities for writing
1063 tests. See L<Test::Simple>.
1067 C<Text::Balanced>, by Damian Conway, has been added, for extracting
1068 delimited text sequences from strings.
1070 use Text::Balanced 'extract_delimited';
1072 ($a, $b) = extract_delimited("'never say never', he never said", "'", '');
1074 $a will be "'never say never'", $b will be ', he never said'.
1076 In addition to extract_delimited(), there are also extract_bracketed(),
1077 extract_quotelike(), extract_codeblock(), extract_variable(),
1078 extract_tagged(), extract_multiple(), gen_delimited_pat(), and
1079 gen_extract_tagged(). With these, you can implement rather advanced
1080 parsing algorithms. See L<Text::Balanced>.
1084 C<threads>, by Arthur Bergman, is an interface to interpreter threads.
1085 Interpreter threads (ithreads) is the new thread model introduced in
1086 Perl 5.6 but only available as an internal interface for extension
1087 writers (and for Win32 Perl for C<fork()> emulation). See L<threads>,
1088 L<threads::shared>, and L<perlthrtut>.
1092 C<threads::shared>, by Arthur Bergman, allows data sharing for
1093 interpreter threads. In the ithreads model any data sharing between
1094 threads must be explicit, as opposed to the old 5.005 thread model
1095 where data sharing was implicit. See L<threads::shared>.
1099 C<Tie::File>, by Mark-Jason Dominus, associates a Perl array with the
1100 lines of a file. See L<Tie::File>.
1104 C<Tie::Memoize>, by Ilya Zakharevich, provides on-demand loaded hashes.
1105 See L<Tie::Memoize>.
1109 C<Tie::RefHash::Nestable>, by Edward Avis, allows storing hash
1110 references (unlike the standard Tie::RefHash) The module is contained
1111 within Tie::RefHash. See L<Tie::RefHash>.
1115 C<Time::HiRes>, by Douglas E. Wegscheid, provides high resolution
1116 timing (ualarm, usleep, and gettimeofday). See L<Time::HiRes>.
1120 C<Unicode::UCD> offers a querying interface to the Unicode Character
1121 Database. See L<Unicode::UCD>.
1125 C<Unicode::Collate>, by SADAHIRO Tomoyuki, implements the UCA
1126 (Unicode Collation Algorithm) for sorting Unicode strings.
1127 See L<Unicode::Collate>.
1131 C<Unicode::Normalize>, by SADAHIRO Tomoyuki, implements the various
1132 Unicode normalization forms. See L<Unicode::Normalize>.
1136 C<XS::Typemap>, by Tim Jenness, is a test extension that exercises
1137 XS typemaps. Nothing gets installed, but the code is worth studying
1138 for extension writers.
1142 =head2 Updated And Improved Modules and Pragmata
1148 The following independently supported modules have been updated to the
1149 newest versions from CPAN: CGI, CPAN, DB_File, File::Spec, File::Temp,
1150 Getopt::Long, Math::BigFloat, Math::BigInt, the podlators bundle
1151 (Pod::Man, Pod::Text), Pod::LaTeX [561+], Pod::Parser, Storable,
1152 Term::ANSIColor, Test, Text-Tabs+Wrap.
1156 attributes::reftype() now works on tied arguments.
1160 AutoLoader can now be disabled with C<no AutoLoader;>.
1164 B::Deparse has been significantly enhanced by Robin Houston. It can
1165 now deparse almost all of the standard test suite (so that the tests
1166 still succeed). There is a make target "test.deparse" for trying this
1171 Carp now has better interface documentation, and the @CARP_NOT
1172 interface has been added to get optional control over where errors
1173 are reported independently of @ISA, by Ben Tilly.
1177 Class::Struct can now define the classes in compile time.
1181 Class::Struct now assigns the array/hash element if the accessor
1182 is called with an array/hash element as the B<sole> argument.
1186 The return value of Cwd::fastcwd() is now tainted.
1190 Data::Dumper now has an option to sort hashes.
1194 Data::Dumper now has an option to dump code references
1199 DB_File now supports newer Berkeley DB versions, among
1204 Devel::Peek now has an interface for the Perl memory statistics
1205 (this works only if you are using perl's malloc, and if you have
1206 compiled with debugging).
1210 The English module can now be used without the infamous performance
1213 use English '-no_match_vars';
1215 (Assuming, of course, that you don't need the troublesome variables
1216 C<$`>, C<$&>, or C<$'>.) Also, introduced C<@LAST_MATCH_START> and
1217 C<@LAST_MATCH_END> English aliases for C<@-> and C<@+>.
1221 ExtUtils::MakeMaker now uses File::Spec internally, which hopefully
1222 leads to better portability.
1226 Fcntl, Socket, and Sys::Syslog have been rewritten by Nicholas Clark
1227 to use the new-style constant dispatch section (see L<ExtUtils::Constant>).
1228 This means that they will be more robust and hopefully faster.
1232 File::Find now chdir()s correctly when chasing symbolic links. [561]
1236 File::Find now has pre- and post-processing callbacks. It also
1237 correctly changes directories when chasing symbolic links. Callbacks
1238 (naughtily) exiting with "next;" instead of "return;" now work.
1242 File::Find is now (again) reentrant. It also has been made
1247 The warnings issued by File::Find now belong to their own category.
1248 You can enable/disable them with C<use/no warnings 'File::Find';>.
1252 File::Glob::glob() has been renamed to File::Glob::bsd_glob()
1253 because the name clashes with the builtin glob(). The older
1254 name is still available for compatibility, but is deprecated. [561]
1258 File::Glob now supports C<GLOB_LIMIT> constant to limit the size of
1259 the returned list of filenames.
1263 IPC::Open3 now allows the use of numeric file descriptors.
1267 IO::Socket now has an atmark() method, which returns true if the socket
1268 is positioned at the out-of-band mark. The method is also exportable
1269 as a sockatmark() function.
1273 IO::Socket::INET failed to open the specified port if the service name
1274 was not known. It now correctly uses the supplied port number as is. [561]
1278 IO::Socket::INET has support for the ReusePort option (if your
1279 platform supports it). The Reuse option now has an alias, ReuseAddr.
1280 For clarity, you may want to prefer ReuseAddr.
1284 IO::Socket::INET now supports a value of zero for C<LocalPort>
1285 (usually meaning that the operating system will make one up.)
1289 'use lib' now works identically to @INC. Removing directories
1290 with 'no lib' now works.
1294 Math::BigFloat and Math::BigInt have undergone a full rewrite by Tels.
1295 They are now magnitudes faster, and they support various bignum
1296 libraries such as GMP and PARI as their backends.
1300 Math::Complex handles inf, NaN etc., better.
1304 Net::Ping has been considerably enhanced by Rob Brown: multihoming is
1305 now supported, Win32 functionality is better, there is now time
1306 measuring functionality (optionally high-resolution using
1307 Time::HiRes), and there is now "external" protocol which uses
1308 Net::Ping::External module which runs your external ping utility and
1309 parses the output. A version of Net::Ping::External is available in
1312 Note that some of the Net::Ping tests are disabled when running
1313 under the Perl distribution since one cannot assume one or more
1314 of the following: enabled echo port at localhost, full Internet
1315 connectivity, or sympathetic firewalls. You can set the environment
1316 variable PERL_TEST_Net_Ping to "1" (one) before running the Perl test
1317 suite to enable all the Net::Ping tests.
1321 POSIX::sigaction() is now much more flexible and robust.
1322 You can now install coderef handlers, 'DEFAULT', and 'IGNORE'
1323 handlers, installing new handlers was not atomic.
1327 In Safe, C<%INC> is now localised in a Safe compartment so that
1332 In SDBM_File on dosish platforms, some keys went missing because of
1333 lack of support for files with "holes". A workaround for the problem
1338 In Search::Dict one can now have a pre-processing hook for the
1339 lines being searched.
1343 The Shell module now has an OO interface.
1347 In Sys::Syslog there is now a failover mechanism that will go
1348 through alternative connection mechanisms until the message
1349 is successfully logged.
1353 The Test module has been significantly enhanced.
1357 Time::Local::timelocal() does not handle fractional seconds anymore.
1358 The rationale is that neither does localtime(), and timelocal() and
1359 localtime() are supposed to be inverses of each other.
