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 PerlIO C<:raw> discipline (as described in Camel III) is deprecated
237 because its definition (as either the discipline version of C<binmode(FH)>
238 or as the opposite of C<:crlf>) didn't really work: most importantly
239 because turning off "clrfness" is not enough to make a stream truly
240 binary. Instead of C<:raw> use one of the following: C<open(..., ':bytes')>,
241 C<binmode(FH)>, C<sysopen()> + C<sysread()>.
245 The current user-visible implementation of pseudo-hashes (the weird
246 use of the first array element) is deprecated starting from Perl 5.8.0
247 and will be removed in Perl 5.10.0, and the feature will be
248 implemented differently. Not only is the current interface rather
249 ugly, but the current implementation slows down normal array and hash
250 use quite noticeably. The C<fields> pragma interface will remain
251 available. The I<restricted hashes> interface is expected to
252 be the replacement interface (see L<Hash::Util>).
256 The syntaxes C<< @a->[...] >> and C<< %h->{...} >> have now been deprecated.
260 After years of trying, suidperl is considered to be too complex to
261 ever be considered truly secure. The suidperl functionality is likely
262 to be removed in a future release.
266 The 5.005 threads model (module C<Thread>) is deprecated and expected
267 to be removed in Perl 5.10. Multithreaded code should be migrated to
268 the new ithreads model (see L<threads>, L<threads::shared> and
273 The long deprecated uppercase aliases for the string comparison
274 operators (EQ, NE, LT, LE, GE, GT) have now been removed.
278 The tr///C and tr///U features have been removed and will not return;
279 the interface was a mistake. Sorry about that. For similar
280 functionality, see pack('U0', ...) and pack('C0', ...). [561]
284 Earlier Perls treated "sub foo (@bar)" as equivalent to "sub foo (@)".
285 The prototypes are now checked better at compile-time for invalid
286 syntax. An optional warning is generated ("Illegal character in
287 prototype...") but this may be upgraded to a fatal error in a future
292 The C<exec LIST> and C<system LIST> operations will produce fatal
293 errors on tainted data in some future release.
297 The existing behaviour when localising tied arrays and hashes is wrong,
298 and will be changed in a future release, so do not rely on the existing
299 behaviour. See L<"Localising Tied Arrays and Hashes Is Broken">.
303 =head1 Core Enhancements
305 =head2 PerlIO is Now The Default
311 IO is now by default done via PerlIO rather than system's "stdio".
312 PerlIO allows "layers" to be "pushed" onto a file handle to alter the
313 handle's behaviour. Layers can be specified at open time via 3-arg
316 open($fh,'>:crlf :utf8', $path) || ...
318 or on already opened handles via extended C<binmode>:
320 binmode($fh,':encoding(iso-8859-7)');
322 The built-in layers are: unix (low level read/write), stdio (as in
323 previous Perls), perlio (re-implementation of stdio buffering in a
324 portable manner), crlf (does CRLF <=> "\n" translation as on Win32,
325 but available on any platform). A mmap layer may be available if
326 platform supports it (mostly UNIXes).
328 Layers to be applied by default may be specified via the 'open' pragma.
330 See L</"Installation and Configuration Improvements"> for the effects
331 of PerlIO on your architecture name.
335 File handles can be marked as accepting Perl's internal encoding of Unicode
336 (UTF-8 or UTF-EBCDIC depending on platform) by a pseudo layer ":utf8" :
338 open($fh,">:utf8","Uni.txt");
340 Note for EBCDIC users: the pseudo layer ":utf8" is erroneously named
341 for you since it's not UTF-8 what you will be getting but instead
342 UTF-EBCDIC. See L<perlunicode>, L<utf8>, and
343 http://www.unicode.org/unicode/reports/tr16/ for more information.
344 In future releases this naming may change. See L<perluniintro>
345 for more information about UTF-8.
349 If your environment variables (LC_ALL, LC_CTYPE, LANG, LANGUAGE) look
350 like you want to use UTF-8 (any of the the variables match C</utf-?8/i>),
351 your STDIN, STDOUT, STDERR handles and the default open discipline
352 (see L<open>) are marked as UTF-8. (This feature, like other new
353 features that combine Unicode and I/O, work only if you are using
354 PerlIO, but that's is the default.)
356 Note that after this Perl really does assume that everything is UTF-8:
357 for example if some input handle is not, Perl will probably very soon
358 complain about the input data like this "Malformed UTF-8 ..." since
359 any old eight-bit data is not legal UTF-8.
361 Note for code authors: if you want to enable your users to use UTF-8
362 as their default encoding but in your code still have eight-bit I/O streams
363 (such as images or zip files), you need to explicitly open() or binmode()
364 with C<:bytes> (see L<perlfunc/open> and L<perlfunc/binmode>), or you
365 can just use C<binmode(FH)> (nice for pre-5.8.0 backward compatibility).
369 File handles can translate character encodings from/to Perl's internal
370 Unicode form on read/write via the ":encoding()" layer.
374 File handles can be opened to "in memory" files held in Perl scalars via:
376 open($fh,'>', \$variable) || ...
380 Anonymous temporary files are available without need to
381 'use FileHandle' or other module via
383 open($fh,"+>", undef) || ...
385 That is a literal undef, not an undefined value.
389 The list form of C<open> is now implemented for pipes (at least on UNIX):
391 open($fh,"-|", 'cat', '/etc/motd')
393 creates a pipe, and runs the equivalent of exec('cat', '/etc/motd') in
398 If your locale environment variables (LANGUAGE, LC_ALL, LC_CTYPE, LANG)
399 contain the strings 'UTF-8' or 'UTF8' (case-insensitive matching),
400 the default encoding of your STDIN, STDOUT, and STDERR, and of
401 B<any subsequent file open>, is UTF-8.
405 =head2 Restricted Hashes
407 A restricted hash is restricted to a certain set of keys, no keys
408 outside the set can be added. Also individual keys can be restricted
409 so that the key cannot be deleted and the value cannot be changed.
410 No new syntax is involved: the Hash::Util module is the interface.
414 Perl used to be fragile in that signals arriving at inopportune moments
415 could corrupt Perl's internal state. Now Perl postpones handling of
416 signals until it's safe (between opcodes).
418 This change may have surprising side effects because signals no longer
419 interrupt Perl instantly. Perl will now first finish whatever it was
420 doing, like finishing an internal operation (like sort()) or an
421 external operation (like an I/O operation), and only then look at any
422 arrived signals (and before starting the next operation). No more corrupt
423 internal state since the current operation is always finished first,
424 but the signal may take more time to get heard. Note that breaking
425 out from potentially blocking operations should still work, though.
427 =head2 Unicode Overhaul
429 Unicode in general should be now much more usable than in Perl 5.6.0
430 (or even in 5.6.1). Unicode can be used in hash keys, Unicode in
431 regular expressions should work now, Unicode in tr/// should work now,
432 Unicode in I/O should work now. See L<perluniintro> for introduction
433 and L<perlunicode> for details.
439 The Unicode Character Database coming with Perl has been upgraded
440 to Unicode 3.2.0. For more information, see http://www.unicode.org/ .
441 [561+] (5.6.1 has UCD 3.0.1.)
445 For developers interested in enhancing Perl's Unicode capabilities:
446 almost all the UCD files are included with the Perl distribution in
447 the F<lib/unicore> subdirectory. The most notable omission, for space
448 considerations, is the Unihan database.
452 The properties \p{Blank} and \p{SpacePerl} have been added. "Blank" is like
453 C isblank(), that is, it contains only "horizontal whitespace" (the space
454 character is, the newline isn't), and the "SpacePerl" is the Unicode
455 equivalent of C<\s> (\p{Space} isn't, since that includes the vertical
456 tabulator character, whereas C<\s> doesn't.)
458 See "New Unicode Properties" earlier in this document for additional
459 information on changes with Unicode properties.
463 =head2 Understanding of Numbers
465 In general a lot of fixing has happened in the area of Perl's
466 understanding of numbers, both integer and floating point. Since in
467 many systems the standard number parsing functions like C<strtoul()>
468 and C<atof()> seem to have bugs, Perl tries to work around their
469 deficiencies. This results hopefully in more accurate numbers.
471 Perl now tries internally to use integer values in numeric conversions
472 and basic arithmetics (+ - * /) if the arguments are integers, and
473 tries also to keep the results stored internally as integers.
474 This change leads to often slightly faster and always less lossy
475 arithmetics. (Previously Perl always preferred floating point numbers
478 =head2 Arrays now always interpolate into double-quoted strings [561]
480 In double-quoted strings, arrays now interpolate, no matter what. The
481 behavior in earlier versions of perl 5 was that arrays would interpolate
482 into strings if the array had been mentioned before the string was
483 compiled, and otherwise Perl would raise a fatal compile-time error.
484 In versions 5.000 through 5.003, the error was
486 Literal @example now requires backslash
488 In versions 5.004_01 through 5.6.0, the error was
490 In string, @example now must be written as \@example
492 The idea here was to get people into the habit of writing
493 C<"fred\@example.com"> when they wanted a literal C<@> sign, just as
494 they have always written C<"Give me back my \$5"> when they wanted a
497 Starting with 5.6.1, when Perl now sees an C<@> sign in a
498 double-quoted string, it I<always> attempts to interpolate an array,
499 regardless of whether or not the array has been used or declared
500 already. The fatal error has been downgraded to an optional warning:
502 Possible unintended interpolation of @example in string
504 This warns you that C<"fred@example.com"> is going to turn into
505 C<fred.com> if you don't backslash the C<@>.
506 See http://www.plover.com/~mjd/perl/at-error.html for more details
507 about the history here.
509 =head2 Miscellaneous Changes
515 AUTOLOAD is now lvaluable, meaning that you can add the :lvalue attribute
516 to AUTOLOAD subroutines and you can assign to the AUTOLOAD return value.
520 The $Config{byteorder} (and corresponding BYTEORDER in config.h) was
521 previously wrong in platforms if sizeof(long) was 4, but sizeof(IV)
522 was 8. The byteorder was only sizeof(long) bytes long (1234 or 4321),
523 but now it is correctly sizeof(IV) bytes long, (12345678 or 87654321).
524 (This problem didn't affect Windows platforms.)
526 Also, $Config{byteorder} is now computed dynamically--this is more
527 robust with "fat binaries" where an executable image contains binaries
528 for more than one binary platform, and when cross-compiling.
532 C<perl -d:Module=arg,arg,arg> now works (previously one couldn't pass
533 in multiple arguments.)
537 C<do> followed by a bareword now ensures that this bareword isn't
538 a keyword (to avoid a bug where C<do q(foo.pl)> tried to call a
539 subroutine called C<q>). This means that for example instead of
540 C<do format()> you must write C<do &format()>.
544 The builtin dump() now gives an optional warning
545 C<dump() better written as CORE::dump()>,
546 meaning that by default C<dump(...)> is resolved as the builtin
547 dump() which dumps core and aborts, not as (possibly) user-defined
548 C<sub dump>. To call the latter, qualify the call as C<&dump(...)>.
549 (The whole dump() feature is to considered deprecated, and possibly
550 removed/changed in future releases.)
554 chomp() and chop() are now overridable. Note, however, that their
555 prototype (as given by C<prototype("CORE::chomp")> is undefined,
556 because it cannot be expressed and therefore one cannot really write
557 replacements to override these builtins.
561 END blocks are now run even if you exit/die in a BEGIN block.
562 Internally, the execution of END blocks is now controlled by
563 PL_exit_flags & PERL_EXIT_DESTRUCT_END. This enables the new
564 behaviour for Perl embedders. This will default in 5.10. See
569 Formats now support zero-padded decimal fields.
573 Lvalue subroutines can now return C<undef> in list context. However,
574 the lvalue subroutine feature still remains experimental. [561+]
578 A lost warning "Can't declare ... dereference in my" has been
579 restored (Perl had it earlier but it became lost in later releases.)
