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
35 New Thread Implementation
39 Better Numeric Accuracy
51 More Extensive Regression Testing
55 =head1 Incompatible Changes
57 =head2 Binary Incompatibility
59 B<Perl 5.8 is not binary compatible with earlier releases of Perl.>
61 B<You have to recompile your XS modules.>
63 (Pure Perl modules should continue to work.)
65 The major reason for the discontinuity is the new IO architecture
66 called PerlIO. PerlIO is the default configuration because without
67 it many new features of Perl 5.8 cannot be used. In other words:
68 you just have to recompile your modules containing XS code, sorry
71 In future releases of Perl, non-PerlIO aware XS modules may become
72 completely unsupported. This shouldn't be too difficult for module
73 authors, however: PerlIO has been designed as a drop-in replacement
74 (at the source code level) for the stdio interface.
76 Depending on your platform, there are also other reasons why
77 we decided to break binary compatibility, please read on.
79 =head2 64-bit platforms and malloc
81 If your pointers are 64 bits wide, the Perl malloc is no longer being
82 used because it does not work well with 8-byte pointers. Also,
83 usually the system mallocs on such platforms are much better optimized
84 for such large memory models than the Perl malloc. Some memory-hungry
85 Perl applications like the PDL don't work well with Perl's malloc.
86 Finally, other applications than Perl (such as mod_perl) tend to prefer
87 the system malloc. Such platforms include Alpha and 64-bit HPPA,
90 =head2 AIX Dynaloading
92 The AIX dynaloading now uses in AIX releases 4.3 and newer the native
93 dlopen interface of AIX instead of the old emulated interface. This
94 change will probably break backward compatibility with compiled
95 modules. The change was made to make Perl more compliant with other
96 applications like mod_perl which are using the AIX native interface.
98 =head2 Attributes for C<my> variables now handled at run-time.
100 The C<my EXPR : ATTRS> syntax now applies variable attributes at
101 run-time. (Subroutine and C<our> variables still get attributes applied
102 at compile-time.) See L<attributes> for additional details. In particular,
103 however, this allows variable attributes to be useful for C<tie> interfaces,
104 which was a deficiency of earlier releases. Note that the new semantics
105 doesn't work with the Attribute::Handlers module (as of version 0.76).
107 =head2 Socket Extension Dynamic in VMS
109 The Socket extension is now dynamically loaded instead of being
110 statically built in. This may or may not be a problem with ancient
111 TCP/IP stacks of VMS: we do not know since we weren't able to test
112 Perl in such configurations.
114 =head2 IEEE-format Floating Point Default on OpenVMS Alpha
116 Perl now uses IEEE format (T_FLOAT) as the default internal floating
117 point format on OpenVMS Alpha, potentially breaking binary compatibility
118 with external libraries or existing data. G_FLOAT is still available as
119 a configuration option. The default on VAX (D_FLOAT) has not changed.
121 =head2 New Unicode Properties
123 Unicode I<scripts> are now supported. Scripts are similar to (and superior
124 to) Unicode I<blocks>. The difference between scripts and blocks is that
125 scripts are the glyphs used by a language or a group of languages, while
126 the blocks are more artificial groupings of (mostly) 256 characters based
127 on the Unicode numbering.
129 In general, scripts are more inclusive, but not universally so. For
130 example, while the script C<Latin> includes all the Latin characters and
131 their various diacritic-adorned versions, it does not include the various
132 punctuation or digits (since they are not solely C<Latin>).
134 A number of other properties are now supported, including C<\p{L&}>,
135 C<\p{Any}> C<\p{Assigned}>, C<\p{Unassigned}>, C<\p{Blank}> [561] and
136 C<\p{SpacePerl}> [561] (along with their C<\P{...}> versions, of course).
137 See L<perlunicode> for details, and more additions.
139 The C<In> or C<Is> prefix to names used with the C<\p{...}> and C<\P{...}>
140 are now almost always optional. The only exception is that a C<In> prefix
141 is required to signify a Unicode block when a block name conflicts with a
142 script name. For example, C<\p{Tibetan}> refers to the script, while
143 C<\p{InTibetan}> refers to the block. When there is no name conflict, you
144 can omit the C<In> from the block name (e.g. C<\p{BraillePatterns}>), but
145 to be safe, it's probably best to always use the C<In>).