1363 The vars pragma now supports declaring fully qualified variables.
1364 (Something that C<our()> does not and will not support.)
1368 The C<utf8::> name space (as in the pragma) provides various
1369 Perl-callable functions to provide low level access to Perl's
1370 internal Unicode representation. At the moment only length()
1371 has been implemented.
1375 =head1 Utility Changes
1381 Emacs perl mode (emacs/cperl-mode.el) has been updated to version
1386 F<emacs/e2ctags.pl> is now much faster.
1390 C<enc2xs> is a tool for people adding their own encodings to the
1395 C<h2ph> now supports C trigraphs.
1399 C<h2xs> now produces a template README.
1403 C<h2xs> now uses C<Devel::PPPort> for better portability between
1404 different versions of Perl.
1408 C<h2xs> uses the new L<ExtUtils::Constant|ExtUtils::Constant> module
1409 which will affect newly created extensions that define constants.
1410 Since the new code is more correct (if you have two constants where the
1411 first one is a prefix of the second one, the first constant B<never>
1412 got defined), less lossy (it uses integers for integer constant,
1413 as opposed to the old code that used floating point numbers even for
1414 integer constants), and slightly faster, you might want to consider
1415 regenerating your extension code (the new scheme makes regenerating
1416 easy). L<h2xs> now also supports C trigraphs.
1420 C<libnetcfg> has been added to configure libnet.
1424 C<perlbug> is now much more robust. It also sends the bug report to
1425 perl.org, not perl.com.
1429 C<perlcc> has been rewritten and its user interface (that is,
1430 command line) is much more like that of the UNIX C compiler, cc.
1431 (The perlbc tools has been removed. Use C<perlcc -B> instead.)
1432 B<Note that perlcc is still considered very experimental and
1437 C<perlivp> is a new Installation Verification Procedure utility
1438 for running any time after installing Perl.
1442 C<piconv> is an implementation of the character conversion utility
1443 C<iconv>, demonstrating the new Encode module.
1447 C<pod2html> now allows specifying a cache directory.
1451 C<pod2html> now produces XHTML 1.0.
1455 C<pod2html> now understands POD written using different line endings
1456 (PC-like CRLF versus UNIX-like LF versus MacClassic-like CR).
1460 C<s2p> has been completely rewritten in Perl. (It is in fact a full
1461 implementation of sed in Perl: you can use the sed functionality by
1462 using the C<psed> utility.)
1466 C<xsubpp> now understands POD documentation embedded in the *.xs
1471 C<xsubpp> now supports the OUT keyword.
1475 =head1 New Documentation
1481 perl56delta details the changes between the 5.005 release and the
1486 perlclib documents the internal replacements for standard C library
1487 functions. (Interesting only for extension writers and Perl core
1492 perldebtut is a Perl debugging tutorial. [561+]
1496 perlebcdic contains considerations for running Perl on EBCDIC
1501 perlintro is a gentle introduction to Perl.
1505 perliol documents the internals of PerlIO with layers.
1509 perlmodstyle is a style guide for writing modules.
1513 perlnewmod tells about writing and submitting a new module. [561+]
1517 perlpacktut is a pack() tutorial.
1521 perlpod has been rewritten to be clearer and to record the best
1522 practices gathered over the years.
1526 perlpodspec is a more formal specification of the pod format,
1527 mainly of interest for writers of pod applications, not to
1528 people writing in pod.
1532 perlretut is a regular expression tutorial. [561+]
1536 perlrequick is a regular expressions quick-start guide.
1537 Yes, much quicker than perlretut. [561]
1541 perltodo has been updated.
1545 perltootc has been renamed as perltooc (to not to conflict
1546 with perltoot in filesystems restricted to "8.3" names).
1550 perluniintro is an introduction to using Unicode in Perl.
1551 (perlunicode is more of a detailed reference and background
1556 perlutil explains the command line utilities packaged with the Perl
1557 distribution. [561+]
1561 The following platform-specific documents are available before
1562 the installation as README.I<platform>, and after the installation
1565 perlaix perlamiga perlapollo perlbeos perlbs2000
1566 perlce perlcygwin perldgux perldos perlepoc perlfreebsd perlhpux
1567 perlhurd perlirix perlmachten perlmacos perlmint perlmpeix
1568 perlnetware perlos2 perlos390 perlplan9 perlqnx perlsolaris
1569 perltru64 perluts perlvmesa perlvms perlvos perlwin32
1571 These documents usually detail one or more of the following subjects:
1572 configuring, building, testing, installing, and sometimes also using
1573 Perl on the said platform.
1575 Eastern Asian Perl users are now welcomed in their own languages:
1576 README.jp (Japanese), README.ko (Korean), README.cn (simplified
1577 Chinese) and README.tw (traditional Chinese), which are written in
1578 normal pod but encoded in EUC-JP, EUC-KR, EUC-CN and Big5. These
1579 will get installed as
1581 perljp perlko perlcn perltw
1587 The documentation for the POSIX-BC platform is called "BS2000", to avoid
1588 confusion with the Perl POSIX module.
1592 The documentation for the WinCE platform is called perlce (README.ce
1593 in the source code kit), to avoid confusion with the perlwin32
1594 documentation on 8.3-restricted filesystems.
1598 =head1 Performance Enhancements
1604 map() could get pathologically slow when the result list it generates
1605 is larger than the source list. The performance has been improved for
1606 common scenarios. [561]
1610 sort() is also fully reentrant, in the sense that the sort function
1611 can itself call sort(). This did not work reliably in previous
1616 sort() has been changed to use primarily mergesort internally as
1617 opposed to the earlier quicksort. For very small lists this may
1618 result in slightly slower sorting times, but in general the speedup
1619 should be at least 20%. Additional bonuses are that the worst case
1620 behaviour of sort() is now better (in computer science terms it now
1621 runs in time O(N log N), as opposed to quicksort's Theta(N**2)
1622 worst-case run time behaviour), and that sort() is now stable
1623 (meaning that elements with identical keys will stay ordered as they
1624 were before the sort). See the C<sort> pragma for information.
1626 The story in more detail: suppose you want to serve yourself a little
1629 @digits = ( 3,1,4,1,5,9 );
1631 A numerical sort of the digits will yield (1,1,3,4,5,9), as expected.
1632 Which C<1> comes first is hard to know, since one C<1> looks pretty
1633 much like any other. You can regard this as totally trivial,
1634 or somewhat profound. However, if you just want to sort the even
1635 digits ahead of the odd ones, then what will
1637 sort { ($a % 2) <=> ($b % 2) } @digits;
1639 yield? The only even digit, C<4>, will come first. But how about
1640 the odd numbers, which all compare equal? With the quicksort algorithm
1641 used to implement Perl 5.6 and earlier, the order of ties is left up
1642 to the sort. So, as you add more and more digits of Pi, the order
1643 in which the sorted even and odd digits appear will change.
1644 and, for sufficiently large slices of Pi, the quicksort algorithm
1645 in Perl 5.8 won't return the same results even if reinvoked with the
1646 same input. The justification for this rests with quicksort's
1647 worst case behavior. If you run
1649 sort { $a <=> $b } ( 1 .. $N , 1 .. $N );
1651 (something you might approximate if you wanted to merge two sorted
1652 arrays using sort), doubling $N doesn't just double the quicksort time,
1653 it I<quadruples> it. Quicksort has a worst case run time that can
1654 grow like N**2, so-called I<quadratic> behaviour, and it can happen
1655 on patterns that may well arise in normal use. You won't notice this
1656 for small arrays, but you I<will> notice it with larger arrays,
1657 and you may not live long enough for the sort to complete on arrays
1658 of a million elements. So the 5.8 quicksort scrambles large arrays
1659 before sorting them, as a statistical defence against quadratic behaviour.
1660 But that means if you sort the same large array twice, ties may be
1661 broken in different ways.
1663 Because of the unpredictability of tie-breaking order, and the quadratic
1664 worst-case behaviour, quicksort was I<almost> replaced completely with
1665 a stable mergesort. I<Stable> means that ties are broken to preserve
1666 the original order of appearance in the input array. So
1668 sort { ($a % 2) <=> ($b % 2) } (3,1,4,1,5,9);
1670 will yield (4,3,1,1,5,9), guaranteed. The even and odd numbers
1671 appear in the output in the same order they appeared in the input.