583 A new special regular expression variable has been introduced:
584 C<$^N>, which contains the most-recently closed group (submatch).
588 C<no Module;> does not produce an error even if Module does not have an
589 unimport() method. This parallels the behavior of C<use> vis-a-vis
594 The numerical comparison operators return C<undef> if either operand
595 is a NaN. Previously the behaviour was unspecified.
599 C<our> can now have an experimental optional attribute C<unique> that
600 affects how global variables are shared among multiple interpreters,
605 The following builtin functions are now overridable: each(), keys(),
606 pop(), push(), shift(), splice(), unshift(). [561]
610 C<pack() / unpack()> can now group template letters with C<()> and then
611 apply repetition/count modifiers on the groups.
615 C<pack() / unpack()> can now process the Perl internal numeric types:
616 IVs, UVs, NVs-- and also long doubles, if supported by the platform.
617 The template letters are C<j>, C<J>, C<F>, and C<D>.
621 C<pack('U0a*', ...)> can now be used to force a string to UTF8.
625 my __PACKAGE__ $obj now works. [561]
629 POSIX::sleep() now returns the number of I<unslept> seconds
630 (as the POSIX standard says), as opposed to CORE::sleep() which
631 returns the number of slept seconds.
635 The printf() and sprintf() now support parameter reordering using the
636 C<%\d+\$> and C<*\d+\$> syntaxes. For example
638 print "%2\$s %1\$s\n", "foo", "bar";
640 will print "bar foo\n". This feature helps in writing
641 internationalised software, and in general when the order
642 of the parameters can vary.
646 The (\&) prototype now works properly. [561]
650 prototype(\[$@%&]) is now available to implicitly create references
651 (useful for example if you want to emulate the tie() interface).
655 A new command-line option, C<-t> is available. It is the
656 little brother of C<-T>: instead of dying on taint violations,
657 lexical warnings are given. B<This is only meant as a temporary
658 debugging aid while securing the code of old legacy applications.
659 This is not a substitute for -T.>
663 In other taint news, the C<exec LIST> and C<system LIST> have now been
664 considered too risky (think C<exec @ARGV>: it can start any program
665 with any arguments), and now the said forms cause a warning under
666 lexical warnings. You should carefully launder the arguments to
667 guarantee their validity. In future releases of Perl the forms will
668 become fatal errors so consider starting laundering now.
672 Tied hash interfaces are now required to have the EXISTS and DELETE
673 methods (either own or inherited).
677 If tr/// is just counting characters, it doesn't attempt to
682 untie() will now call an UNTIE() hook if it exists. See L<perltie>
687 L<utime> now supports C<utime undef, undef, @files> to change the
688 file timestamps to the current time.
692 The rules for allowing underscores (underbars) in numeric constants
693 have been relaxed and simplified: now you can have an underscore
694 simply B<between digits>.
698 Rather than relying on C's argv[0] (which may not contain a full pathname)
699 where possible $^X is now set by asking the operating system.
700 (eg by reading F</proc/self/exe> on Linux, F</proc/curproc/file> on FreeBSD)
704 A new variable, C<${^TAINT}>, indicates whether taint mode is enabled.
708 You can now override the readline() builtin, and this overrides also
709 the <FILEHANDLE> angle bracket operator.
713 The command-line options -s and -F are now recognized on the shebang
718 Use of the C</c> match modifier without an accompanying C</g> modifier
719 elicits a new warning: C<Use of /c modifier is meaningless without /g>.
721 Use of C</c> in substitutions, even with C</g>, elicits
722 C<Use of /c modifier is meaningless in s///>.
724 Use of C</g> with C<split> elicits C<Use of /g modifier is meaningless
729 Support for the C<CLONE> special subroutine had been added.
730 With ithreads, when a new thread is created, all Perl data is cloned,
731 however non-Perl data cannot be cloned automatically. In C<CLONE> you
732 can do whatever you need to do, like for example handle the cloning of
733 non-Perl data, if necessary. C<CLONE> will be executed once for every
734 package that has it defined or inherited. It will be called in the
735 context of the new thread, so all modifications are made in the new area.
741 =head1 Modules and Pragmata
743 =head2 New Modules and Pragmata
749 C<Attribute::Handlers>, originally by Damian Conway and now maintained
750 by Arthur Bergman, allows a class to define attribute handlers.
753 use Attribute::Handlers;
754 sub Wolf :ATTR(SCALAR) { print "howl!\n" }
756 # later, in some package using or inheriting from MyPack...
758 my MyPack $Fluffy : Wolf; # the attribute handler Wolf will be called
760 Both variables and routines can have attribute handlers. Handlers can
761 be specific to type (SCALAR, ARRAY, HASH, or CODE), or specific to the
762 exact compilation phase (BEGIN, CHECK, INIT, or END).
763 See L<Attribute::Handlers>.
767 C<B::Concise>, by Stephen McCamant, is a new compiler backend for
768 walking the Perl syntax tree, printing concise info about ops.
769 The output is highly customisable. See L<B::Concise>. [561+]
773 The new bignum, bigint, and bigrat pragmas, by Tels, implement
774 transparent bignum support (using the Math::BigInt, Math::BigFloat,
775 and Math::BigRat backends).
779 C<Class::ISA>, by Sean Burke, is a module for reporting the search
780 path for a class's ISA tree. See L<Class::ISA>.
784 C<Cwd> now has a split personality: if possible, an XS extension is
785 used, (this will hopefully be faster, more secure, and more robust)
786 but if not possible, the familiar Perl implementation is used.
790 C<Devel::PPPort>, originally by Kenneth Albanowski and now
791 maintained by Paul Marquess, has been added. It is primarily used
792 by C<h2xs> to enhance portability of XS modules between different
793 versions of Perl. See L<Devel::PPPort>.
797 C<Digest>, frontend module for calculating digests (checksums), from
798 Gisle Aas, has been added. See L<Digest>.
802 C<Digest::MD5> for calculating MD5 digests (checksums) as defined in
803 RFC 1321, from Gisle Aas, has been added. See L<Digest::MD5>.
805 use Digest::MD5 'md5_hex';
807 $digest = md5_hex("Thirsty Camel");
809 print $digest, "\n"; # 01d19d9d2045e005c3f1b80e8b164de1
811 NOTE: the C<MD5> backward compatibility module is deliberately not
812 included since its further use is discouraged.
816 C<Encode>, originally by Nick Ing-Simmons and now maintained by Dan
817 Kogai, provides a mechanism to translate between different character
818 encodings. Support for Unicode, ISO-8859-1, and ASCII are compiled in
819 to the module. Several other encodings (like the rest of the
820 ISO-8859, CP*/Win*, Mac, KOI8-R, three variants EBCDIC, Chinese,
821 Japanese, and Korean encodings) are included and can be loaded at
822 runtime. (For space considerations, the largest Chinese encodings
823 have been separated into their own CPAN module, Encode::HanExtra,
824 which Encode will use if available). See L<Encode>.
826 Any encoding supported by Encode module is also available to the
827 ":encoding()" layer if PerlIO is used.
831 C<Hash::Util> is the interface to the new I<restricted hashes>
832 feature. (Implemented by Jeffrey Friedl, Nick Ing-Simmons, and
833 Michael Schwern.) See L<Hash::Util>.
837 C<I18N::Langinfo> can be used to query locale information.
838 See L<I18N::Langinfo>.
842 C<I18N::LangTags>, by Sean Burke, has functions for dealing with
843 RFC3066-style language tags. See L<I18N::LangTags>.
847 C<ExtUtils::Constant>, by Nicholas Clark, is a new tool for extension
848 writers for generating XS code to import C header constants.
849 See L<ExtUtils::Constant>.
853 C<Filter::Simple>, by Damian Conway, is an easy-to-use frontend to
854 Filter::Util::Call. See L<Filter::Simple>.
860 use Filter::Simple sub {
861 while (my ($from, $to) = splice @_, 0, 2) {
870 use MyFilter qr/red/ => 'green';
872 print "red\n"; # this code is filtered, will print "green\n"
873 print "bored\n"; # this code is filtered, will print "bogreen\n"
877 print "red\n"; # this code is not filtered, will print "red\n"
881 C<File::Temp>, by Tim Jenness, allows one to create temporary files
882 and directories in an easy, portable, and secure way. See L<File::Temp>.
887 C<Filter::Util::Call>, by Paul Marquess, provides you with the
888 framework to write I<source filters> in Perl. For most uses, the
889 frontend Filter::Simple is to be preferred. See L<Filter::Util::Call>.
893 C<if>, by Ilya Zakharevich, is a new pragma for conditional inclusion
898 L<libnet>, by Graham Barr, is a collection of perl5 modules related
899 to network programming. See L<Net::FTP>, L<Net::NNTP>, L<Net::Ping>
900 (not part of libnet, but related), L<Net::POP3>, L<Net::SMTP>,
903 Perl installation leaves libnet unconfigured; use F<libnetcfg>
908 C<List::Util>, by Graham Barr, is a selection of general-utility
909 list subroutines, such as sum(), min(), first(), and shuffle().
914 C<Locale::Constants>, C<Locale::Country>, C<Locale::Currency>
915 C<Locale::Language>, and L<Locale::Script>, by Neil Bowers, have
916 been added. They provide the codes for various locale standards, such
917 as "fr" for France, "usd" for US Dollar, and "ja" for Japanese.
921 $country = code2country('jp'); # $country gets 'Japan'
922 $code = country2code('Norway'); # $code gets 'no'
924 See L<Locale::Constants>, L<Locale::Country>, L<Locale::Currency>,
925 and L<Locale::Language>.
929 C<Locale::Maketext>, by Sean Burke, is a localization framework. See
930 L<Locale::Maketext>, and L<Locale::Maketext::TPJ13>. The latter is an
931 article about software localization, originally published in The Perl
932 Journal #13, and republished here with kind permission.
936 C<Math::BigRat> for big rational numbers, to accompany Math::BigInt and
937 Math::BigFloat, from Tels. See L<Math::BigRat>.
941 C<Memoize> can make your functions faster by trading space for time,
942 from Mark-Jason Dominus. See L<Memoize>.
946 C<MIME::Base64>, by Gisle Aas, allows you to encode data in base64,
947 as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail
952 $encoded = encode_base64('Aladdin:open sesame');
953 $decoded = decode_base64($encoded);
955 print $encoded, "\n"; # "QWxhZGRpbjpvcGVuIHNlc2FtZQ=="
961 C<MIME::QuotedPrint>, by Gisle Aas, allows you to encode data
962 in quoted-printable encoding, as defined in RFC 2045 - I<MIME
963 (Multipurpose Internet Mail Extensions)>.
965 use MIME::QuotedPrint;
967 $encoded = encode_qp("Smiley in Unicode: \x{263a}");
968 $decoded = decode_qp($encoded);
970 print $encoded, "\n"; # "Smiley in Unicode: =263A"
972 MIME::QuotedPrint has been enhanced to provide the basic methods
973 necessary to use it with PerlIO::Via as in :
975 use MIME::QuotedPrint;
976 open($fh,">Via(MIME::QuotedPrint)",$path);
978 See L<MIME::QuotedPrint>.
982 C<NEXT>, by Damian Conway, is a pseudo-class for method redispatch.
987 C<open> is a new pragma for setting the default I/O disciplines
992 C<PerlIO::Scalar>, by Nick Ing-Simmons, provides the implementation
993 of IO to "in memory" Perl scalars as discussed above. It also serves
994 as an example of a loadable PerlIO layer. Other future possibilities
995 include PerlIO::Array and PerlIO::Code. See L<PerlIO::Scalar>.
999 C<PerlIO::Via>, by Nick Ing-Simmons, acts as a PerlIO layer and wraps
1000 PerlIO layer functionality provided by a class (typically implemented
1003 use MIME::QuotedPrint;
1004 open($fh,">Via(MIME::QuotedPrint)",$path);
1006 This will automatically convert everything output to C<$fh>
1007 to Quoted-Printable. See L<PerlIO::Via>.