147 =head2 REF(...) Instead Of SCALAR(...)
149 A reference to a reference now stringifies as "REF(0x81485ec)" instead
150 of "SCALAR(0x81485ec)" in order to be more consistent with the return
153 =head2 pack/unpack D/F recycled
155 The undocumented pack/unpack template letters D/F have been recycled
156 for better use: now they stand for long double (if supported by the
157 platform) and NV (Perl internal floating point type). (They used
158 to be aliases for d/f, but you never knew that.)
166 The semantics of bless(REF, REF) were unclear and until someone proves
167 it to make some sense, it is forbidden.
171 The obsolete chat2 library that should never have been allowed
172 to escape the laboratory has been decommissioned.
176 The builtin dump() function has probably outlived most of its
177 usefulness. The core-dumping functionality will remain in future
178 available as an explicit call to C<CORE::dump()>, but in future
179 releases the behaviour of an unqualified C<dump()> call may change.
183 The very dusty examples in the eg/ directory have been removed.
184 Suggestions for new shiny examples welcome but the main issue is that
185 the examples need to be documented, tested and (most importantly)
190 The (bogus) escape sequences \8 and \9 now give an optional warning
191 ("Unrecognized escape passed through"). There is no need to \-escape
196 The list of filenames from glob() (or <...>) is now by default sorted
197 alphabetically to be csh-compliant (which is what happened before
198 in most UNIX platforms). (bsd_glob() does still sort platform
199 natively, ASCII or EBCDIC, unless GLOB_ALPHASORT is specified.) [561]
203 The *glob{FILEHANDLE} is deprecated, use *glob{IO} instead.
207 Spurious syntax errors generated in certain situations, when glob()
208 caused File::Glob to be loaded for the first time, have been fixed. [561]
212 Although "you shouldn't do that", it was possible to write code that
213 depends on Perl's hashed key order (Data::Dumper does this). The new
214 algorithm "One-at-a-Time" produces a different hashed key order.
215 More details are in L</"Performance Enhancements">.
219 lstat(FILEHANDLE) now gives a warning because the operation makes no sense.
220 In future releases this may become a fatal error.
224 The C<package;> syntax (C<package> without an argument) has been
225 deprecated. Its semantics were never that clear and its
226 implementation even less so. If you have used that feature to
227 disallow all but fully qualified variables, C<use strict;> instead.
231 The unimplemented POSIX regex features [[.cc.]] and [[=c=]] are still
232 recognised but now cause fatal errors. The previous behaviour of
233 ignoring them by default and warning if requested was unacceptable
234 since it, in a way, falsely promised that the features could be used.
238 In future releases, non-PerlIO aware XS modules may become completely
239 unsupported. Since PerlIO is a drop-in replacement for stdio at the
240 source code level, this shouldn't be that drastic a change.
244 Previous versions of perl and some readings of some sections of Camel III
245 implied that the C<:raw> "discipline" was the inverse of C<:crlf>.
246 Turning off "clrfness" is no longer enough to make a stream truly
247 binary. So the PerlIO C<:raw> layer (or "discipline", to use the
248 Camel book's older terminology) is now formally defined as being
249 equivalent to binmode(FH) - which is in turn defined as doing whatever
250 is necessary to pass each byte as-is without any translation.
251 In particular binmode(FH) - and hence C<:raw> - will now turn off both CRLF
252 and UTF-8 translation and remove other layers (e.g. :encoding()) which
253 would modify byte stream.
257 The current user-visible implementation of pseudo-hashes (the weird
258 use of the first array element) is deprecated starting from Perl 5.8.0
259 and will be removed in Perl 5.10.0, and the feature will be
260 implemented differently. Not only is the current interface rather
261 ugly, but the current implementation slows down normal array and hash
262 use quite noticeably. The C<fields> pragma interface will remain
263 available. The I<restricted hashes> interface is expected to
264 be the replacement interface (see L<Hash::Util>). If your existing
265 programs depends on the underlying implementation, consider using
266 L<Class::PseudoHash> from CPAN.
270 The syntaxes C<< @a->[...] >> and C<< %h->{...} >> have now been deprecated.
274 After years of trying, suidperl is considered to be too complex to
275 ever be considered truly secure. The suidperl functionality is likely
276 to be removed in a future release.