1672 Mergesort has worst case O(N log N) behaviour, the best value
1673 attainable. And, ironically, this mergesort does particularly
1674 well where quicksort goes quadratic: mergesort sorts (1..$N, 1..$N)
1675 in O(N) time. But quicksort was rescued at the last moment because
1676 it is faster than mergesort on certain inputs and platforms.
1677 For example, if you really I<don't> care about the order of even
1678 and odd digits, quicksort will run in O(N) time; it's very good
1679 at sorting many repetitions of a small number of distinct elements.
1680 The quicksort divide and conquer strategy works well on platforms
1681 with relatively small, very fast, caches. Eventually, the problem gets
1682 whittled down to one that fits in the cache, from which point it
1683 benefits from the increased memory speed.
1685 Quicksort was rescued by implementing a sort pragma to control aspects
1686 of the sort. The B<stable> subpragma forces stable behaviour,
1687 regardless of algorithm. The B<_quicksort> and B<_mergesort>
1688 subpragmas are heavy-handed ways to select the underlying implementation.
1689 The leading C<_> is a reminder that these subpragmas may not survive
1690 beyond 5.8. More appropriate mechanisms for selecting the implementation
1691 exist, but they wouldn't have arrived in time to save quicksort.
1695 Hashes now use Bob Jenkins "One-at-a-Time" hashing key algorithm
1696 ( http://burtleburtle.net/bob/hash/doobs.html ). This algorithm is
1697 reasonably fast while producing a much better spread of values than
1698 the old hashing algorithm (originally by Chris Torek, later tweaked by
1699 Ilya Zakharevich). Hash values output from the algorithm on a hash of
1700 all 3-char printable ASCII keys comes much closer to passing the
1701 DIEHARD random number generation tests. According to perlbench, this
1702 change has not affected the overall speed of Perl.
1706 unshift() should now be noticeably faster.
1710 =head1 Installation and Configuration Improvements
1712 =head2 Generic Improvements
1718 INSTALL now explains how you can configure Perl to use 64-bit
1719 integers even on non-64-bit platforms.
1723 Policy.sh policy change: if you are reusing a Policy.sh file
1724 (see INSTALL) and you use Configure -Dprefix=/foo/bar and in the old
1725 Policy $prefix eq $siteprefix and $prefix eq $vendorprefix, all of
1726 them will now be changed to the new prefix, /foo/bar. (Previously
1727 only $prefix changed.) If you do not like this new behaviour,
1728 specify prefix, siteprefix, and vendorprefix explicitly.
1732 A new optional location for Perl libraries, otherlibdirs, is available.
1733 It can be used for example for vendor add-ons without disturbing Perl's
1734 own library directories.
1738 In many platforms, the vendor-supplied 'cc' is too stripped-down to
1739 build Perl (basically, 'cc' doesn't do ANSI C). If this seems
1740 to be the case and 'cc' does not seem to be the GNU C compiler
1741 'gcc', an automatic attempt is made to find and use 'gcc' instead.
1745 gcc needs to closely track the operating system release to avoid
1746 build problems. If Configure finds that gcc was built for a different
1747 operating system release than is running, it now gives a clearly visible
1748 warning that there may be trouble ahead.
1752 Since Perl 5.8 is not binary-compatible with previous releases
1753 of Perl, Configure no longer suggests including the 5.005
1758 Configure C<-S> can now run non-interactively. [561]
1762 Configure support for pdp11-style memory models has been removed due
1763 to obsolescence. [561]
1767 configure.gnu now works with options with whitespace in them.
1771 installperl now outputs everything to STDERR.
1775 Because PerlIO is now the default on most platforms, "-perlio" doesn't
1776 get appended to the $Config{archname} (also known as $^O) anymore.
1777 Instead, if you explicitly choose not to use perlio (Configure command
1778 line option -Uuseperlio), you will get "-stdio" appended.
1782 Another change related to the architecture name is that "-64all"
1783 (-Duse64bitall, or "maximally 64-bit") is appended only if your
1784 pointers are 64 bits wide. (To be exact, the use64bitall is ignored.)
1788 In AFS installations, one can configure the root of the AFS to be
1789 somewhere else than the default F</afs> by using the Configure
1790 parameter C<-Dafsroot=/some/where/else>.
1794 APPLLIB_EXP, a lesser-known configuration-time definition, has been
1795 documented. It can be used to prepend site-specific directories
1796 to Perl's default search path (@INC); see INSTALL for information.
1800 The version of Berkeley DB used when the Perl (and, presumably, the
1801 DB_File extension) was built is now available as
1802 C<@Config{qw(db_version_major db_version_minor db_version_patch)}>
1803 from Perl and as C<DB_VERSION_MAJOR_CFG DB_VERSION_MINOR_CFG
1804 DB_VERSION_PATCH_CFG> from C.
1808 Building Berkeley DB3 for compatibility modes for DB, NDBM, and ODBM
1809 has been documented in INSTALL.
1813 If you have CPAN access (either network or a local copy such as a
1814 CD-ROM) you can during specify extra modules to Configure to build and
1815 install with Perl using the -Dextras=... option. See INSTALL for
1820 In addition to config.over, a new override file, config.arch, is
1821 available. This file is supposed to be used by hints file writers
1822 for architecture-wide changes (as opposed to config.over which is
1823 for site-wide changes).
1827 If your file system supports symbolic links, you can build Perl outside
1828 of the source directory by
1830 mkdir /tmp/perl/build/directory
1831 cd /tmp/perl/build/directory
1832 sh /path/to/perl/source/Configure -Dmksymlinks ...
1834 This will create in /tmp/perl/build/directory a tree of symbolic links
1835 pointing to files in /path/to/perl/source. The original files are left
1836 unaffected. After Configure has finished, you can just say
1840 and Perl will be built and tested, all in /tmp/perl/build/directory.
1845 For Perl developers, several new make targets for profiling
1846 and debugging have been added; see L<perlhack>.
1852 Use of the F<gprof> tool to profile Perl has been documented in
1853 L<perlhack>. There is a make target called "perl.gprof" for
1854 generating a gprofiled Perl executable.
1858 If you have GCC 3, there is a make target called "perl.gcov" for
1859 creating a gcoved Perl executable for coverage analysis. See
1864 If you are on IRIX or Tru64 platforms, new profiling/debugging options
1865 have been added; see L<perlhack> for more information about pixie and
1872 Guidelines of how to construct minimal Perl installations have
1873 been added to INSTALL.
1877 The Thread extension is now not built at all under ithreads
1878 (C<Configure -Duseithreads>) because it wouldn't work anyway (the
1879 Thread extension requires being Configured with C<-Duse5005threads>).
1881 But note that the Thread.pm interface is now shared by both
1886 The Gconvert macro ($Config{d_Gconvert}) used by perl for stringifying
1887 floating-point numbers is now more picky about using sprintf %.*g
1888 rules for the conversion. Some platforms that used to use gcvt may
1889 now resort to the slower sprintf.
1893 The obsolete method of making a special (e.g., debugging) flavor
1896 make LIBPERL=libperld.a
1898 has been removed. Use -DDEBUGGING instead.
1902 =head2 New Or Improved Platforms
1904 For the list of platforms known to support Perl,
1905 see L<perlport/"Supported Platforms">.
1911 AIX dynamic loading should be now better supported.
1915 AIX should now work better with gcc, threads, and 64-bitness. Also the
1916 long doubles support in AIX should be better now. See L<perlaix>.
1920 AtheOS ( http://www.atheos.cx/ ) is a new platform.
1924 BeOS has been reclaimed.
1928 The DG/UX platform now supports 5.005-style threads.
1933 The DYNIX/ptx platform (a.k.a. dynixptx) is supported at or near
1938 EBCDIC platforms (z/OS (also known as OS/390), POSIX-BC, and VM/ESA)
1939 have been regained. Many test suite tests still fail and the
1940 co-existence of Unicode and EBCDIC isn't quite settled, but the
1941 situation is much better than with Perl 5.6. See L<perlos390>,
1942 L<perlbs2000> (for POSIX-BC), and L<perlvmesa> for more information.
1946 Building perl with -Duseithreads or -Duse5005threads now works under
1947 HP-UX 10.20 (previously it only worked under 10.30 or later). You will
1948 need a thread library package installed. See README.hpux. [561]
1952 Mac OS Classic is now supported in the mainstream source package
1953 (MacPerl has of course been available since perl 5.004 but now the
1954 source code bases of standard Perl and MacPerl have been synchronised)
1959 Mac OS X (or Darwin) should now be able to build Perl even on HFS+
1960 filesystems. (The case-insensitivity used to confuse the Perl build
1965 NCR MP-RAS is now supported. [561]
1969 All the NetBSD specific patches (except for the installation
1970 specific ones) have been merged back to the main distribution.