1011 C<Pod::ParseLink>, by Russ Allbery, has been added,
1012 to parse LZ<><> links in pods as described in the new
1017 C<Pod::Text::Overstrike>, by Joe Smith, has been added.
1018 It converts POD data to formatted overstrike text.
1019 See L<Pod::Text::Overstrike>. [561+]
1023 C<Scalar::Util> is a selection of general-utility scalar subroutines,
1024 such as blessed(), reftype(), and tainted(). See L<Scalar::Util>.
1028 C<sort> is a new pragma for controlling the behaviour of sort().
1032 C<Storable> gives persistence to Perl data structures by allowing the
1033 storage and retrieval of Perl data to and from files in a fast and
1034 compact binary format. Because in effect Storable does serialisation
1035 of Perl data structues, with it you can also clone deep, hierarchical
1036 datastructures. Storable was originally created by Raphael Manfredi,
1037 but it is now maintained by Abhijit Menon-Sen. Storable has been
1038 enhanced to understand the two new hash features, Unicode keys and
1039 restricted hashes. See L<Storable>.
1043 C<Switch>, by Damian Conway, has been added. Just by saying
1047 you have C<switch> and C<case> available in Perl.
1053 case 1 { print "number 1" }
1054 case "a" { print "string a" }
1055 case [1..10,42] { print "number in list" }
1056 case (@array) { print "number in list" }
1057 case /\w+/ { print "pattern" }
1058 case qr/\w+/ { print "pattern" }
1059 case (%hash) { print "entry in hash" }
1060 case (\%hash) { print "entry in hash" }
1061 case (\&sub) { print "arg to subroutine" }
1062 else { print "previous case not true" }
1069 C<Test::More>, by Michael Schwern, is yet another framework for writing
1070 test scripts, more extensive than Test::Simple. See L<Test::More>.
1074 C<Test::Simple>, by Michael Schwern, has basic utilities for writing
1075 tests. See L<Test::Simple>.
1079 C<Text::Balanced>, by Damian Conway, has been added, for extracting
1080 delimited text sequences from strings.
1082 use Text::Balanced 'extract_delimited';
1084 ($a, $b) = extract_delimited("'never say never', he never said", "'", '');
1086 $a will be "'never say never'", $b will be ', he never said'.
1088 In addition to extract_delimited(), there are also extract_bracketed(),
1089 extract_quotelike(), extract_codeblock(), extract_variable(),
1090 extract_tagged(), extract_multiple(), gen_delimited_pat(), and
1091 gen_extract_tagged(). With these, you can implement rather advanced
1092 parsing algorithms. See L<Text::Balanced>.
1096 C<threads>, by Arthur Bergman, is an interface to interpreter threads.
1097 Interpreter threads (ithreads) is the new thread model introduced in
1098 Perl 5.6 but only available as an internal interface for extension
1099 writers (and for Win32 Perl for C<fork()> emulation). See L<threads>,
1100 L<threads::shared>, and L<perlthrtut>.
1104 C<threads::shared>, by Arthur Bergman, allows data sharing for
1105 interpreter threads. In the ithreads model any data sharing between
1106 threads must be explicit, as opposed to the old 5.005 thread model
1107 where data sharing was implicit. See L<threads::shared>.
1111 C<Tie::File>, by Mark-Jason Dominus, associates a Perl array with the
1112 lines of a file. See L<Tie::File>.
1116 C<Tie::Memoize>, by Ilya Zakharevich, provides on-demand loaded hashes.
1117 See L<Tie::Memoize>.
1121 C<Tie::RefHash::Nestable>, by Edward Avis, allows storing hash
1122 references (unlike the standard Tie::RefHash) The module is contained
1123 within Tie::RefHash. See L<Tie::RefHash>.
1127 C<Time::HiRes>, by Douglas E. Wegscheid, provides high resolution
1128 timing (ualarm, usleep, and gettimeofday). See L<Time::HiRes>.
1132 C<Unicode::UCD> offers a querying interface to the Unicode Character
1133 Database. See L<Unicode::UCD>.
1137 C<Unicode::Collate>, by SADAHIRO Tomoyuki, implements the UCA
1138 (Unicode Collation Algorithm) for sorting Unicode strings.
1139 See L<Unicode::Collate>.
1143 C<Unicode::Normalize>, by SADAHIRO Tomoyuki, implements the various
1144 Unicode normalization forms. See L<Unicode::Normalize>.
1148 C<XS::APItest>, by Tim Jenness, is a test extension that exercises XS
1149 APIs. Currently only C<printf()> is tested: how to output various
1150 basic data types from XS.
1154 C<XS::Typemap>, by Tim Jenness, is a test extension that exercises
1155 XS typemaps. Nothing gets installed, but the code is worth studying
1156 for extension writers.
1160 =head2 Updated And Improved Modules and Pragmata
1166 The following independently supported modules have been updated to the
1167 newest versions from CPAN: CGI, CPAN, DB_File, File::Spec, File::Temp,
1168 Getopt::Long, Math::BigFloat, Math::BigInt, the podlators bundle
1169 (Pod::Man, Pod::Text), Pod::LaTeX [561+], Pod::Parser, Storable,
1170 Term::ANSIColor, Test, Text-Tabs+Wrap.
1174 attributes::reftype() now works on tied arguments.
1178 AutoLoader can now be disabled with C<no AutoLoader;>.
1182 B::Deparse has been significantly enhanced by Robin Houston. It can
1183 now deparse almost all of the standard test suite (so that the tests
1184 still succeed). There is a make target "test.deparse" for trying this
1189 Carp now has better interface documentation, and the @CARP_NOT
1190 interface has been added to get optional control over where errors
1191 are reported independently of @ISA, by Ben Tilly.
1195 Class::Struct can now define the classes in compile time.
1199 Class::Struct now assigns the array/hash element if the accessor
1200 is called with an array/hash element as the B<sole> argument.
1204 The return value of Cwd::fastcwd() is now tainted.
1208 Data::Dumper now has an option to sort hashes.
1212 Data::Dumper now has an option to dump code references
1217 DB_File now supports newer Berkeley DB versions, among
1222 Devel::Peek now has an interface for the Perl memory statistics
1223 (this works only if you are using perl's malloc, and if you have
1224 compiled with debugging).
1228 The English module can now be used without the infamous performance
1231 use English '-no_match_vars';
1233 (Assuming, of course, that you don't need the troublesome variables
1234 C<$`>, C<$&>, or C<$'>.) Also, introduced C<@LAST_MATCH_START> and
1235 C<@LAST_MATCH_END> English aliases for C<@-> and C<@+>.
1239 ExtUtils::MakeMaker has been significantly cleaned up and fixed.
1240 The enhanced version has also been backported to earlier releases
1241 of Perl and submitted to CPAN so that the earlier releases can
1246 The arguments of WriteMakefile() in Makefile.PL are now checked
1247 for sanity much more carefully than before. This may cause new
1248 warnings when modules are being insalled. See L<ExtUtils::MakeMaker>
1253 ExtUtils::MakeMaker now uses File::Spec internally, which hopefully
1254 leads to better portability.
1258 Fcntl, Socket, and Sys::Syslog have been rewritten by Nicholas Clark
1259 to use the new-style constant dispatch section (see L<ExtUtils::Constant>).
1260 This means that they will be more robust and hopefully faster.
1264 File::Find now chdir()s correctly when chasing symbolic links. [561]
1268 File::Find now has pre- and post-processing callbacks. It also
1269 correctly changes directories when chasing symbolic links. Callbacks
1270 (naughtily) exiting with "next;" instead of "return;" now work.
1274 File::Find is now (again) reentrant. It also has been made
1279 The warnings issued by File::Find now belong to their own category.
1280 You can enable/disable them with C<use/no warnings 'File::Find';>.
1284 File::Glob::glob() has been renamed to File::Glob::bsd_glob()
1285 because the name clashes with the builtin glob(). The older
1286 name is still available for compatibility, but is deprecated. [561]
1290 File::Glob now supports C<GLOB_LIMIT> constant to limit the size of
1291 the returned list of filenames.
1295 IPC::Open3 now allows the use of numeric file descriptors.
1299 IO::Socket now has an atmark() method, which returns true if the socket
1300 is positioned at the out-of-band mark. The method is also exportable
1301 as a sockatmark() function.
1305 IO::Socket::INET failed to open the specified port if the service name
1306 was not known. It now correctly uses the supplied port number as is. [561]
1310 IO::Socket::INET has support for the ReusePort option (if your
1311 platform supports it). The Reuse option now has an alias, ReuseAddr.
1312 For clarity, you may want to prefer ReuseAddr.
1316 IO::Socket::INET now supports a value of zero for C<LocalPort>
1317 (usually meaning that the operating system will make one up.)
1321 'use lib' now works identically to @INC. Removing directories
1322 with 'no lib' now works.
1326 Math::BigFloat and Math::BigInt have undergone a full rewrite by Tels.
1327 They are now magnitudes faster, and they support various bignum
1328 libraries such as GMP and PARI as their backends.
1332 Math::Complex handles inf, NaN etc., better.
1336 Net::Ping has been considerably enhanced by Rob Brown: multihoming is
1337 now supported, Win32 functionality is better, there is now time
1338 measuring functionality (optionally high-resolution using
1339 Time::HiRes), and there is now "external" protocol which uses
1340 Net::Ping::External module which runs your external ping utility and
1341 parses the output. A version of Net::Ping::External is available in
1344 Note that some of the Net::Ping tests are disabled when running
1345 under the Perl distribution since one cannot assume one or more
1346 of the following: enabled echo port at localhost, full Internet
1347 connectivity, or sympathetic firewalls. You can set the environment
1348 variable PERL_TEST_Net_Ping to "1" (one) before running the Perl test
1349 suite to enable all the Net::Ping tests.
1353 POSIX::sigaction() is now much more flexible and robust.
1354 You can now install coderef handlers, 'DEFAULT', and 'IGNORE'
1355 handlers, installing new handlers was not atomic.
1359 In Safe, C<%INC> is now localised in a Safe compartment so that
1364 In SDBM_File on dosish platforms, some keys went missing because of
1365 lack of support for files with "holes". A workaround for the problem
1370 In Search::Dict one can now have a pre-processing hook for the
1371 lines being searched.
1375 The Shell module now has an OO interface.
1379 In Sys::Syslog there is now a failover mechanism that will go
1380 through alternative connection mechanisms until the message
1381 is successfully logged.
1385 The Test module has been significantly enhanced.
1389 Time::Local::timelocal() does not handle fractional seconds anymore.
1390 The rationale is that neither does localtime(), and timelocal() and
1391 localtime() are supposed to be inverses of each other.
1395 The vars pragma now supports declaring fully qualified variables.
1396 (Something that C<our()> does not and will not support.)
1400 The C<utf8::> name space (as in the pragma) provides various
1401 Perl-callable functions to provide low level access to Perl's
1402 internal Unicode representation. At the moment only length()
1403 has been implemented.
1407 =head1 Utility Changes
1413 Emacs perl mode (emacs/cperl-mode.el) has been updated to version
1418 F<emacs/e2ctags.pl> is now much faster.
1422 C<enc2xs> is a tool for people adding their own encodings to the
1427 C<h2ph> now supports C trigraphs.
1431 C<h2xs> now produces a template README.
1435 C<h2xs> now uses C<Devel::PPPort> for better portability between
1436 different versions of Perl.
1440 C<h2xs> uses the new L<ExtUtils::Constant|ExtUtils::Constant> module
1441 which will affect newly created extensions that define constants.