280 The 5.005 threads model (module C<Thread>) is deprecated and expected
281 to be removed in Perl 5.10. Multithreaded code should be migrated to
282 the new ithreads model (see L<threads>, L<threads::shared> and
287 The long deprecated uppercase aliases for the string comparison
288 operators (EQ, NE, LT, LE, GE, GT) have now been removed.
292 The tr///C and tr///U features have been removed and will not return;
293 the interface was a mistake. Sorry about that. For similar
294 functionality, see pack('U0', ...) and pack('C0', ...). [561]
298 Earlier Perls treated "sub foo (@bar)" as equivalent to "sub foo (@)".
299 The prototypes are now checked better at compile-time for invalid
300 syntax. An optional warning is generated ("Illegal character in
301 prototype...") but this may be upgraded to a fatal error in a future
306 The C<exec LIST> and C<system LIST> operations will produce fatal
307 errors on tainted data in some future release.
311 The existing behaviour when localising tied arrays and hashes is wrong,
312 and will be changed in a future release, so do not rely on the existing
313 behaviour. See L<"Localising Tied Arrays and Hashes Is Broken">.
317 =head1 Core Enhancements
319 =head2 PerlIO is Now The Default
325 IO is now by default done via PerlIO rather than system's "stdio".
326 PerlIO allows "layers" to be "pushed" onto a file handle to alter the
327 handle's behaviour. Layers can be specified at open time via 3-arg
330 open($fh,'>:crlf :utf8', $path) || ...
332 or on already opened handles via extended C<binmode>:
334 binmode($fh,':encoding(iso-8859-7)');
336 The built-in layers are: unix (low level read/write), stdio (as in
337 previous Perls), perlio (re-implementation of stdio buffering in a
338 portable manner), crlf (does CRLF <=> "\n" translation as on Win32,
339 but available on any platform). A mmap layer may be available if
340 platform supports it (mostly UNIXes).
342 Layers to be applied by default may be specified via the 'open' pragma.
344 See L</"Installation and Configuration Improvements"> for the effects
345 of PerlIO on your architecture name.
349 If your platform supports fork(), you can use the list form of C<open>
350 for pipes. For example:
352 open KID_PS, "-|", "ps", "aux" or die $!;
354 forks the ps(1) command (without spawning a shell, as there are more
355 than three arguments to open()), and reads its standard output via the
356 C<KID_PS> filehandle. See L<perlipc>.
360 File handles can be marked as accepting Perl's internal encoding of Unicode
361 (UTF-8 or UTF-EBCDIC depending on platform) by a pseudo layer ":utf8" :
363 open($fh,">:utf8","Uni.txt");
365 Note for EBCDIC users: the pseudo layer ":utf8" is erroneously named
366 for you since it's not UTF-8 what you will be getting but instead
367 UTF-EBCDIC. See L<perlunicode>, L<utf8>, and
368 http://www.unicode.org/unicode/reports/tr16/ for more information.
369 In future releases this naming may change. See L<perluniintro>
370 for more information about UTF-8.
374 If your environment variables (LC_ALL, LC_CTYPE, LANG, LANGUAGE) look
375 like you want to use UTF-8 (any of the the variables match C</utf-?8/i>),
376 your STDIN, STDOUT, STDERR handles and the default open layer
377 (see L<open>) are marked as UTF-8. (This feature, like other new
378 features that combine Unicode and I/O, work only if you are using
379 PerlIO, but that's the default.)
381 Note that after this Perl really does assume that everything is UTF-8:
382 for example if some input handle is not, Perl will probably very soon
383 complain about the input data like this "Malformed UTF-8 ..." since
384 any old eight-bit data is not legal UTF-8.