1974 NetWare from Novell is now supported. See L<perlnetware>.
1978 NonStop-UX is now supported. [561]
1982 NEC SUPER-UX is now supported.
1986 All the OpenBSD specific patches (except for the installation
1987 specific ones) have been merged back to the main distribution.
1991 Perl has been tested with the GNU pth userlevel thread package
1992 ( http://www.gnu.org/software/pth/pth.html ). All thread tests
1993 of Perl now work, but not without adding some yield()s to the tests,
1994 so while pth (and other userlevel thread implementations) can be
1995 considered to be "working" with Perl ithreads, keep in mind the
1996 possible non-preemptability of the underlying thread implementation.
2000 Stratus VOS is now supported using Perl's native build method
2001 (Configure). This is the recommended method to build Perl on
2002 VOS. The older methods, which build miniperl, are still
2003 available. See L<perlvos>. [561+]
2007 The Amdahl UTS UNIX mainframe platform is now supported. [561]
2011 WinCE is now supported. See L<perlce>.
2015 z/OS (formerly known as OS/390, formerly known as MVS OE) now has
2016 support for dynamic loading. This is not selected by default,
2017 however, you must specify -Dusedl in the arguments of Configure. [561]
2021 =head1 Selected Bug Fixes
2023 Numerous memory leaks and uninitialized memory accesses have been
2024 hunted down. Most importantly, anonymous subs used to leak quite
2031 The autouse pragma didn't work for Multi::Part::Function::Names.
2035 caller() could cause core dumps in certain situations. Carp was
2036 sometimes affected by this problem. In particular, caller() now
2037 returns a subroutine name of C<(unknown)> for subroutines that have
2038 been removed from the symbol table.
2042 chop(@list) in list context returned the characters chopped in
2043 reverse order. This has been reversed to be in the right order. [561]
2047 Configure no longer includes the DBM libraries (dbm, gdbm, db, ndbm)
2048 when building the Perl binary. The only exception to this is SunOS 4.x,
2049 which needs them. [561]
2053 The behaviour of non-decimal but numeric string constants such as
2054 "0x23" was platform-dependent: in some platforms that was seen as 35,
2055 in some as 0, in some as a floating point number (don't ask). This
2056 was caused by Perl's using the operating system libraries in a situation
2057 where the result of the string to number conversion is undefined: now
2058 Perl consistently handles such strings as zero in numeric contexts.
2062 The order of DESTROYs has been made more predictable.
2066 Several debugger fixes: exit code now reflects the script exit code,
2067 condition C<"0"> now treated correctly, the C<d> command now checks
2068 line number, C<$.> no longer gets corrupted, and all debugger output
2069 now goes correctly to the socket if RemotePort is set. [561]
2073 Perl 5.6.0 could emit spurious warnings about redefinition of
2074 dl_error() when statically building extensions into perl.
2075 This has been corrected. [561]
2079 L<dprofpp> -R didn't work.
2083 C<*foo{FORMAT}> now works.
2087 Infinity is now recognized as a number.
2091 UNIVERSAL::isa no longer caches methods incorrectly. (This broke
2092 the Tk extension with 5.6.0.) [561]
2096 Lexicals I: lexicals outside an eval "" weren't resolved
2097 correctly inside a subroutine definition inside the eval "" if they
2098 were not already referenced in the top level of the eval""ed code.
2102 Lexicals II: lexicals leaked at file scope into subroutines that
2103 were declared before the lexicals.
2107 Lexical warnings now propagating correctly between scopes
2108 and into C<eval "...">.
2112 C<use warnings qw(FATAL all)> did not work as intended. This has been
2117 warnings::enabled() now reports the state of $^W correctly if the caller
2118 isn't using lexical warnings. [561]
2122 Line renumbering with eval and C<#line> now works. [561]
2126 Fixed numerous memory leaks, especially in eval "".
2130 Localised tied variables no longer leak memory
2133 tie my %tied_hash => 'Tie::StdHash';
2137 # Used to leak memory every time local() was called;
2138 # in a loop, this added up.
2139 local($tied_hash{Foo}) = 1;
2143 Localised hash elements (and %ENV) are correctly unlocalised to not
2144 exist, if they didn't before they were localised.
2148 tie my %tied_hash => 'Tie::StdHash';
2152 # Nothing has set the FOO element so far
2154 { local $tied_hash{FOO} = 'Bar' }
2156 # This used to print, but not now.
2157 print "exists!\n" if exists $tied_hash{FOO};
2159 As a side effect of this fix, tied hash interfaces B<must> define
2160 the EXISTS and DELETE methods.
2164 mkdir() now ignores trailing slashes in the directory name,
2165 as mandated by POSIX.
2169 Some versions of glibc have a broken modfl(). This affects builds
2170 with C<-Duselongdouble>. This version of Perl detects this brokenness
2171 and has a workaround for it. The glibc release 2.2.2 is known to have
2172 fixed the modfl() bug.
2176 Modulus of unsigned numbers now works (4063328477 % 65535 used to
2177 return 27406, instead of 27047). [561]
2181 Some "not a number" warnings introduced in 5.6.0 eliminated to be
2182 more compatible with 5.005. Infinity is now recognised as a number. [561]
2186 Numeric conversions did not recognize changes in the string value
2187 properly in certain circumstances. [561]
2191 Attributes (such as :shared) didn't work with our().
2195 our() variables will not cause bogus "Variable will not stay shared"
2200 "our" variables of the same name declared in two sibling blocks
2201 resulted in bogus warnings about "redeclaration" of the variables.
2202 The problem has been corrected. [561]
2206 pack "Z" now correctly terminates the string with "\0".
2210 Fix password routines which in some shadow password platforms
2211 (e.g. HP-UX) caused getpwent() to return every other entry.
2215 The PERL5OPT environment variable (for passing command line arguments
2216 to Perl) didn't work for more than a single group of options. [561]
2220 PERL5OPT with embedded spaces didn't work.
2224 printf() no longer resets the numeric locale to "C".
2228 C<qw(a\\b)> now parses correctly as C<'a\\b'>: that is, as three
2229 characters, not four. [561]
2233 pos() did not return the correct value within s///ge in earlier
2234 versions. This is now handled correctly. [561]
2238 Printing quads (64-bit integers) with printf/sprintf now works
2239 without the q L ll prefixes (assuming you are on a quad-capable platform).
2243 Regular expressions on references and overloaded scalars now work. [561+]
2247 Right-hand side magic (GMAGIC) could in many cases such as string
2248 concatenation be invoked too many times.
2252 scalar() now forces scalar context even when used in void context.
2256 SOCKS support is now much more robust.
2260 sort() arguments are now compiled in the right wantarray context
2261 (they were accidentally using the context of the sort() itself).
2262 The comparison block is now run in scalar context, and the arguments
2263 to be sorted are always provided list context. [561]
2267 Changed the POSIX character class C<[[:space:]]> to include the (very
2268 rarely used) vertical tab character. Added a new POSIX-ish character
2269 class C<[[:blank:]]> which stands for horizontal whitespace
2270 (currently, the space and the tab).
2274 The tainting behaviour of sprintf() has been rationalized. It does
2275 not taint the result of floating point formats anymore, making the
2276 behaviour consistent with that of string interpolation. [561]
2280 Some cases of inconsistent taint propagation (such as within hash
2281 values) have been fixed.