1442 Since the new code is more correct (if you have two constants where the
1443 first one is a prefix of the second one, the first constant B<never>
1444 got defined), less lossy (it uses integers for integer constant,
1445 as opposed to the old code that used floating point numbers even for
1446 integer constants), and slightly faster, you might want to consider
1447 regenerating your extension code (the new scheme makes regenerating
1448 easy). L<h2xs> now also supports C trigraphs.
1452 C<libnetcfg> has been added to configure libnet.
1456 C<perlbug> is now much more robust. It also sends the bug report to
1457 perl.org, not perl.com.
1461 C<perlcc> has been rewritten and its user interface (that is,
1462 command line) is much more like that of the UNIX C compiler, cc.
1463 (The perlbc tools has been removed. Use C<perlcc -B> instead.)
1464 B<Note that perlcc is still considered very experimental and
1469 C<perlivp> is a new Installation Verification Procedure utility
1470 for running any time after installing Perl.
1474 C<piconv> is an implementation of the character conversion utility
1475 C<iconv>, demonstrating the new Encode module.
1479 C<pod2html> now allows specifying a cache directory.
1483 C<pod2html> now produces XHTML 1.0.
1487 C<pod2html> now understands POD written using different line endings
1488 (PC-like CRLF versus UNIX-like LF versus MacClassic-like CR).
1492 C<s2p> has been completely rewritten in Perl. (It is in fact a full
1493 implementation of sed in Perl: you can use the sed functionality by
1494 using the C<psed> utility.)
1498 C<xsubpp> now understands POD documentation embedded in the *.xs
1503 C<xsubpp> now supports the OUT keyword.
1507 =head1 New Documentation
1513 perl56delta details the changes between the 5.005 release and the
1518 perlclib documents the internal replacements for standard C library
1519 functions. (Interesting only for extension writers and Perl core
1524 perldebtut is a Perl debugging tutorial. [561+]
1528 perlebcdic contains considerations for running Perl on EBCDIC
1533 perlintro is a gentle introduction to Perl.
1537 perliol documents the internals of PerlIO with layers.
1541 perlmodstyle is a style guide for writing modules.
1545 perlnewmod tells about writing and submitting a new module. [561+]
1549 perlpacktut is a pack() tutorial.
1553 perlpod has been rewritten to be clearer and to record the best
1554 practices gathered over the years.
1558 perlpodspec is a more formal specification of the pod format,
1559 mainly of interest for writers of pod applications, not to
1560 people writing in pod.
1564 perlretut is a regular expression tutorial. [561+]
1568 perlrequick is a regular expressions quick-start guide.
1569 Yes, much quicker than perlretut. [561]
1573 perltodo has been updated.
1577 perltootc has been renamed as perltooc (to not to conflict
1578 with perltoot in filesystems restricted to "8.3" names).
1582 perluniintro is an introduction to using Unicode in Perl.
1583 (perlunicode is more of a detailed reference and background
1588 perlutil explains the command line utilities packaged with the Perl
1589 distribution. [561+]
1593 The following platform-specific documents are available before
1594 the installation as README.I<platform>, and after the installation
1597 perlaix perlamiga perlapollo perlbeos perlbs2000
1598 perlce perlcygwin perldgux perldos perlepoc perlfreebsd perlhpux
1599 perlhurd perlirix perlmachten perlmacos perlmint perlmpeix
1600 perlnetware perlos2 perlos390 perlplan9 perlqnx perlsolaris
1601 perltru64 perluts perlvmesa perlvms perlvos perlwin32
1603 These documents usually detail one or more of the following subjects:
1604 configuring, building, testing, installing, and sometimes also using
1605 Perl on the said platform.
1607 Eastern Asian Perl users are now welcomed in their own languages:
1608 README.jp (Japanese), README.ko (Korean), README.cn (simplified
1609 Chinese) and README.tw (traditional Chinese), which are written in
1610 normal pod but encoded in EUC-JP, EUC-KR, EUC-CN and Big5. These
1611 will get installed as
1613 perljp perlko perlcn perltw
1619 The documentation for the POSIX-BC platform is called "BS2000", to avoid
1620 confusion with the Perl POSIX module.
1624 The documentation for the WinCE platform is called perlce (README.ce
1625 in the source code kit), to avoid confusion with the perlwin32
1626 documentation on 8.3-restricted filesystems.
1630 =head1 Performance Enhancements
1636 map() could get pathologically slow when the result list it generates
1637 is larger than the source list. The performance has been improved for
1638 common scenarios. [561]
1642 sort() is also fully reentrant, in the sense that the sort function
1643 can itself call sort(). This did not work reliably in previous
1648 sort() has been changed to use primarily mergesort internally as
1649 opposed to the earlier quicksort. For very small lists this may
1650 result in slightly slower sorting times, but in general the speedup
1651 should be at least 20%. Additional bonuses are that the worst case
1652 behaviour of sort() is now better (in computer science terms it now
1653 runs in time O(N log N), as opposed to quicksort's Theta(N**2)
1654 worst-case run time behaviour), and that sort() is now stable
1655 (meaning that elements with identical keys will stay ordered as they
1656 were before the sort). See the C<sort> pragma for information.
1658 The story in more detail: suppose you want to serve yourself a little
1661 @digits = ( 3,1,4,1,5,9 );
1663 A numerical sort of the digits will yield (1,1,3,4,5,9), as expected.
1664 Which C<1> comes first is hard to know, since one C<1> looks pretty
1665 much like any other. You can regard this as totally trivial,
1666 or somewhat profound. However, if you just want to sort the even
1667 digits ahead of the odd ones, then what will
1669 sort { ($a % 2) <=> ($b % 2) } @digits;
1671 yield? The only even digit, C<4>, will come first. But how about
1672 the odd numbers, which all compare equal? With the quicksort algorithm
1673 used to implement Perl 5.6 and earlier, the order of ties is left up
1674 to the sort. So, as you add more and more digits of Pi, the order
1675 in which the sorted even and odd digits appear will change.
1676 and, for sufficiently large slices of Pi, the quicksort algorithm
1677 in Perl 5.8 won't return the same results even if reinvoked with the
1678 same input. The justification for this rests with quicksort's
1679 worst case behavior. If you run
1681 sort { $a <=> $b } ( 1 .. $N , 1 .. $N );
1683 (something you might approximate if you wanted to merge two sorted
1684 arrays using sort), doubling $N doesn't just double the quicksort time,
1685 it I<quadruples> it. Quicksort has a worst case run time that can
1686 grow like N**2, so-called I<quadratic> behaviour, and it can happen
1687 on patterns that may well arise in normal use. You won't notice this
1688 for small arrays, but you I<will> notice it with larger arrays,
1689 and you may not live long enough for the sort to complete on arrays
1690 of a million elements. So the 5.8 quicksort scrambles large arrays
1691 before sorting them, as a statistical defence against quadratic behaviour.
1692 But that means if you sort the same large array twice, ties may be
1693 broken in different ways.
1695 Because of the unpredictability of tie-breaking order, and the quadratic
1696 worst-case behaviour, quicksort was I<almost> replaced completely with
1697 a stable mergesort. I<Stable> means that ties are broken to preserve
1698 the original order of appearance in the input array. So
1700 sort { ($a % 2) <=> ($b % 2) } (3,1,4,1,5,9);
1702 will yield (4,3,1,1,5,9), guaranteed. The even and odd numbers
1703 appear in the output in the same order they appeared in the input.
1704 Mergesort has worst case O(N log N) behaviour, the best value
1705 attainable. And, ironically, this mergesort does particularly
1706 well where quicksort goes quadratic: mergesort sorts (1..$N, 1..$N)
1707 in O(N) time. But quicksort was rescued at the last moment because
1708 it is faster than mergesort on certain inputs and platforms.
1709 For example, if you really I<don't> care about the order of even
1710 and odd digits, quicksort will run in O(N) time; it's very good
1711 at sorting many repetitions of a small number of distinct elements.
1712 The quicksort divide and conquer strategy works well on platforms
1713 with relatively small, very fast, caches. Eventually, the problem gets
1714 whittled down to one that fits in the cache, from which point it
1715 benefits from the increased memory speed.
1717 Quicksort was rescued by implementing a sort pragma to control aspects
1718 of the sort. The B<stable> subpragma forces stable behaviour,
1719 regardless of algorithm. The B<_quicksort> and B<_mergesort>
1720 subpragmas are heavy-handed ways to select the underlying implementation.
1721 The leading C<_> is a reminder that these subpragmas may not survive
1722 beyond 5.8. More appropriate mechanisms for selecting the implementation
1723 exist, but they wouldn't have arrived in time to save quicksort.
1727 Hashes now use Bob Jenkins "One-at-a-Time" hashing key algorithm
1728 ( http://burtleburtle.net/bob/hash/doobs.html ). This algorithm is
1729 reasonably fast while producing a much better spread of values than
1730 the old hashing algorithm (originally by Chris Torek, later tweaked by
1731 Ilya Zakharevich). Hash values output from the algorithm on a hash of
1732 all 3-char printable ASCII keys comes much closer to passing the
1733 DIEHARD random number generation tests. According to perlbench, this
1734 change has not affected the overall speed of Perl.
1738 unshift() should now be noticeably faster.
1742 =head1 Installation and Configuration Improvements
1744 =head2 Generic Improvements
1750 INSTALL now explains how you can configure Perl to use 64-bit
1751 integers even on non-64-bit platforms.
1755 Policy.sh policy change: if you are reusing a Policy.sh file
1756 (see INSTALL) and you use Configure -Dprefix=/foo/bar and in the old
1757 Policy $prefix eq $siteprefix and $prefix eq $vendorprefix, all of
1758 them will now be changed to the new prefix, /foo/bar. (Previously
1759 only $prefix changed.) If you do not like this new behaviour,
1760 specify prefix, siteprefix, and vendorprefix explicitly.
1764 A new optional location for Perl libraries, otherlibdirs, is available.
1765 It can be used for example for vendor add-ons without disturbing Perl's
1766 own library directories.
1770 In many platforms, the vendor-supplied 'cc' is too stripped-down to
1771 build Perl (basically, 'cc' doesn't do ANSI C). If this seems
1772 to be the case and 'cc' does not seem to be the GNU C compiler
1773 'gcc', an automatic attempt is made to find and use 'gcc' instead.
1777 gcc needs to closely track the operating system release to avoid
1778 build problems. If Configure finds that gcc was built for a different
1779 operating system release than is running, it now gives a clearly visible
1780 warning that there may be trouble ahead.
1784 Since Perl 5.8 is not binary-compatible with previous releases
1785 of Perl, Configure no longer suggests including the 5.005
1790 Configure C<-S> can now run non-interactively. [561]
1794 Configure support for pdp11-style memory models has been removed due
1795 to obsolescence. [561]
1799 configure.gnu now works with options with whitespace in them.
1803 installperl now outputs everything to STDERR.
1807 Because PerlIO is now the default on most platforms, "-perlio" doesn't
1808 get appended to the $Config{archname} (also known as $^O) anymore.
1809 Instead, if you explicitly choose not to use perlio (Configure command
1810 line option -Uuseperlio), you will get "-stdio" appended.
1814 Another change related to the architecture name is that "-64all"
1815 (-Duse64bitall, or "maximally 64-bit") is appended only if your
1816 pointers are 64 bits wide. (To be exact, the use64bitall is ignored.)
1820 In AFS installations, one can configure the root of the AFS to be
1821 somewhere else than the default F</afs> by using the Configure
1822 parameter C<-Dafsroot=/some/where/else>.
1826 APPLLIB_EXP, a lesser-known configuration-time definition, has been
1827 documented. It can be used to prepend site-specific directories
1828 to Perl's default search path (@INC); see INSTALL for information.
1832 The version of Berkeley DB used when the Perl (and, presumably, the
1833 DB_File extension) was built is now available as
1834 C<@Config{qw(db_version_major db_version_minor db_version_patch)}>
1835 from Perl and as C<DB_VERSION_MAJOR_CFG DB_VERSION_MINOR_CFG
1836 DB_VERSION_PATCH_CFG> from C.
1840 Building Berkeley DB3 for compatibility modes for DB, NDBM, and ODBM
1841 has been documented in INSTALL.