386 Note for code authors: if you want to enable your users to use UTF-8
387 as their default encoding but in your code still have eight-bit I/O streams
388 (such as images or zip files), you need to explicitly open() or binmode()
389 with C<:bytes> (see L<perlfunc/open> and L<perlfunc/binmode>), or you
390 can just use C<binmode(FH)> (nice for pre-5.8.0 backward compatibility).
394 File handles can translate character encodings from/to Perl's internal
395 Unicode form on read/write via the ":encoding()" layer.
399 File handles can be opened to "in memory" files held in Perl scalars via:
401 open($fh,'>', \$variable) || ...
405 Anonymous temporary files are available without need to
406 'use FileHandle' or other module via
408 open($fh,"+>", undef) || ...
410 That is a literal undef, not an undefined value.
414 The list form of C<open> is now implemented for pipes (at least on UNIX):
416 open($fh,"-|", 'cat', '/etc/motd')
418 creates a pipe, and runs the equivalent of exec('cat', '/etc/motd') in
423 If your locale environment variables (LANGUAGE, LC_ALL, LC_CTYPE, LANG)
424 contain the strings 'UTF-8' or 'UTF8' (case-insensitive matching),
425 the default encoding of your STDIN, STDOUT, and STDERR, and of
426 B<any subsequent file open>, is UTF-8.
430 =head2 Restricted Hashes
432 A restricted hash is restricted to a certain set of keys, no keys
433 outside the set can be added. Also individual keys can be restricted
434 so that the key cannot be deleted and the value cannot be changed.
435 No new syntax is involved: the Hash::Util module is the interface.
439 Perl used to be fragile in that signals arriving at inopportune moments
440 could corrupt Perl's internal state. Now Perl postpones handling of
441 signals until it's safe (between opcodes).
443 This change may have surprising side effects because signals no longer
444 interrupt Perl instantly. Perl will now first finish whatever it was
445 doing, like finishing an internal operation (like sort()) or an
446 external operation (like an I/O operation), and only then look at any
447 arrived signals (and before starting the next operation). No more corrupt
448 internal state since the current operation is always finished first,
449 but the signal may take more time to get heard. Note that breaking
450 out from potentially blocking operations should still work, though.
452 =head2 Unicode Overhaul
454 Unicode in general should be now much more usable than in Perl 5.6.0
455 (or even in 5.6.1). Unicode can be used in hash keys, Unicode in
456 regular expressions should work now, Unicode in tr/// should work now,
457 Unicode in I/O should work now. See L<perluniintro> for introduction
458 and L<perlunicode> for details.
464 The Unicode Character Database coming with Perl has been upgraded
465 to Unicode 3.2.0. For more information, see http://www.unicode.org/ .
466 [561+] (5.6.1 has UCD 3.0.1.)
470 For developers interested in enhancing Perl's Unicode capabilities:
471 almost all the UCD files are included with the Perl distribution in
472 the F<lib/unicore> subdirectory. The most notable omission, for space
473 considerations, is the Unihan database.
477 The properties \p{Blank} and \p{SpacePerl} have been added. "Blank" is like
478 C isblank(), that is, it contains only "horizontal whitespace" (the space
479 character is, the newline isn't), and the "SpacePerl" is the Unicode
480 equivalent of C<\s> (\p{Space} isn't, since that includes the vertical
481 tabulator character, whereas C<\s> doesn't.)
483 See "New Unicode Properties" earlier in this document for additional
484 information on changes with Unicode properties.
488 =head2 Understanding of Numbers
490 In general a lot of fixing has happened in the area of Perl's
491 understanding of numbers, both integer and floating point. Since in
492 many systems the standard number parsing functions like C<strtoul()>
493 and C<atof()> seem to have bugs, Perl tries to work around their
494 deficiencies. This results hopefully in more accurate numbers.
496 Perl now tries internally to use integer values in numeric conversions
497 and basic arithmetics (+ - * /) if the arguments are integers, and
498 tries also to keep the results stored internally as integers.
499 This change leads to often slightly faster and always less lossy
500 arithmetics. (Previously Perl always preferred floating point numbers
503 =head2 Arrays now always interpolate into double-quoted strings [561]
505 In double-quoted strings, arrays now interpolate, no matter what. The
506 behavior in earlier versions of perl 5 was that arrays would interpolate
507 into strings if the array had been mentioned before the string was
508 compiled, and otherwise Perl would raise a fatal compile-time error.