2285 The RE engine found in Perl 5.6.0 accidentally pessimised certain kinds
2286 of simple pattern matches. These are now handled better. [561]
2290 Regular expression debug output (whether through C<use re 'debug'>
2291 or via C<-Dr>) now looks better. [561]
2295 Multi-line matches like C<"a\nxb\n" =~ /(?!\A)x/m> were flawed. The
2296 bug has been fixed. [561]
2300 Use of $& could trigger a core dump under some situations. This
2301 is now avoided. [561]
2305 The regular expression captured submatches ($1, $2, ...) are now
2306 more consistently unset if the match fails, instead of leaving false
2307 data lying around in them. [561]
2311 readline() on files opened in "slurp" mode could return an extra
2312 "" (blank line) at the end in certain situations. This has been
2317 Autovivification of symbolic references of special variables described
2318 in L<perlvar> (as in C<${$num}>) was accidentally disabled. This works
2323 Sys::Syslog ignored the C<LOG_AUTH> constant.
2327 $AUTOLOAD, sort(), lock(), and spawning subprocesses
2328 in multiple threads simultaneously are now thread-safe.
2332 Tie::Array's SPLICE method was broken.
2336 Allow a read-only string on the left-hand side of a non-modifying tr///.
2340 If C<STDERR> is tied, warnings caused by C<warn> and C<die> now
2341 correctly pass to it.
2345 Several Unicode fixes.
2351 BOMs (byte order marks) at the beginning of Perl files
2352 (scripts, modules) should now be transparently skipped.
2353 UTF-16 and UCS-2 encoded Perl files should now be read correctly.
2357 The character tables have been updated to Unicode 3.2.0.
2361 Comparing with utf8 data does not magically upgrade non-utf8 data
2362 into utf8. (This was a problem for example if you were mixing data
2363 from I/O and Unicode data: your output might have got magically encoded
2368 Generating illegal Unicode code points such as U+FFFE, or the UTF-16
2369 surrogates, now also generates an optional warning.
2373 C<IsAlnum>, C<IsAlpha>, and C<IsWord> now match titlecase.
2377 Concatenation with the C<.> operator or via variable interpolation,
2378 C<eq>, C<substr>, C<reverse>, C<quotemeta>, the C<x> operator,
2379 substitution with C<s///>, single-quoted UTF8, should now work.
2383 The C<tr///> operator now works. Note that the C<tr///CU>
2384 functionality has been removed (but see pack('U0', ...)).
2388 C<eval "v200"> now works.
2392 Perl 5.6.0 parsed m/\x{ab}/ incorrectly, leading to spurious warnings.
2393 This has been corrected. [561]
2397 Zero entries were missing from the Unicode classes such as C<IsDigit>.
2403 Large unsigned numbers (those above 2**31) could sometimes lose their
2404 unsignedness, causing bogus results in arithmetic operations. [561]
2408 The Perl parser has been stress tested using both random input and
2409 Markov chain input and the few found crashes and lockups have been
2414 =head2 Platform Specific Changes and Fixes
2422 Perl now works on post-4.0 BSD/OSes.
2428 Setting C<$0> now works (as much as possible; see L<perlvar> for details).
2434 Numerous updates; currently synchronised with Cygwin 1.3.10.
2438 Previously DYNIX/ptx had problems in its Configure probe for non-blocking I/O.
2444 EPOC now better supported. See README.epoc. [561]
2450 Perl now works on post-3.0 FreeBSDs.
2456 README.hpux updated; C<Configure -Duse64bitall> now works;
2457 now uses HP-UX malloc instead of Perl malloc.
2463 Numerous compilation flag and hint enhancements; accidental mixing
2464 of 32-bit and 64-bit libraries (a doomed attempt) made much harder.
2474 Long doubles should now work (see INSTALL). [561]
2478 Linux previously had problems related to sockaddrlen when using
2479 accept(), recvfrom() (in Perl: recv()), getpeername(), and
2488 Compilation of the standard Perl distribution in Mac OS Classic should
2489 now work if you have the Metrowerks development environment and the
2490 missing Mac-specific toolkit bits. Contact the macperl mailing list
2497 MPE/iX update after Perl 5.6.0. See README.mpeix. [561]
2501 NetBSD/threads: try installing the GNU pth (should be in the
2502 packages collection, or http://www.gnu.org/software/pth/),
2503 and Configure with -Duseithreads.
2509 Perl now works on NetBSD/sparc.
2515 Now works with usethreads (see INSTALL). [561]
2521 64-bitness using the Sun Workshop compiler now works.
2527 The native build method requires at least VOS Release 14.5.0
2528 and GNU C++/GNU Tools 2.0.1 or later. The Perl pack function
2529 now maps overflowed values to +infinity and underflowed values
2534 Tru64 (aka Digital UNIX, aka DEC OSF/1)
2536 The operating system version letter now recorded in $Config{osvers}.
2537 Allow compiling with gcc (previously explicitly forbidden). Compiling
2538 with gcc still not recommended because buggy code results, even with
2545 Fixed various alignment problems that lead into core dumps either
2546 during build or later; no longer dies on math errors at runtime;
2547 now using full quad integers (64 bits), previously was using
2548 only 46 bit integers for speed.
2554 See L</"Socket Extension Dynamic in VMS"> and L</"IEEE-format Floating Point
2555 Default on OpenVMS Alpha"> for important changes not otherwise listed here.
2557 chdir() now works better despite a CRT bug; now works with MULTIPLICITY
2558 (see INSTALL); now works with Perl's malloc.
2560 The tainting of C<%ENV> elements via C<keys> or C<values> was previously
2561 unimplemented. It now works as documented.
2563 The C<waitpid> emulation has been improved. The worst bug (now fixed)
2564 was that a pid of -1 would cause a wildcard search of all processes on
2567 POSIX-style signals are now emulated much better on VMS versions prior
2570 The C<system> function and backticks operator have improved
2571 functionality and better error handling. [561]
2573 File access tests now use current process privileges rather than the
2574 user's default privileges, which could sometimes result in a mismatch
2575 between reported access and actual access.
2577 There is a new C<kill> implementation based on C<sys$sigprc> that allows
2578 older VMS systems (pre-7.0) to use C<kill> to send signals rather than
2579 simply force exit. This implementation also allows later systems to
2580 call C<kill> from within a signal handler.
2582 Iterative logical name translations are now limited to 10 iterations in
2583 imitation of SHOW LOGICAL and other OpenVMS facilities.
2593 accept() no longer leaks memory. [561]
2597 Borland C++ v5.5 is now a supported compiler that can build Perl.
2598 However, the generated binaries continue to be incompatible with those
2599 generated by the other supported compilers (GCC and Visual C++). [561]
2603 Better chdir() return value for a non-existent directory.
2607 Duping socket handles with open(F, ">&MYSOCK") now works under Windows
2612 New %ENV entries now propagate to subprocesses. [561]
2616 Current directory entries in %ENV are now correctly propagated to child
2621 $ENV{LIB} now used to search for libs under Visual C.
2625 fork() emulation has been improved in various ways, but still continues
2626 to be experimental. See L<perlfork> for known bugs and caveats. [561+]
2630 A failed (pseudo)fork now returns undef and sets errno to EAGAIN.
2634 Win32::GetCwd() correctly returns C:\ instead of C: when at the drive root.
2635 Other bugs in chdir() and Cwd::cwd() have also been fixed. [561]
2639 HTML files will be installed in c:\perl\html instead of c:\perl\lib\pod\html
2643 The makefiles now provide a single switch to bulk-enable all the
2644 features enabled in ActiveState ActivePerl (a popular Win32 binary
2645 distribution). [561]
2649 Allow REG_EXPAND_SZ keys in the registry.
2653 Can now send() from all threads, not just the first one. [561]
2657 ExtUtils::MakeMaker now uses $ENV{LIB} to search for libraries.
2661 Fake signal handling reenabled, bugs and all.
2665 %SIG has been enabled under USE_ITHREADS, but its use is completely
2666 unsupported under all configurations. [561]
2670 Less stack reserved per thread so that more threads can run
2671 concurrently. (Still 16M per thread.)
2675 C<< File::Spec->tmpdir() >> now prefers C:/temp over /tmp
2676 (works better when perl is running as service).
2680 Better UNC path handling under ithreads. [561]
2684 wait(), waitpid(), and backticks now return the correct exit status
2685 under Windows 9x. [561]
2689 Non-blocking waits for child processes (or pseudo-processes) are
2690 supported via C<waitpid($pid, &POSIX::WNOHANG)>.