1845 If you have CPAN access (either network or a local copy such as a
1846 CD-ROM) you can during specify extra modules to Configure to build and
1847 install with Perl using the -Dextras=... option. See INSTALL for
1852 In addition to config.over, a new override file, config.arch, is
1853 available. This file is supposed to be used by hints file writers
1854 for architecture-wide changes (as opposed to config.over which is
1855 for site-wide changes).
1859 If your file system supports symbolic links, you can build Perl outside
1860 of the source directory by
1862 mkdir /tmp/perl/build/directory
1863 cd /tmp/perl/build/directory
1864 sh /path/to/perl/source/Configure -Dmksymlinks ...
1866 This will create in /tmp/perl/build/directory a tree of symbolic links
1867 pointing to files in /path/to/perl/source. The original files are left
1868 unaffected. After Configure has finished, you can just say
1872 and Perl will be built and tested, all in /tmp/perl/build/directory.
1877 For Perl developers, several new make targets for profiling
1878 and debugging have been added; see L<perlhack>.
1884 Use of the F<gprof> tool to profile Perl has been documented in
1885 L<perlhack>. There is a make target called "perl.gprof" for
1886 generating a gprofiled Perl executable.
1890 If you have GCC 3, there is a make target called "perl.gcov" for
1891 creating a gcoved Perl executable for coverage analysis. See
1896 If you are on IRIX or Tru64 platforms, new profiling/debugging options
1897 have been added; see L<perlhack> for more information about pixie and
1904 Guidelines of how to construct minimal Perl installations have
1905 been added to INSTALL.
1909 The Thread extension is now not built at all under ithreads
1910 (C<Configure -Duseithreads>) because it wouldn't work anyway (the
1911 Thread extension requires being Configured with C<-Duse5005threads>).
1913 But note that the Thread.pm interface is now shared by both
1918 The Gconvert macro ($Config{d_Gconvert}) used by perl for stringifying
1919 floating-point numbers is now more picky about using sprintf %.*g
1920 rules for the conversion. Some platforms that used to use gcvt may
1921 now resort to the slower sprintf.
1925 The obsolete method of making a special (e.g., debugging) flavor
1928 make LIBPERL=libperld.a
1930 has been removed. Use -DDEBUGGING instead.
1934 =head2 New Or Improved Platforms
1936 For the list of platforms known to support Perl,
1937 see L<perlport/"Supported Platforms">.
1943 AIX dynamic loading should be now better supported.
1947 AIX should now work better with gcc, threads, and 64-bitness. Also the
1948 long doubles support in AIX should be better now. See L<perlaix>.
1952 AtheOS ( http://www.atheos.cx/ ) is a new platform.
1956 BeOS has been reclaimed.
1960 The DG/UX platform now supports 5.005-style threads.
1965 The DYNIX/ptx platform (a.k.a. dynixptx) is supported at or near
1970 EBCDIC platforms (z/OS (also known as OS/390), POSIX-BC, and VM/ESA)
1971 have been regained. Many test suite tests still fail and the
1972 co-existence of Unicode and EBCDIC isn't quite settled, but the
1973 situation is much better than with Perl 5.6. See L<perlos390>,
1974 L<perlbs2000> (for POSIX-BC), and L<perlvmesa> for more information.
1978 Building perl with -Duseithreads or -Duse5005threads now works under
1979 HP-UX 10.20 (previously it only worked under 10.30 or later). You will
1980 need a thread library package installed. See README.hpux. [561]
1984 Mac OS Classic is now supported in the mainstream source package
1985 (MacPerl has of course been available since perl 5.004 but now the
1986 source code bases of standard Perl and MacPerl have been synchronised)
1991 Mac OS X (or Darwin) should now be able to build Perl even on HFS+
1992 filesystems. (The case-insensitivity used to confuse the Perl build
1997 NCR MP-RAS is now supported. [561]
2001 All the NetBSD specific patches (except for the installation
2002 specific ones) have been merged back to the main distribution.
2006 NetWare from Novell is now supported. See L<perlnetware>.
2010 NonStop-UX is now supported. [561]
2014 NEC SUPER-UX is now supported.
2018 All the OpenBSD specific patches (except for the installation
2019 specific ones) have been merged back to the main distribution.
2023 Perl has been tested with the GNU pth userlevel thread package
2024 ( http://www.gnu.org/software/pth/pth.html ). All thread tests
2025 of Perl now work, but not without adding some yield()s to the tests,
2026 so while pth (and other userlevel thread implementations) can be
2027 considered to be "working" with Perl ithreads, keep in mind the
2028 possible non-preemptability of the underlying thread implementation.
2032 Stratus VOS is now supported using Perl's native build method
2033 (Configure). This is the recommended method to build Perl on
2034 VOS. The older methods, which build miniperl, are still
2035 available. See L<perlvos>. [561+]
2039 The Amdahl UTS UNIX mainframe platform is now supported. [561]
2043 WinCE is now supported. See L<perlce>.
2047 z/OS (formerly known as OS/390, formerly known as MVS OE) now has
2048 support for dynamic loading. This is not selected by default,
2049 however, you must specify -Dusedl in the arguments of Configure. [561]
2053 =head1 Selected Bug Fixes
2055 Numerous memory leaks and uninitialized memory accesses have been
2056 hunted down. Most importantly, anonymous subs used to leak quite
2063 The autouse pragma didn't work for Multi::Part::Function::Names.
2067 caller() could cause core dumps in certain situations. Carp was
2068 sometimes affected by this problem. In particular, caller() now
2069 returns a subroutine name of C<(unknown)> for subroutines that have
2070 been removed from the symbol table.
2074 chop(@list) in list context returned the characters chopped in
2075 reverse order. This has been reversed to be in the right order. [561]
2079 Configure no longer includes the DBM libraries (dbm, gdbm, db, ndbm)
2080 when building the Perl binary. The only exception to this is SunOS 4.x,
2081 which needs them. [561]
2085 The behaviour of non-decimal but numeric string constants such as
2086 "0x23" was platform-dependent: in some platforms that was seen as 35,
2087 in some as 0, in some as a floating point number (don't ask). This
2088 was caused by Perl's using the operating system libraries in a situation
2089 where the result of the string to number conversion is undefined: now
2090 Perl consistently handles such strings as zero in numeric contexts.
2094 The order of DESTROYs has been made more predictable.
2098 Perl 5.6.0 could emit spurious warnings about redefinition of
2099 dl_error() when statically building extensions into perl.
2100 This has been corrected. [561]
2104 L<dprofpp> -R didn't work.
2108 C<*foo{FORMAT}> now works.
2112 Infinity is now recognized as a number.
2116 UNIVERSAL::isa no longer caches methods incorrectly. (This broke
2117 the Tk extension with 5.6.0.) [561]
2121 Lexicals I: lexicals outside an eval "" weren't resolved
2122 correctly inside a subroutine definition inside the eval "" if they
2123 were not already referenced in the top level of the eval""ed code.
2127 Lexicals II: lexicals leaked at file scope into subroutines that
2128 were declared before the lexicals.
2132 Lexical warnings now propagating correctly between scopes
2133 and into C<eval "...">.
2137 C<use warnings qw(FATAL all)> did not work as intended. This has been
2142 warnings::enabled() now reports the state of $^W correctly if the caller
2143 isn't using lexical warnings. [561]
2147 Line renumbering with eval and C<#line> now works. [561]
2151 Fixed numerous memory leaks, especially in eval "".
2155 Localised tied variables no longer leak memory
2158 tie my %tied_hash => 'Tie::StdHash';
2162 # Used to leak memory every time local() was called;
2163 # in a loop, this added up.
2164 local($tied_hash{Foo}) = 1;
2168 Localised hash elements (and %ENV) are correctly unlocalised to not
2169 exist, if they didn't before they were localised.
2173 tie my %tied_hash => 'Tie::StdHash';
2177 # Nothing has set the FOO element so far
2179 { local $tied_hash{FOO} = 'Bar' }
2181 # This used to print, but not now.
2182 print "exists!\n" if exists $tied_hash{FOO};
2184 As a side effect of this fix, tied hash interfaces B<must> define
2185 the EXISTS and DELETE methods.
2189 mkdir() now ignores trailing slashes in the directory name,
2190 as mandated by POSIX.
2194 Some versions of glibc have a broken modfl(). This affects builds
2195 with C<-Duselongdouble>. This version of Perl detects this brokenness
2196 and has a workaround for it. The glibc release 2.2.2 is known to have
2197 fixed the modfl() bug.
2201 Modulus of unsigned numbers now works (4063328477 % 65535 used to
2202 return 27406, instead of 27047). [561]
2206 Some "not a number" warnings introduced in 5.6.0 eliminated to be
2207 more compatible with 5.005. Infinity is now recognised as a number. [561]
2211 Numeric conversions did not recognize changes in the string value
2212 properly in certain circumstances. [561]
2216 Attributes (such as :shared) didn't work with our().
2220 our() variables will not cause bogus "Variable will not stay shared"
2225 "our" variables of the same name declared in two sibling blocks
2226 resulted in bogus warnings about "redeclaration" of the variables.
2227 The problem has been corrected. [561]
2231 pack "Z" now correctly terminates the string with "\0".
2235 Fix password routines which in some shadow password platforms
2236 (e.g. HP-UX) caused getpwent() to return every other entry.
2240 The PERL5OPT environment variable (for passing command line arguments
2241 to Perl) didn't work for more than a single group of options. [561]
2245 PERL5OPT with embedded spaces didn't work.
2249 printf() no longer resets the numeric locale to "C".
2253 C<qw(a\\b)> now parses correctly as C<'a\\b'>: that is, as three
2254 characters, not four. [561]
2258 pos() did not return the correct value within s///ge in earlier
2259 versions. This is now handled correctly. [561]
2263 Printing quads (64-bit integers) with printf/sprintf now works
2264 without the q L ll prefixes (assuming you are on a quad-capable platform).
2268 Regular expressions on references and overloaded scalars now work. [561+]
2272 Right-hand side magic (GMAGIC) could in many cases such as string
2273 concatenation be invoked too many times.
2277 scalar() now forces scalar context even when used in void context.
2281 SOCKS support is now much more robust.
2285 sort() arguments are now compiled in the right wantarray context
2286 (they were accidentally using the context of the sort() itself).
2287 The comparison block is now run in scalar context, and the arguments
2288 to be sorted are always provided list context. [561]
2292 Changed the POSIX character class C<[[:space:]]> to include the (very
2293 rarely used) vertical tab character. Added a new POSIX-ish character
2294 class C<[[:blank:]]> which stands for horizontal whitespace
2295 (currently, the space and the tab).
2299 The tainting behaviour of sprintf() has been rationalized. It does
2300 not taint the result of floating point formats anymore, making the
2301 behaviour consistent with that of string interpolation. [561]
2305 Some cases of inconsistent taint propagation (such as within hash
2306 values) have been fixed.
2310 The RE engine found in Perl 5.6.0 accidentally pessimised certain kinds
2311 of simple pattern matches. These are now handled better. [561]
2315 Regular expression debug output (whether through C<use re 'debug'>
2316 or via C<-Dr>) now looks better. [561]
2320 Multi-line matches like C<"a\nxb\n" =~ /(?!\A)x/m> were flawed. The
2321 bug has been fixed. [561]
2325 Use of $& could trigger a core dump under some situations. This
2326 is now avoided. [561]
2330 The regular expression captured submatches ($1, $2, ...) are now
2331 more consistently unset if the match fails, instead of leaving false
2332 data lying around in them. [561]
2336 readline() on files opened in "slurp" mode could return an extra
2337 "" (blank line) at the end in certain situations. This has been
2342 Autovivification of symbolic references of special variables described
2343 in L<perlvar> (as in C<${$num}>) was accidentally disabled. This works
2348 Sys::Syslog ignored the C<LOG_AUTH> constant.
2352 $AUTOLOAD, sort(), lock(), and spawning subprocesses
2353 in multiple threads simultaneously are now thread-safe.