509 In versions 5.000 through 5.003, the error was
511 Literal @example now requires backslash
513 In versions 5.004_01 through 5.6.0, the error was
515 In string, @example now must be written as \@example
517 The idea here was to get people into the habit of writing
518 C<"fred\@example.com"> when they wanted a literal C<@> sign, just as
519 they have always written C<"Give me back my \$5"> when they wanted a
522 Starting with 5.6.1, when Perl now sees an C<@> sign in a
523 double-quoted string, it I<always> attempts to interpolate an array,
524 regardless of whether or not the array has been used or declared
525 already. The fatal error has been downgraded to an optional warning:
527 Possible unintended interpolation of @example in string
529 This warns you that C<"fred@example.com"> is going to turn into
530 C<fred.com> if you don't backslash the C<@>.
531 See http://www.plover.com/~mjd/perl/at-error.html for more details
532 about the history here.
534 =head2 Miscellaneous Changes
540 AUTOLOAD is now lvaluable, meaning that you can add the :lvalue attribute
541 to AUTOLOAD subroutines and you can assign to the AUTOLOAD return value.
545 The $Config{byteorder} (and corresponding BYTEORDER in config.h) was
546 previously wrong in platforms if sizeof(long) was 4, but sizeof(IV)
547 was 8. The byteorder was only sizeof(long) bytes long (1234 or 4321),
548 but now it is correctly sizeof(IV) bytes long, (12345678 or 87654321).
549 (This problem didn't affect Windows platforms.)
551 Also, $Config{byteorder} is now computed dynamically--this is more
552 robust with "fat binaries" where an executable image contains binaries
553 for more than one binary platform, and when cross-compiling.
557 C<perl -d:Module=arg,arg,arg> now works (previously one couldn't pass
558 in multiple arguments.)
562 C<do> followed by a bareword now ensures that this bareword isn't
563 a keyword (to avoid a bug where C<do q(foo.pl)> tried to call a
564 subroutine called C<q>). This means that for example instead of
565 C<do format()> you must write C<do &format()>.
569 The builtin dump() now gives an optional warning
570 C<dump() better written as CORE::dump()>,
571 meaning that by default C<dump(...)> is resolved as the builtin
572 dump() which dumps core and aborts, not as (possibly) user-defined
573 C<sub dump>. To call the latter, qualify the call as C<&dump(...)>.
574 (The whole dump() feature is to considered deprecated, and possibly
575 removed/changed in future releases.)
579 chomp() and chop() are now overridable. Note, however, that their
580 prototype (as given by C<prototype("CORE::chomp")> is undefined,
581 because it cannot be expressed and therefore one cannot really write
582 replacements to override these builtins.
586 END blocks are now run even if you exit/die in a BEGIN block.
587 Internally, the execution of END blocks is now controlled by
588 PL_exit_flags & PERL_EXIT_DESTRUCT_END. This enables the new
589 behaviour for Perl embedders. This will default in 5.10. See
594 Formats now support zero-padded decimal fields.
598 Lvalue subroutines can now return C<undef> in list context. However,
599 the lvalue subroutine feature still remains experimental. [561+]
603 A lost warning "Can't declare ... dereference in my" has been
604 restored (Perl had it earlier but it became lost in later releases.)
608 A new special regular expression variable has been introduced:
609 C<$^N>, which contains the most-recently closed group (submatch).
613 C<no Module;> does not produce an error even if Module does not have an
614 unimport() method. This parallels the behavior of C<use> vis-a-vis
619 The numerical comparison operators return C<undef> if either operand
620 is a NaN. Previously the behaviour was unspecified.
624 C<our> can now have an experimental optional attribute C<unique> that
625 affects how global variables are shared among multiple interpreters,
630 The following builtin functions are now overridable: each(), keys(),
631 pop(), push(), shift(), splice(), unshift(). [561]
635 C<pack() / unpack()> can now group template letters with C<()> and then
636 apply repetition/count modifiers on the groups.