2694 Win64 compilation is now supported.
2698 winsock handle leak fixed. [561]
2704 =head1 New or Changed Diagnostics
2710 The lexical warnings category "deprecated" is no longer a sub-category
2711 of the "syntax" category. It is now a top-level category in its own
2716 All regular expression compilation error messages are now hopefully
2717 easier to understand both because the error message now comes before
2718 the failed regex and because the point of failure is now clearly
2719 marked by a C<E<lt>-- HERE> marker.
2723 The various "opened only for", "on closed", "never opened" warnings
2724 drop the C<main::> prefix for filehandles in the C<main> package,
2725 for example C<STDIN> instead of C<main::STDIN>.
2729 The "Unrecognized escape" warning has been extended to include C<\8>,
2730 C<\9>, and C<\_>. There is no need to escape any of the C<\w> characters.
2734 Two new debugging options have been added: if you have compiled your
2735 Perl with debugging, you can use the -DT [561] and -DR options to trace
2736 tokenising and to add reference counts to displaying variables,
2741 The debugger (perl5db.pl) has been modified to present a more
2742 consistent commands interface, via (CommandSet=580). perl5db.t was
2743 also added to test the changes, and as a placeholder for further tests.
2749 The debugger has a new C<dumpDepth> option to control the maximum
2750 depth to which nested structures are dumped. The C<x> command has
2751 been extended so that C<x N EXPR> dumps out the value of I<EXPR> to a
2752 depth of at most I<N> levels.
2756 The debugger can now show lexical variables if you have the CPAN
2757 module PadWalker installed.
2761 If an attempt to use a (non-blessed) reference as an array index
2762 is made, a warning is given.
2766 C<push @a;> and C<unshift @a;> (with no values to push or unshift)
2767 now give a warning. This may be a problem for generated and evaled
2772 If you try to L<perlfunc/pack> a number less than 0 or larger than 255
2773 using the C<"C"> format you will get an optional warning. Similarly
2774 for the C<"c"> format and a number less than -128 or more than 127.
2778 Certain regex modifiers such as C<(?o)> make sense only if applied to
2779 the entire regex. You will get an optional warning if you try to do
2784 Using arrays or hashes as references (e.g. C<< %foo->{bar} >>
2785 has been deprecated for a while. Now you will get an optional warning.
2789 Using C<sort> in scalar context now issues an optional warning.
2790 This didn't do anything useful, as the sort was not performed.
2794 =head1 Changed Internals
2800 perlapi.pod (a companion to perlguts) now attempts to document the
2805 You can now build a really minimal perl called microperl.
2806 Building microperl does not require even running Configure;
2807 C<make -f Makefile.micro> should be enough. Beware: microperl makes
2808 many assumptions, some of which may be too bold; the resulting
2809 executable may crash or otherwise misbehave in wondrous ways.
2810 For careful hackers only.
2814 Added rsignal(), whichsig(), do_join(), op_clear, op_null,
2815 ptr_table_clear(), ptr_table_free(), sv_setref_uv(), and several UTF-8
2816 interfaces to the publicised API. For the full list of the available
2817 APIs see L<perlapi>.
2821 Made possible to propagate customised exceptions via croak()ing.
2825 Now xsubs can have attributes just like subs. (Well, at least the
2826 built-in attributes.)
2830 dTHR and djSP have been obsoleted; the former removed (because it's
2831 a no-op) and the latter replaced with dSP.
2835 PERL_OBJECT has been completely removed.
2839 The MAGIC constants (e.g. C<'P'>) have been macrofied
2840 (e.g. C<PERL_MAGIC_TIED>) for better source code readability
2841 and maintainability.
2845 The regex compiler now maintains a structure that identifies nodes in
2846 the compiled bytecode with the corresponding syntactic features of the
2847 original regex expression. The information is attached to the new
2848 C<offsets> member of the C<struct regexp>. See L<perldebguts> for more
2849 complete information.
2853 The C code has been made much more C<gcc -Wall> clean. Some warning
2854 messages still remain in some platforms, so if you are compiling with
2855 gcc you may see some warnings about dubious practices. The warnings
2856 are being worked on.
2860 F<perly.c>, F<sv.c>, and F<sv.h> have now been extensively commented.
2864 Documentation on how to use the Perl source repository has been added
2865 to F<Porting/repository.pod>.
2869 There are now several profiling make targets.
2873 =head1 Security Vulnerability Closed [561]
2875 (This change was already made in 5.7.0 but bears repeating here.)
2877 A potential security vulnerability in the optional suidperl component
2878 of Perl was identified in August 2000. suidperl is neither built nor
2879 installed by default. As of November 2001 the only known vulnerable
2880 platform is Linux, most likely all Linux distributions. CERT and
2881 various vendors and distributors have been alerted about the vulnerability.
2882 See http://www.cpan.org/src/5.0/sperl-2000-08-05/sperl-2000-08-05.txt
2883 for more information.
2885 The problem was caused by Perl trying to report a suspected security
2886 exploit attempt using an external program, /bin/mail. On Linux
2887 platforms the /bin/mail program had an undocumented feature which
2888 when combined with suidperl gave access to a root shell, resulting in
2889 a serious compromise instead of reporting the exploit attempt. If you
2890 don't have /bin/mail, or if you have 'safe setuid scripts', or if
2891 suidperl is not installed, you are safe.
2893 The exploit attempt reporting feature has been completely removed from
2894 Perl 5.8.0 (and the maintenance release 5.6.1, and it was removed also
2895 from all the Perl 5.7 releases), so that particular vulnerability
2896 isn't there anymore. However, further security vulnerabilities are,
2897 unfortunately, always possible. The suidperl functionality is most
2898 probably going to be removed in Perl 5.10. In any case, suidperl
2899 should only be used by security experts who know exactly what they are
2900 doing and why they are using suidperl instead of some other solution
2901 such as sudo ( see http://www.courtesan.com/sudo/ ).
2905 Several new tests have been added, especially for the F<lib> and F<ext>
2906 subsections. There are now about 65 000 individual tests (spread over
2907 about 700 test scripts), in the regression suite (5.6.1 has about
2908 11700 tests, in 258 test scripts) Many of the new tests are of course
2909 introduced by the new modules, but still in general Perl is now more
2912 Because of the large number of tests, running the regression suite
2913 will take considerably longer time than it used to: expect the suite
2914 to take up to 4-5 times longer to run than in perl 5.6. On a really
2915 fast machine you can hope to finish the suite in about 6-8 minutes
2918 The tests are now reported in a different order than in earlier Perls.
2919 (This happens because the test scripts from under t/lib have been moved
2920 to be closer to the library/extension they are testing.)
2922 =head1 Known Problems
2930 If using the AIX native make command, instead of just "make" issue
2931 "make all". In some setups the former has been known to spuriously
2932 also try to run "make install". Alternatively, you may want to use
2937 In AIX 4.2, Perl extensions that use C++ functions that use statics
2938 may have problems in that the statics are not getting initialized.
2939 In newer AIX releases, this has been solved by linking Perl with
2940 the libC_r library, but unfortunately in AIX 4.2 the said library
2941 has an obscure bug where the various functions related to time
2942 (such as time() and gettimeofday()) return broken values, and
2943 therefore in AIX 4.2 Perl is not linked against libC_r.
2947 vac 5.0.0.0 May Produce Buggy Code For Perl
2949 The AIX C compiler vac version 5.0.0.0 may produce buggy code,
2950 resulting in a few random tests failing when run as part of "make
2951 test", but when the failing tests are run by hand, they succeed.
2952 We suggest upgrading to at least vac version 5.0.1.0, that has been
2953 known to compile Perl correctly. "lslpp -L|grep vac.C" will tell
2954 you the vac version. See README.aix.
2958 If building threaded Perl, you may get compilation warning from pp_sys.c:
2960 "pp_sys.c", line 4651.39: 1506-280 (W) Function argument assignment between types "unsigned char*" and "const void*" is not allowed.
2962 This is harmless; it is caused by the getnetbyaddr() and getnetbyaddr_r()
2963 having slightly different types for their first argument.
2967 =head2 Alpha systems with old gccs fail several tests
2969 If you see op/pack, op/pat, op/regexp, or ext/Storable tests failing
2970 in a Linux/alpha or *BSD/Alpha, it's probably time to upgrade your gcc.