2357 Tie::Array's SPLICE method was broken.
2361 Allow a read-only string on the left-hand side of a non-modifying tr///.
2365 If C<STDERR> is tied, warnings caused by C<warn> and C<die> now
2366 correctly pass to it.
2370 Several Unicode fixes.
2376 BOMs (byte order marks) at the beginning of Perl files
2377 (scripts, modules) should now be transparently skipped.
2378 UTF-16 and UCS-2 encoded Perl files should now be read correctly.
2382 The character tables have been updated to Unicode 3.2.0.
2386 Comparing with utf8 data does not magically upgrade non-utf8 data
2387 into utf8. (This was a problem for example if you were mixing data
2388 from I/O and Unicode data: your output might have got magically encoded
2393 Generating illegal Unicode code points such as U+FFFE, or the UTF-16
2394 surrogates, now also generates an optional warning.
2398 C<IsAlnum>, C<IsAlpha>, and C<IsWord> now match titlecase.
2402 Concatenation with the C<.> operator or via variable interpolation,
2403 C<eq>, C<substr>, C<reverse>, C<quotemeta>, the C<x> operator,
2404 substitution with C<s///>, single-quoted UTF8, should now work.
2408 The C<tr///> operator now works. Note that the C<tr///CU>
2409 functionality has been removed (but see pack('U0', ...)).
2413 C<eval "v200"> now works.
2417 Perl 5.6.0 parsed m/\x{ab}/ incorrectly, leading to spurious warnings.
2418 This has been corrected. [561]
2422 Zero entries were missing from the Unicode classes such as C<IsDigit>.
2428 Large unsigned numbers (those above 2**31) could sometimes lose their
2429 unsignedness, causing bogus results in arithmetic operations. [561]
2433 The Perl parser has been stress tested using both random input and
2434 Markov chain input and the few found crashes and lockups have been
2439 =head2 Platform Specific Changes and Fixes
2447 Perl now works on post-4.0 BSD/OSes.
2453 Setting C<$0> now works (as much as possible; see L<perlvar> for details).
2459 Numerous updates; currently synchronised with Cygwin 1.3.10.
2463 Previously DYNIX/ptx had problems in its Configure probe for non-blocking I/O.
2469 EPOC now better supported. See README.epoc. [561]
2475 Perl now works on post-3.0 FreeBSDs.
2481 README.hpux updated; C<Configure -Duse64bitall> now works;
2482 now uses HP-UX malloc instead of Perl malloc.
2488 Numerous compilation flag and hint enhancements; accidental mixing
2489 of 32-bit and 64-bit libraries (a doomed attempt) made much harder.
2499 Long doubles should now work (see INSTALL). [561]
2503 Linux previously had problems related to sockaddrlen when using
2504 accept(), recvfrom() (in Perl: recv()), getpeername(), and
2513 Compilation of the standard Perl distribution in Mac OS Classic should
2514 now work if you have the Metrowerks development environment and the
2515 missing Mac-specific toolkit bits. Contact the macperl mailing list
2522 MPE/iX update after Perl 5.6.0. See README.mpeix. [561]
2526 NetBSD/threads: try installing the GNU pth (should be in the
2527 packages collection, or http://www.gnu.org/software/pth/),
2528 and Configure with -Duseithreads.
2534 Perl now works on NetBSD/sparc.
2540 Now works with usethreads (see INSTALL). [561]
2546 64-bitness using the Sun Workshop compiler now works.
2552 The native build method requires at least VOS Release 14.5.0
2553 and GNU C++/GNU Tools 2.0.1 or later. The Perl pack function
2554 now maps overflowed values to +infinity and underflowed values
2559 Tru64 (aka Digital UNIX, aka DEC OSF/1)
2561 The operating system version letter now recorded in $Config{osvers}.
2562 Allow compiling with gcc (previously explicitly forbidden). Compiling
2563 with gcc still not recommended because buggy code results, even with
2570 Fixed various alignment problems that lead into core dumps either
2571 during build or later; no longer dies on math errors at runtime;
2572 now using full quad integers (64 bits), previously was using
2573 only 46 bit integers for speed.
2579 See L</"Socket Extension Dynamic in VMS"> and L</"IEEE-format Floating Point
2580 Default on OpenVMS Alpha"> for important changes not otherwise listed here.
2582 chdir() now works better despite a CRT bug; now works with MULTIPLICITY
2583 (see INSTALL); now works with Perl's malloc.
2585 The tainting of C<%ENV> elements via C<keys> or C<values> was previously
2586 unimplemented. It now works as documented.
2588 The C<waitpid> emulation has been improved. The worst bug (now fixed)
2589 was that a pid of -1 would cause a wildcard search of all processes on
2592 POSIX-style signals are now emulated much better on VMS versions prior
2595 The C<system> function and backticks operator have improved
2596 functionality and better error handling. [561]
2598 File access tests now use current process privileges rather than the
2599 user's default privileges, which could sometimes result in a mismatch
2600 between reported access and actual access. This improvement is only
2601 available on VMS v6.0 and later.
2603 There is a new C<kill> implementation based on C<sys$sigprc> that allows
2604 older VMS systems (pre-7.0) to use C<kill> to send signals rather than
2605 simply force exit. This implementation also allows later systems to
2606 call C<kill> from within a signal handler.
2608 Iterative logical name translations are now limited to 10 iterations in
2609 imitation of SHOW LOGICAL and other OpenVMS facilities.
2619 Signal handling now works better than it used to. It is now implemented
2620 using a Windows message loop, and is therefore less prone to random
2625 fork() emulation is now more robust, but still continues to have a few
2626 esoteric bugs and caveats. See L<perlfork> for details. [561+]
2630 A failed (pseudo)fork now returns undef and sets errno to EAGAIN. [561]
2634 The following modules now work on Windows:
2636 ExtUtils::Embed [561]
2643 IO::File::new_tmpfile() is no longer limited to 32767 invocations
2648 Better chdir() return value for a non-existent directory.
2652 Compiling perl using the 64-bit Platform SDK tools is now supported.
2656 The Win32::SetChildShowWindow() builtin can be used to control the
2657 visibility of windows created by child processes. See L<Win32> for
2662 Non-blocking waits for child processes (or pseudo-processes) are
2663 supported via C<waitpid($pid, &POSIX::WNOHANG)>.
2667 The behavior of system() with multiple arguments has been rationalized.
2668 Each unquoted argument will be automatically quoted to protect whitespace,
2669 and any existing whitespace in the arguments will be preserved. This
2670 improves the portability of system(@args) by avoiding the need for
2671 Windows C<cmd> shell specific quoting in perl programs.
2673 Note that this means that some scripts that may have relied on earlier
2674 buggy behavior may no longer work correctly. For example,
2675 C<system("nmake /nologo", @args)> will now attempt to run the file
2676 C<nmake /nologo> and will fail when such a file isn't found.
2677 On the other hand, perl will now execute code such as
2678 C<system("c:/Program Files/MyApp/foo.exe", @args)> correctly.
2682 The perl header files no longer suppress common warnings from the
2683 Microsoft Visual C++ compiler. This means that additional warnings may
2684 now show up when compiling XS code.
2688 Borland C++ v5.5 is now a supported compiler that can build Perl.
2689 However, the generated binaries continue to be incompatible with those
2690 generated by the other supported compilers (GCC and Visual C++). [561]
2694 Duping socket handles with open(F, ">&MYSOCK") now works under Windows 9x.
2699 Current directory entries in %ENV are now correctly propagated to child
2704 New %ENV entries now propagate to subprocesses. [561]
2708 Win32::GetCwd() correctly returns C:\ instead of C: when at the drive root.
2709 Other bugs in chdir() and Cwd::cwd() have also been fixed. [561]
2713 The makefiles now default to the features enabled in ActiveState ActivePerl
2714 (a popular Win32 binary distribution). [561]
2718 HTML files will now be installed in c:\perl\html instead of
2719 c:\perl\lib\pod\html
2723 REG_EXPAND_SZ keys are now allowed in registry settings used by perl. [561]
2727 Can now send() from all threads, not just the first one. [561]
2731 ExtUtils::MakeMaker now uses $ENV{LIB} to search for libraries. [561]
2735 Less stack reserved per thread so that more threads can run
2736 concurrently. (Still 16M per thread.) [561]
2740 C<< File::Spec->tmpdir() >> now prefers C:/temp over /tmp
2741 (works better when perl is running as service).
2745 Better UNC path handling under ithreads. [561]
2749 wait(), waitpid(), and backticks now return the correct exit status
2750 under Windows 9x. [561]
2754 A socket handle leak in accept() has been fixed. [561]
2760 =head1 New or Changed Diagnostics
2766 The lexical warnings category "deprecated" is no longer a sub-category
2767 of the "syntax" category. It is now a top-level category in its own
2772 All regular expression compilation error messages are now hopefully
2773 easier to understand both because the error message now comes before
2774 the failed regex and because the point of failure is now clearly
2775 marked by a C<E<lt>-- HERE> marker.
2779 The various "opened only for", "on closed", "never opened" warnings
2780 drop the C<main::> prefix for filehandles in the C<main> package,
2781 for example C<STDIN> instead of C<main::STDIN>.
2785 The "Unrecognized escape" warning has been extended to include C<\8>,
2786 C<\9>, and C<\_>. There is no need to escape any of the C<\w> characters.
2790 Two new debugging options have been added: if you have compiled your
2791 Perl with debugging, you can use the -DT [561] and -DR options to trace
2792 tokenising and to add reference counts to displaying variables,
2797 Several debugger fixes: exit code now reflects the script exit code,
2798 condition C<"0"> now treated correctly, the C<d> command now checks
2799 line number, C<$.> no longer gets corrupted, and all debugger output
2800 now goes correctly to the socket if RemotePort is set. [561]
2804 The debugger (perl5db.pl) has been modified to present a more
2805 consistent commands interface, via (CommandSet=580). perl5db.t was
2806 also added to test the changes, and as a placeholder for further tests.
2812 The debugger has a new C<dumpDepth> option to control the maximum
2813 depth to which nested structures are dumped. The C<x> command has
2814 been extended so that C<x N EXPR> dumps out the value of I<EXPR> to a
2815 depth of at most I<N> levels.
2819 The debugger can now show lexical variables if you have the CPAN
2820 module PadWalker installed.
2824 If an attempt to use a (non-blessed) reference as an array index
2825 is made, a warning is given.
2829 C<push @a;> and C<unshift @a;> (with no values to push or unshift)
2830 now give a warning. This may be a problem for generated and evaled
2835 If you try to L<perlfunc/pack> a number less than 0 or larger than 255
2836 using the C<"C"> format you will get an optional warning. Similarly
2837 for the C<"c"> format and a number less than -128 or more than 127.
2841 Certain regex modifiers such as C<(?o)> make sense only if applied to
2842 the entire regex. You will get an optional warning if you try to do
2847 Using arrays or hashes as references (e.g. C<< %foo->{bar} >>
2848 has been deprecated for a while. Now you will get an optional warning.
2852 Using C<sort> in scalar context now issues an optional warning.
2853 This didn't do anything useful, as the sort was not performed.
2857 =head1 Changed Internals
2863 perlapi.pod (a companion to perlguts) now attempts to document the
2868 You can now build a really minimal perl called microperl.
2869 Building microperl does not require even running Configure;
2870 C<make -f Makefile.micro> should be enough. Beware: microperl makes
2871 many assumptions, some of which may be too bold; the resulting
2872 executable may crash or otherwise misbehave in wondrous ways.