2971 gccs prior to 2.95.3 are definitely not good enough, and gcc 3.1 may
2972 be even better. (RedHat Linux/alpha with gcc 3.1 reported no problems,
2973 as did Linux 2.4.18 with gcc 2.95.4.) (In Tru64, it is preferable to
2974 use the bundled C compiler.)
2978 Perl 5.8.0 doesn't build in AmigaOS. It broke at some point during
2979 the ithreads work and we could not find Amiga experts to unbreak the
2980 problems. Perl 5.6.1 still works for AmigaOS (as does the the 5.7.2
2981 development release).
2985 The following tests fail on 5.8.0 Perl in BeOS Personal 5.03:
2987 t/op/lfs............................FAILED at test 17
2988 t/op/magic..........................FAILED at test 24
2989 ext/POSIX/t/sigaction...............FAILED at test 13
2990 ext/POSIX/t/waitpid.................FAILED at test 1
2992 See L<perlbeos> (README.beos) for more details.
2994 =head2 Cygwin "unable to remap"
2996 For example when building the Tk extension for Cygwin,
2997 you may get an error message saying "unable to remap".
2998 This is known problem with Cygwin, and a workaround is
2999 detailed in here: http://sources.redhat.com/ml/cygwin/2001-12/msg00894.html
3001 =head2 ext/threads/t/libc
3003 If this test fails, it indicates that your libc (C library) is not
3004 threadsafe. This particular test stress tests the localtime() call to
3005 find out whether it is threadsafe. See L<perlthrtut> for more information.
3007 =head2 FreeBSD built with ithreads coredumps reading large directories
3009 This is a known bug in FreeBSD's readdir_r() (see L<perlfreebsd>
3010 (README.freebsd)), which hopefully will be fixed in FreeBSD 4.6.
3012 =head2 FreeBSD Failing locale Test 117 For ISO8859-15 Locales
3014 The ISO8859-15 locales may fail the locale test 117 in FreeBSD.
3015 This is caused by the characters \xFF (y with diaeresis) and \xBE
3016 (Y with diaeresis) not behaving correctly when being matched
3019 =head2 IRIX fails ext/List/Util/t/shuffle.t
3021 IRIX with MIPSpro 7.3.1.3m compiler may fail the said List::Util test
3022 by dumping core. This seems to be a compiler error since if compiled
3023 with gcc no core dump ensues, and no failures on the said test on any
3026 =head2 Modifying $_ Inside for(..)
3030 works without complaint. It shouldn't. (You should be able to
3031 modify only lvalue elements inside the loops.) You can see the
3032 correct behaviour by replacing the 1..5 with 1, 2, 3, 4, 5.
3034 =head2 mod_perl 1.26 Doesn't Build With Threaded Perl
3036 Use mod_perl 1.27 or higher.
3038 =head2 lib/ftmp-security tests warn 'system possibly insecure'
3040 Don't panic. Read the 'make test' section of INSTALL instead.
3042 =head2 HP-UX lib/posix Subtest 9 Fails When LP64-Configured
3044 If perl is configured with -Duse64bitall, the successful result of the
3045 subtest 10 of lib/posix may arrive before the successful result of the
3046 subtest 9, which confuses the test harness so much that it thinks the
3049 =head2 Linux with glibc 2.2.5 fails t/op/int subtest #6 with -Duse64bitint
3051 This is a known bug in the glibc 2.2.5 with long long integers.
3052 ( http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=65612 )
3054 =head2 Linux With Sfio Fails op/misc Test 48
3058 =head2 libwww-perl (LWP) fails base/date #51
3060 Use libwww-perl 5.65 or later.
3064 Please remember to set your environment variable LC_ALL to "C"
3065 (setenv LC_ALL C) before running "make test" to avoid a lot of
3066 warnings about the broken locales of Mac OS X.
3068 The following tests are known to fail in Mac OS X 10.1.5 because of
3069 buggy (old) implementations of Berkeley DB included in Mac OS X:
3071 Failed Test Stat Wstat Total Fail Failed List of Failed
3072 -------------------------------------------------------------------------
3073 ../ext/DB_File/t/db-btree.t 0 11 ?? ?? % ??
3074 ../ext/DB_File/t/db-recno.t 149 3 2.01% 61 63 65
3076 If you are building on a UFS partition, you will also probably see
3077 t/op/stat.t subtest #9 fail. This is caused by Darwin's UFS not
3078 supporting inode change time.
3080 Also the ext/POSIX/t/posix.t subtest #10 fails but it is skipped for
3081 now because the failure is Apple's fault, not Perl's (blocked signals
3084 If you Configure with ithreads, ext/threads/t/libc.t will fail. Again,
3085 this is not Perl's fault-- the libc of Mac OS X is not threadsafe
3086 (in this particular test, the localtime() call is found to be
3089 =head2 op/sprintf tests 91, 129, and 130
3091 The op/sprintf tests 91, 129, and 130 are known to fail on some platforms.
3092 Examples include any platform using sfio, and Compaq/Tandem's NonStop-UX.
3094 Test 91 is known to fail on QNX6 (nto), because C<sprintf '%e',0>
3095 incorrectly produces C<0.000000e+0> instead of C<0.000000e+00>.
3097 For tests 129 and 130, the failing platforms do not comply with
3098 the ANSI C Standard: lines 19ff on page 134 of ANSI X3.159 1989, to
3099 be exact. (They produce something other than "1" and "-1" when
3100 formatting 0.6 and -0.6 using the printf format "%.0f"; most often,
3101 they produce "0" and "-0".)
3105 In case you are still using Solaris 2.5 (aka SunOS 5.5), you may
3106 experience failures (the test core dumping) in lib/locale.t.
3107 The suggested cure is to upgrade your Solaris.
3109 =head2 SUPER-UX (NEC SX)
3111 The following tests are known to fail on SUPER-UX:
3113 op/64bitint...........................FAILED tests 29-30, 32-33, 35-36
3114 op/arith..............................FAILED tests 128-130
3115 op/pack...............................FAILED tests 25-5625
3116 op/pow................................
3117 op/taint..............................# msgsnd failed
3118 ../ext/IO/lib/IO/t/io_poll............FAILED tests 3-4
3119 ../ext/IPC/SysV/ipcsysv...............FAILED tests 2, 5-6
3120 ../ext/IPC/SysV/t/msg.................FAILED tests 2, 4-6
3121 ../ext/Socket/socketpair..............FAILED tests 12
3122 ../lib/IPC/SysV.......................FAILED tests 2, 5-6
3123 ../lib/warnings.......................FAILED tests 115-116, 118-119
3125 The op/pack failure ("Cannot compress negative numbers at op/pack.t line 126")
3126 is serious but as of yet unsolved. It points at some problems with the
3127 signedness handling of the C compiler, as do the 64bitint, arith, and pow
3128 failures. Most of the rest point at problems with SysV IPC.
3130 =head2 Term::ReadKey not working on Win32
3132 Use Term::ReadKey 2.20 or later.
3134 =head2 Failure of Thread (5.005-style) tests
3136 B<Note that support for 5.005-style threading is deprecated,
3137 experimental and practically unsupported. In 5.10, it is expected
3140 The following tests are known to fail due to fundamental problems in
3141 the 5.005 threading implementation. These are not new failures--Perl
3142 5.005_0x has the same bugs, but didn't have these tests.
3144 ../ext/B/t/xref.t 255 65280 14 12 85.71% 3-14
3145 ../ext/List/Util/t/first.t 255 65280 7 4 57.14% 2 5-7
3146 ../lib/English.t 2 512 54 2 3.70% 2-3
3147 ../lib/ExtUtils/t/basic.t 1 256 17 1 5.88% 14
3148 ../lib/FileCache.t 5 1 20.00% 5
3149 ../lib/Filter/Simple/t/data.t 6 3 50.00% 1-3
3150 ../lib/Filter/Simple/t/filter_onl 9 3 33.33% 1-2 5
3151 ../lib/Tie/File/t/31_autodefer.t 255 65280 65 32 49.23% 34-65
3152 ../lib/autouse.t 10 1 10.00% 4
3153 op/flip.t 15 1 6.67% 15
3155 These failures are unlikely to get fixed as 5.005-style threads
3156 are considered fundamentally broken. (Basically what happens is that
3157 competing threads can corrupt shared global state.)