2873 For careful hackers only.
2877 Added rsignal(), whichsig(), do_join(), op_clear, op_null,
2878 ptr_table_clear(), ptr_table_free(), sv_setref_uv(), and several UTF-8
2879 interfaces to the publicised API. For the full list of the available
2880 APIs see L<perlapi>.
2884 Made possible to propagate customised exceptions via croak()ing.
2888 Now xsubs can have attributes just like subs. (Well, at least the
2889 built-in attributes.)
2893 dTHR and djSP have been obsoleted; the former removed (because it's
2894 a no-op) and the latter replaced with dSP.
2898 PERL_OBJECT has been completely removed.
2902 The MAGIC constants (e.g. C<'P'>) have been macrofied
2903 (e.g. C<PERL_MAGIC_TIED>) for better source code readability
2904 and maintainability.
2908 The regex compiler now maintains a structure that identifies nodes in
2909 the compiled bytecode with the corresponding syntactic features of the
2910 original regex expression. The information is attached to the new
2911 C<offsets> member of the C<struct regexp>. See L<perldebguts> for more
2912 complete information.
2916 The C code has been made much more C<gcc -Wall> clean. Some warning
2917 messages still remain in some platforms, so if you are compiling with
2918 gcc you may see some warnings about dubious practices. The warnings
2919 are being worked on.
2923 F<perly.c>, F<sv.c>, and F<sv.h> have now been extensively commented.
2927 Documentation on how to use the Perl source repository has been added
2928 to F<Porting/repository.pod>.
2932 There are now several profiling make targets.
2936 =head1 Security Vulnerability Closed [561]
2938 (This change was already made in 5.7.0 but bears repeating here.)
2939 (5.7.0 came out before 5.6.1: the development branch 5.7 released
2940 sooner than the maintenance branch 5.6)
2942 A potential security vulnerability in the optional suidperl component
2943 of Perl was identified in August 2000. suidperl is neither built nor
2944 installed by default. As of November 2001 the only known vulnerable
2945 platform is Linux, most likely all Linux distributions. CERT and
2946 various vendors and distributors have been alerted about the vulnerability.
2947 See http://www.cpan.org/src/5.0/sperl-2000-08-05/sperl-2000-08-05.txt
2948 for more information.
2950 The problem was caused by Perl trying to report a suspected security
2951 exploit attempt using an external program, /bin/mail. On Linux
2952 platforms the /bin/mail program had an undocumented feature which
2953 when combined with suidperl gave access to a root shell, resulting in
2954 a serious compromise instead of reporting the exploit attempt. If you
2955 don't have /bin/mail, or if you have 'safe setuid scripts', or if
2956 suidperl is not installed, you are safe.
2958 The exploit attempt reporting feature has been completely removed from
2959 Perl 5.8.0 (and the maintenance release 5.6.1, and it was removed also
2960 from all the Perl 5.7 releases), so that particular vulnerability
2961 isn't there anymore. However, further security vulnerabilities are,
2962 unfortunately, always possible. The suidperl functionality is most
2963 probably going to be removed in Perl 5.10. In any case, suidperl
2964 should only be used by security experts who know exactly what they are
2965 doing and why they are using suidperl instead of some other solution
2966 such as sudo ( see http://www.courtesan.com/sudo/ ).
2970 Several new tests have been added, especially for the F<lib> and F<ext>
2971 subsections. There are now about 65 000 individual tests (spread over
2972 about 700 test scripts), in the regression suite (5.6.1 has about
2973 11700 tests, in 258 test scripts) Many of the new tests are of course
2974 introduced by the new modules, but still in general Perl is now more
2977 Because of the large number of tests, running the regression suite
2978 will take considerably longer time than it used to: expect the suite
2979 to take up to 4-5 times longer to run than in perl 5.6. On a really
2980 fast machine you can hope to finish the suite in about 6-8 minutes
2983 The tests are now reported in a different order than in earlier Perls.
2984 (This happens because the test scripts from under t/lib have been moved
2985 to be closer to the library/extension they are testing.)
2987 =head1 Known Problems
2995 If using the AIX native make command, instead of just "make" issue
2996 "make all". In some setups the former has been known to spuriously
2997 also try to run "make install". Alternatively, you may want to use
3002 In AIX 4.2, Perl extensions that use C++ functions that use statics
3003 may have problems in that the statics are not getting initialized.
3004 In newer AIX releases, this has been solved by linking Perl with
3005 the libC_r library, but unfortunately in AIX 4.2 the said library
3006 has an obscure bug where the various functions related to time
3007 (such as time() and gettimeofday()) return broken values, and
3008 therefore in AIX 4.2 Perl is not linked against libC_r.
3012 vac 5.0.0.0 May Produce Buggy Code For Perl
3014 The AIX C compiler vac version 5.0.0.0 may produce buggy code,
3015 resulting in a few random tests failing when run as part of "make
3016 test", but when the failing tests are run by hand, they succeed.
3017 We suggest upgrading to at least vac version 5.0.1.0, that has been
3018 known to compile Perl correctly. "lslpp -L|grep vac.C" will tell
3019 you the vac version. See README.aix.
3023 If building threaded Perl, you may get compilation warning from pp_sys.c:
3025 "pp_sys.c", line 4651.39: 1506-280 (W) Function argument assignment between types "unsigned char*" and "const void*" is not allowed.
3027 This is harmless; it is caused by the getnetbyaddr() and getnetbyaddr_r()
3028 having slightly different types for their first argument.
3032 =head2 Alpha systems with old gccs fail several tests
3034 If you see op/pack, op/pat, op/regexp, or ext/Storable tests failing
3035 in a Linux/alpha or *BSD/Alpha, it's probably time to upgrade your gcc.
3036 gccs prior to 2.95.3 are definitely not good enough, and gcc 3.1 may
3037 be even better. (RedHat Linux/alpha with gcc 3.1 reported no problems,
3038 as did Linux 2.4.18 with gcc 2.95.4.) (In Tru64, it is preferable to
3039 use the bundled C compiler.)
3043 Perl 5.8.0 doesn't build in AmigaOS. It broke at some point during
3044 the ithreads work and we could not find Amiga experts to unbreak the
3045 problems. Perl 5.6.1 still works for AmigaOS (as does the the 5.7.2
3046 development release).
3050 The following tests fail on 5.8.0 Perl in BeOS Personal 5.03:
3052 t/op/lfs............................FAILED at test 17
3053 t/op/magic..........................FAILED at test 24
3054 ext/POSIX/t/sigaction...............FAILED at test 13
3055 ext/POSIX/t/waitpid.................FAILED at test 1
3057 See L<perlbeos> (README.beos) for more details.
3059 =head2 Cygwin "unable to remap"
3061 For example when building the Tk extension for Cygwin,
3062 you may get an error message saying "unable to remap".
3063 This is known problem with Cygwin, and a workaround is
3064 detailed in here: http://sources.redhat.com/ml/cygwin/2001-12/msg00894.html
3066 =head2 ext/threads/t/libc
3068 If this test fails, it indicates that your libc (C library) is not
3069 threadsafe. This particular test stress tests the localtime() call to
3070 find out whether it is threadsafe. See L<perlthrtut> for more information.
3072 =head2 FreeBSD built with ithreads coredumps reading large directories
3074 This is a known bug in FreeBSD's readdir_r() (see L<perlfreebsd>
3075 (README.freebsd)), which hopefully will be fixed in FreeBSD 4.6.
3077 =head2 FreeBSD Failing locale Test 117 For ISO 8859-15 Locales
3079 The ISO 8859-15 locales may fail the locale test 117 in FreeBSD.
3080 This is caused by the characters \xFF (y with diaeresis) and \xBE
3081 (Y with diaeresis) not behaving correctly when being matched
3082 case-insensitively. Apparently this problem has been fixed in
3083 the latest FreeBSD releases.
3084 ( http://www.freebsd.org/cgi/query-pr.cgi?pr=34308 )
3086 =head2 IRIX fails ext/List/Util/t/shuffle.t
3088 IRIX with MIPSpro 7.3.1.3m compiler may fail the said List::Util test
3089 by dumping core. This seems to be a compiler error since if compiled
3090 with gcc no core dump ensues, and no failures on the said test on any
3093 =head2 Modifying $_ Inside for(..)
3097 works without complaint. It shouldn't. (You should be able to
3098 modify only lvalue elements inside the loops.) You can see the
3099 correct behaviour by replacing the 1..5 with 1, 2, 3, 4, 5.
3101 =head2 mod_perl 1.26 Doesn't Build With Threaded Perl
3103 Use mod_perl 1.27 or higher.
3105 =head2 lib/ftmp-security tests warn 'system possibly insecure'
3107 Don't panic. Read the 'make test' section of INSTALL instead.
3109 =head2 HP-UX lib/posix Subtest 9 Fails When LP64-Configured
3111 If perl is configured with -Duse64bitall, the successful result of the
3112 subtest 10 of lib/posix may arrive before the successful result of the
3113 subtest 9, which confuses the test harness so much that it thinks the
3116 =head2 Linux with glibc 2.2.5 fails t/op/int subtest #6 with -Duse64bitint
3118 This is a known bug in the glibc 2.2.5 with long long integers.
3119 ( http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=65612 )
3121 =head2 Linux With Sfio Fails op/misc Test 48
3125 =head2 libwww-perl (LWP) fails base/date #51
3127 Use libwww-perl 5.65 or later.
3131 Please remember to set your environment variable LC_ALL to "C"
3132 (setenv LC_ALL C) before running "make test" to avoid a lot of
3133 warnings about the broken locales of Mac OS X.
3135 The following tests are known to fail in Mac OS X 10.1.5 because of
3136 buggy (old) implementations of Berkeley DB included in Mac OS X:
3138 Failed Test Stat Wstat Total Fail Failed List of Failed
3139 -------------------------------------------------------------------------
3140 ../ext/DB_File/t/db-btree.t 0 11 ?? ?? % ??
3141 ../ext/DB_File/t/db-recno.t 149 3 2.01% 61 63 65
3143 If you are building on a UFS partition, you will also probably see
3144 t/op/stat.t subtest #9 fail. This is caused by Darwin's UFS not
3145 supporting inode change time.
3147 Also the ext/POSIX/t/posix.t subtest #10 fails but it is skipped for
3148 now because the failure is Apple's fault, not Perl's (blocked signals
3151 If you Configure with ithreads, ext/threads/t/libc.t will fail. Again,
3152 this is not Perl's fault-- the libc of Mac OS X is not threadsafe
3153 (in this particular test, the localtime() call is found to be
3156 =head2 OS/2 Test Failures
3158 The following tests are known to fail on OS/2 (for clarity
3159 only the failures are shown, not the full error messages):
3161 t/io/utf8............................FAILED at test 19
3162 t/op/grent...........................FAILED at test 2
3163 t/op/pwent...........................FAILED at test 1
3164 t/lib/os2_base.......................FAILED at test 13
3165 t/lib/os2_process....................FAILED at test 10
3166 t/lib/os2_process_kid................FAILED at test 10
3167 t/lib/rx_cmprt.......................FAILED at test 16
3168 ext/DB_File/t/db-btree...............FAILED at test 0
3169 ext/DB_File/t/db-hash................FAILED at test 0
3170 ext/DB_File/t/db-recno...............FAILED at test 0
3171 lib/ExtUtils/t/basic.................FAILED at test 14
3172 lib/ExtUtils/t/Constant..............FAILED at test 4
3173 lib/Memoize/t/errors.................FAILED at test 4
3175 =head2 op/sprintf tests 91, 129, and 130
3177 The op/sprintf tests 91, 129, and 130 are known to fail on some platforms.
3178 Examples include any platform using sfio, and Compaq/Tandem's NonStop-UX.
3180 Test 91 is known to fail on QNX6 (nto), because C<sprintf '%e',0>
3181 incorrectly produces C<0.000000e+0> instead of C<0.000000e+00>.