3159 =head2 Timing problems
3161 The following tests may fail intermittently because of timing
3162 problems, for example if the system is heavily loaded.
3165 ext/Time/HiRes/HiRes.t
3167 lib/Memoize/t/expmod_t.t
3168 lib/Memoize/t/speed.t
3170 In case of failure please try running them manually, for example
3172 ./perl -Ilib ext/Time/HiRes/HiRes.t
3176 ../lib/Math/Trig.t 26 1 3.85% 25
3177 ../lib/warnings.t 470 1 0.21% 429
3179 The Trig.t failure is caused by the slighly differing (from IEEE)
3180 floating point implementation of UNICOS. The warnings.t failure is
3181 also related: the test assumes a certain floating point output format;
3182 this assumption fails in UNICOS.
3190 During Configure, the test
3192 Guessing which symbols your C compiler and preprocessor define...
3194 will probably fail with error messages like
3196 CC-20 cc: ERROR File = try.c, Line = 3
3197 The identifier "bad" is undefined.
3199 bad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K
3202 CC-65 cc: ERROR File = try.c, Line = 3
3203 A semicolon is expected at this point.
3205 This is caused by a bug in the awk utility of UNICOS/mk. You can ignore
3206 the error, but it does cause a slight problem: you cannot fully
3207 benefit from the h2ph utility (see L<h2ph>) that can be used to
3208 convert C headers to Perl libraries, mainly used to be able to access
3209 from Perl the constants defined using C preprocessor, cpp. Because of
3210 the above error, parts of the converted headers will be invisible.
3211 Luckily, these days the need for h2ph is rare.
3215 If building Perl with interpreter threads (ithreads), the
3216 getgrent(), getgrnam(), and getgrgid() functions cannot return the
3217 list of the group members due to a bug in the multithreaded support of
3218 UNICOS/mk. What this means is that in list context the functions will
3219 return only three values, not four.
3225 There are a few known test failures, see L<perluts> (README.uts).
3227 =head2 VOS (Stratus)
3229 When Perl is built using the native build process on VOS Release
3230 14.5.0 and GNU C++/GNU Tools 2.0.1, all attempted tests either
3231 pass or result in TODO (ignored) failures.
3235 There should be no reported test failures with a default configuration,
3236 though there are a number of tests marked TODO that point to areas
3237 needing further debugging and/or porting work.
3241 In multi-CPU boxes, there are some problems with the I/O buffering:
3242 some output may appear twice.
3244 =head2 XML::Parser not working
3246 Use XML::Parser 2.31 or later.
3248 =head2 z/OS (OS/390)
3250 z/OS has rather many test failures but the situation is actually
3251 better than it was in 5.6.0; it's just that so many new modules and
3252 tests have been added.
3254 Failed Test Stat Wstat Total Fail Failed List of Failed
3255 ---------------------------------------------------------------------------
3256 ../ext/Data/Dumper/t/dumper.t 357 8 2.24% 311 314 325 327
3258 ../ext/IO/lib/IO/t/io_unix.t 5 4 80.00% 2-5
3259 ../ext/Storable/t/downgrade.t 12 3072 169 12 7.10% 14-15 46-47 78-79
3261 ../lib/ExtUtils/t/Constant.t 121 30976 48 48 100.00% 1-48
3262 ../lib/ExtUtils/t/Embed.t 9 9 100.00% 1-9
3263 op/pat.t 910 7 0.77% 665 776 785 832-
3265 op/sprintf.t 224 3 1.34% 98 100 136
3266 op/tr.t 97 5 5.15% 63 71-74
3267 uni/fold.t 780 6 0.77% 61 169 196 661
3270 The failures in dumper.t and downgrade.t are problems in the tests,
3271 those in io_unix and sprintf are problems in the USS (UDP sockets
3272 and printf formats). The pat, tr, and fold failures are genuine Perl
3273 problems caused by EBCDIC (and in the pat and fold cases, combining
3274 that with Unicode). The Constant and Embed are probably problems
3275 in the tests (since they test Perl's ability to build extensions,
3276 and that seems to be working reasonably well.)
3278 =head2 Localising Tied Arrays and Hashes Is Broken
3282 doesn't work as one would expect: the old value is restored
3283 incorrectly. This will be changed in a future release, but we don't
3284 know yet what the new semantics will exactly be. In any case, the
3285 change will break existing code that relies on the current
3286 (ill-defined) semantics, so just avoid doing this in general.
3288 =head2 Self-tying Problems
3290 Self-tying of arrays and hashes is broken in rather deep and
3291 hard-to-fix ways. As a stop-gap measure to avoid people from getting
3292 frustrated at the mysterious results (core dumps, most often), it is
3293 forbidden for now (you will get a fatal error even from an attempt).
3295 A change to self-tying of globs has caused them to be recursively
3296 referenced (see: L<perlobj/"Two-Phased Garbage Collection">). You
3297 will now need an explicit untie to destroy a self-tied glob. This
3298 behaviour may be fixed at a later date.
3300 Self-tying of scalars and IO thingies works.
3302 =head2 Building Extensions Can Fail Because Of Largefiles
3304 Some extensions like mod_perl are known to have issues with
3305 `largefiles', a change brought by Perl 5.6.0 in which file offsets
3306 default to 64 bits wide, where supported. Modules may fail to compile
3307 at all, or they may compile and work incorrectly. Currently, there
3308 is no good solution for the problem, but Configure now provides
3309 appropriate non-largefile ccflags, ldflags, libswanted, and libs
3310 in the %Config hash (e.g., $Config{ccflags_nolargefiles}) so the
3311 extensions that are having problems can try configuring themselves
3312 without the largefileness. This is admittedly not a clean solution,
3313 and the solution may not even work at all. One potential failure is
3314 whether one can (or, if one can, whether it's a good idea to) link
3315 together at all binaries with different ideas about file offsets;
3316 all this is platform-dependent.
3318 =head2 Unicode Support on EBCDIC Still Spotty
3320 Though mostly working, Unicode support still has problem spots on
3321 EBCDIC platforms. One such known spot are the C<\p{}> and C<\P{}>
3322 regular expression constructs for code points less than 256: the
3323 C<pP> are testing for Unicode code points, not knowing about EBCDIC.
3325 =head2 The Compiler Suite Is Still Very Experimental
3327 The compiler suite is slowly getting better but it continues to be
3328 highly experimental. Use in production environments is discouraged.
3330 =head2 The Long Double Support Is Still Experimental
3332 The ability to configure Perl's numbers to use "long doubles",
3333 floating point numbers of hopefully better accuracy, is still
3334 experimental. The implementations of long doubles are not yet
3335 widespread and the existing implementations are not quite mature
3336 or standardised, therefore trying to support them is a rare
3337 and moving target. The gain of more precision may also be offset
3338 by slowdown in computations (more bits to move around, and the
3339 operations are more likely to be executed by less optimised
3342 =head2 Seen In Perl 5.7 But Gone Now
3344 C<Time::Piece> (previously known as C<Time::Object>) was removed
3345 because it was felt that it didn't have enough value in it to be a
3346 core module. It is still a useful module, though, and is available
3349 Perl 5.8 unfortunately does not build anymore on AmigaOS; this broke
3350 accidentally at some point. Since there are not that many Amiga
3351 developers available, we could not get this fixed and tested in time
3352 for 5.8.0. Perl 5.6.1 still works for AmigaOS (as does the the 5.7.2
3353 development release).
3355 =head1 Reporting Bugs
3357 If you find what you think is a bug, you might check the articles
3358 recently posted to the comp.lang.perl.misc newsgroup and the perl
3359 bug database at http://bugs.perl.org/ . There may also be
3360 information at http://www.perl.com/ , the Perl Home Page.
3362 If you believe you have an unreported bug, please run the B<perlbug>
3363 program included with your release. Be sure to trim your bug down
3364 to a tiny but sufficient test case. Your bug report, along with the
3365 output of C<perl -V>, will be sent off to perlbug@perl.org to be
3366 analysed by the Perl porting team.
3370 The F<Changes> file for exhaustive details on what changed.
3372 The F<INSTALL> file for how to build Perl.
3374 The F<README> file for general stuff.
3376 The F<Artistic> and F<Copying> files for copyright information.
3380 Written by Jarkko Hietaniemi <F<jhi@iki.fi>>.