3183 For tests 129 and 130, the failing platforms do not comply with
3184 the ANSI C Standard: lines 19ff on page 134 of ANSI X3.159 1989, to
3185 be exact. (They produce something other than "1" and "-1" when
3186 formatting 0.6 and -0.6 using the printf format "%.0f"; most often,
3187 they produce "0" and "-0".)
3191 In case you are still using Solaris 2.5 (aka SunOS 5.5), you may
3192 experience failures (the test core dumping) in lib/locale.t.
3193 The suggested cure is to upgrade your Solaris.
3195 =head2 Solaris x86 Fails Tests With -Duse64bitint
3197 The following tests are known to fail in Solaris x86 with Perl
3198 configured to use 64 bit integers:
3200 ext/Data/Dumper/t/dumper.............FAILED at test 268
3201 ext/Devel/Peek/Peek..................FAILED at test 7
3203 =head2 SUPER-UX (NEC SX)
3205 The following tests are known to fail on SUPER-UX:
3207 op/64bitint...........................FAILED tests 29-30, 32-33, 35-36
3208 op/arith..............................FAILED tests 128-130
3209 op/pack...............................FAILED tests 25-5625
3210 op/pow................................
3211 op/taint..............................# msgsnd failed
3212 ../ext/IO/lib/IO/t/io_poll............FAILED tests 3-4
3213 ../ext/IPC/SysV/ipcsysv...............FAILED tests 2, 5-6
3214 ../ext/IPC/SysV/t/msg.................FAILED tests 2, 4-6
3215 ../ext/Socket/socketpair..............FAILED tests 12
3216 ../lib/IPC/SysV.......................FAILED tests 2, 5-6
3217 ../lib/warnings.......................FAILED tests 115-116, 118-119
3219 The op/pack failure ("Cannot compress negative numbers at op/pack.t line 126")
3220 is serious but as of yet unsolved. It points at some problems with the
3221 signedness handling of the C compiler, as do the 64bitint, arith, and pow
3222 failures. Most of the rest point at problems with SysV IPC.
3224 =head2 PDL failing some tests
3226 Use PDL 2.3.4 or later.
3228 =head2 Term::ReadKey not working on Win32
3230 Use Term::ReadKey 2.20 or later.
3232 =head2 Failure of Thread (5.005-style) tests
3234 B<Note that support for 5.005-style threading is deprecated,
3235 experimental and practically unsupported. In 5.10, it is expected
3238 The following tests are known to fail due to fundamental problems in
3239 the 5.005 threading implementation. These are not new failures--Perl
3240 5.005_0x has the same bugs, but didn't have these tests.
3242 ../ext/B/t/xref.t 255 65280 14 12 85.71% 3-14
3243 ../ext/List/Util/t/first.t 255 65280 7 4 57.14% 2 5-7
3244 ../lib/English.t 2 512 54 2 3.70% 2-3
3245 ../lib/ExtUtils/t/basic.t 1 256 17 1 5.88% 14
3246 ../lib/FileCache.t 5 1 20.00% 5
3247 ../lib/Filter/Simple/t/data.t 6 3 50.00% 1-3
3248 ../lib/Filter/Simple/t/filter_onl 9 3 33.33% 1-2 5
3249 ../lib/Tie/File/t/31_autodefer.t 255 65280 65 32 49.23% 34-65
3250 ../lib/autouse.t 10 1 10.00% 4
3251 op/flip.t 15 1 6.67% 15
3253 These failures are unlikely to get fixed as 5.005-style threads
3254 are considered fundamentally broken. (Basically what happens is that
3255 competing threads can corrupt shared global state.)
3257 =head2 Timing problems
3259 The following tests may fail intermittently because of timing
3260 problems, for example if the system is heavily loaded.
3263 ext/Time/HiRes/HiRes.t
3265 lib/Memoize/t/expmod_t.t
3266 lib/Memoize/t/speed.t
3268 In case of failure please try running them manually, for example
3270 ./perl -Ilib ext/Time/HiRes/HiRes.t
3278 During Configure, the test
3280 Guessing which symbols your C compiler and preprocessor define...
3282 will probably fail with error messages like
3284 CC-20 cc: ERROR File = try.c, Line = 3
3285 The identifier "bad" is undefined.
3287 bad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K
3290 CC-65 cc: ERROR File = try.c, Line = 3
3291 A semicolon is expected at this point.
3293 This is caused by a bug in the awk utility of UNICOS/mk. You can ignore
3294 the error, but it does cause a slight problem: you cannot fully
3295 benefit from the h2ph utility (see L<h2ph>) that can be used to
3296 convert C headers to Perl libraries, mainly used to be able to access
3297 from Perl the constants defined using C preprocessor, cpp. Because of
3298 the above error, parts of the converted headers will be invisible.
3299 Luckily, these days the need for h2ph is rare.
3303 If building Perl with interpreter threads (ithreads), the
3304 getgrent(), getgrnam(), and getgrgid() functions cannot return the
3305 list of the group members due to a bug in the multithreaded support of
3306 UNICOS/mk. What this means is that in list context the functions will
3307 return only three values, not four.
3313 There are a few known test failures, see L<perluts> (README.uts).
3315 =head2 VOS (Stratus)
3317 When Perl is built using the native build process on VOS Release
3318 14.5.0 and GNU C++/GNU Tools 2.0.1, all attempted tests either
3319 pass or result in TODO (ignored) failures.
3323 There should be no reported test failures with a default configuration,
3324 though there are a number of tests marked TODO that point to areas
3325 needing further debugging and/or porting work.
3329 In multi-CPU boxes, there are some problems with the I/O buffering:
3330 some output may appear twice.
3332 =head2 XML::Parser not working
3334 Use XML::Parser 2.31 or later.
3336 =head2 z/OS (OS/390)
3338 z/OS has rather many test failures but the situation is actually
3339 better than it was in 5.6.0; it's just that so many new modules and
3340 tests have been added.
3342 Failed Test Stat Wstat Total Fail Failed List of Failed
3343 ---------------------------------------------------------------------------
3344 ../ext/Data/Dumper/t/dumper.t 357 8 2.24% 311 314 325 327
3346 ../ext/IO/lib/IO/t/io_unix.t 5 4 80.00% 2-5
3347 ../ext/Storable/t/downgrade.t 12 3072 169 12 7.10% 14-15 46-47 78-79
3349 ../lib/ExtUtils/t/Constant.t 121 30976 48 48 100.00% 1-48
3350 ../lib/ExtUtils/t/Embed.t 9 9 100.00% 1-9
3351 op/pat.t 910 7 0.77% 665 776 785 832-
3353 op/sprintf.t 224 3 1.34% 98 100 136
3354 op/tr.t 97 5 5.15% 63 71-74
3355 uni/fold.t 780 6 0.77% 61 169 196 661
3358 The failures in dumper.t and downgrade.t are problems in the tests,
3359 those in io_unix and sprintf are problems in the USS (UDP sockets
3360 and printf formats). The pat, tr, and fold failures are genuine Perl
3361 problems caused by EBCDIC (and in the pat and fold cases, combining
3362 that with Unicode). The Constant and Embed are probably problems
3363 in the tests (since they test Perl's ability to build extensions,
3364 and that seems to be working reasonably well.)
3366 =head2 Localising Tied Arrays and Hashes Is Broken
3370 doesn't work as one would expect: the old value is restored
3371 incorrectly. This will be changed in a future release, but we don't
3372 know yet what the new semantics will exactly be. In any case, the
3373 change will break existing code that relies on the current
3374 (ill-defined) semantics, so just avoid doing this in general.
3376 =head2 Self-tying Problems
3378 Self-tying of arrays and hashes is broken in rather deep and
3379 hard-to-fix ways. As a stop-gap measure to avoid people from getting
3380 frustrated at the mysterious results (core dumps, most often), it is
3381 forbidden for now (you will get a fatal error even from an attempt).
3383 A change to self-tying of globs has caused them to be recursively
3384 referenced (see: L<perlobj/"Two-Phased Garbage Collection">). You
3385 will now need an explicit untie to destroy a self-tied glob. This
3386 behaviour may be fixed at a later date.
3388 Self-tying of scalars and IO thingies works.
3390 =head2 Tied/Magical Array/Hash Elements Do Not Autovivify
3392 For normal arrays C<$foo = \$bar[1]> will assign C<undef> to
3393 C<$bar[1]> (assuming that it didn't exist before), but for
3394 tied/magical arrays and hashes such autovivification does not happen
3395 because there is currently no way to catch the reference creation.
3396 The same problem affects slicing over non-existent indices/keys of
3397 a tied/magical array/hash.
3399 =head2 Building Extensions Can Fail Because Of Largefiles
3401 Some extensions like mod_perl are known to have issues with
3402 `largefiles', a change brought by Perl 5.6.0 in which file offsets
3403 default to 64 bits wide, where supported. Modules may fail to compile
3404 at all, or they may compile and work incorrectly. Currently, there
3405 is no good solution for the problem, but Configure now provides
3406 appropriate non-largefile ccflags, ldflags, libswanted, and libs
3407 in the %Config hash (e.g., $Config{ccflags_nolargefiles}) so the
3408 extensions that are having problems can try configuring themselves
3409 without the largefileness. This is admittedly not a clean solution,
3410 and the solution may not even work at all. One potential failure is
3411 whether one can (or, if one can, whether it's a good idea to) link
3412 together at all binaries with different ideas about file offsets;
3413 all this is platform-dependent.
3415 =head2 Unicode Support on EBCDIC Still Spotty
3417 Though mostly working, Unicode support still has problem spots on
3418 EBCDIC platforms. One such known spot are the C<\p{}> and C<\P{}>
3419 regular expression constructs for code points less than 256: the
3420 C<pP> are testing for Unicode code points, not knowing about EBCDIC.
3422 =head2 The Compiler Suite Is Still Very Experimental
3424 The compiler suite is slowly getting better but it continues to be
3425 highly experimental. Use in production environments is discouraged.
3427 =head2 The Long Double Support Is Still Experimental
3429 The ability to configure Perl's numbers to use "long doubles",
3430 floating point numbers of hopefully better accuracy, is still
3431 experimental. The implementations of long doubles are not yet
3432 widespread and the existing implementations are not quite mature
3433 or standardised, therefore trying to support them is a rare
3434 and moving target. The gain of more precision may also be offset
3435 by slowdown in computations (more bits to move around, and the
3436 operations are more likely to be executed by less optimised
3439 =head2 Seen In Perl 5.7 But Gone Now
3441 C<Time::Piece> (previously known as C<Time::Object>) was removed
3442 because it was felt that it didn't have enough value in it to be a
3443 core module. It is still a useful module, though, and is available
3446 Perl 5.8 unfortunately does not build anymore on AmigaOS; this broke
3447 accidentally at some point. Since there are not that many Amiga
3448 developers available, we could not get this fixed and tested in time
3449 for 5.8.0. Perl 5.6.1 still works for AmigaOS (as does the the 5.7.2
3450 development release).
3452 =head1 Reporting Bugs
3454 If you find what you think is a bug, you might check the articles
3455 recently posted to the comp.lang.perl.misc newsgroup and the perl
3456 bug database at http://bugs.perl.org/ . There may also be
3457 information at http://www.perl.com/ , the Perl Home Page.
3459 If you believe you have an unreported bug, please run the B<perlbug>
3460 program included with your release. Be sure to trim your bug down
3461 to a tiny but sufficient test case. Your bug report, along with the
3462 output of C<perl -V>, will be sent off to perlbug@perl.org to be
3463 analysed by the Perl porting team.
3467 The F<Changes> file for exhaustive details on what changed.
3469 The F<INSTALL> file for how to build Perl.
3471 The F<README> file for general stuff.
3473 The F<Artistic> and F<Copying> files for copyright information.
3477 Written by Jarkko Hietaniemi <F<jhi@iki.fi>>.