640 C<pack() / unpack()> can now process the Perl internal numeric types:
641 IVs, UVs, NVs-- and also long doubles, if supported by the platform.
642 The template letters are C<j>, C<J>, C<F>, and C<D>.
646 C<pack('U0a*', ...)> can now be used to force a string to UTF8.
650 my __PACKAGE__ $obj now works. [561]
654 POSIX::sleep() now returns the number of I<unslept> seconds
655 (as the POSIX standard says), as opposed to CORE::sleep() which
656 returns the number of slept seconds.
660 The printf() and sprintf() now support parameter reordering using the
661 C<%\d+\$> and C<*\d+\$> syntaxes. For example
663 print "%2\$s %1\$s\n", "foo", "bar";
665 will print "bar foo\n". This feature helps in writing
666 internationalised software, and in general when the order
667 of the parameters can vary.
671 The (\&) prototype now works properly. [561]
675 prototype(\[$@%&]) is now available to implicitly create references
676 (useful for example if you want to emulate the tie() interface).
680 A new command-line option, C<-t> is available. It is the
681 little brother of C<-T>: instead of dying on taint violations,
682 lexical warnings are given. B<This is only meant as a temporary
683 debugging aid while securing the code of old legacy applications.
684 This is not a substitute for -T.>
688 In other taint news, the C<exec LIST> and C<system LIST> have now been
689 considered too risky (think C<exec @ARGV>: it can start any program
690 with any arguments), and now the said forms cause a warning under
691 lexical warnings. You should carefully launder the arguments to
692 guarantee their validity. In future releases of Perl the forms will
693 become fatal errors so consider starting laundering now.
697 Tied hash interfaces are now required to have the EXISTS and DELETE
698 methods (either own or inherited).
702 If tr/// is just counting characters, it doesn't attempt to
707 untie() will now call an UNTIE() hook if it exists. See L<perltie>
712 L<utime> now supports C<utime undef, undef, @files> to change the
713 file timestamps to the current time.
717 The rules for allowing underscores (underbars) in numeric constants
718 have been relaxed and simplified: now you can have an underscore
719 simply B<between digits>.
723 Rather than relying on C's argv[0] (which may not contain a full pathname)
724 where possible $^X is now set by asking the operating system.
725 (eg by reading F</proc/self/exe> on Linux, F</proc/curproc/file> on FreeBSD)
729 A new variable, C<${^TAINT}>, indicates whether taint mode is enabled.
733 You can now override the readline() builtin, and this overrides also
734 the <FILEHANDLE> angle bracket operator.
738 The command-line options -s and -F are now recognized on the shebang
743 Use of the C</c> match modifier without an accompanying C</g> modifier
744 elicits a new warning: C<Use of /c modifier is meaningless without /g>.
746 Use of C</c> in substitutions, even with C</g>, elicits
747 C<Use of /c modifier is meaningless in s///>.
749 Use of C</g> with C<split> elicits C<Use of /g modifier is meaningless
754 Support for the C<CLONE> special subroutine had been added.
755 With ithreads, when a new thread is created, all Perl data is cloned,
756 however non-Perl data cannot be cloned automatically. In C<CLONE> you
757 can do whatever you need to do, like for example handle the cloning of
758 non-Perl data, if necessary. C<CLONE> will be executed once for every
759 package that has it defined or inherited. It will be called in the
760 context of the new thread, so all modifications are made in the new area.
766 =head1 Modules and Pragmata
768 =head2 New Modules and Pragmata
774 C<Attribute::Handlers>, originally by Damian Conway and now maintained
775 by Arthur Bergman, allows a class to define attribute handlers.
778 use Attribute::Handlers;
779 sub Wolf :ATTR(SCALAR) { print "howl!\n" }
781 # later, in some package using or inheriting from MyPack...
783 my MyPack $Fluffy : Wolf; # the attribute handler Wolf will be called
785 Both variables and routines can have attribute handlers. Handlers can
786 be specific to type (SCALAR, ARRAY, HASH, or CODE), or specific to the
787 exact compilation phase (BEGIN, CHECK, INIT, or END).
788 See L<Attribute::Handlers>.
792 C<B::Concise>, by Stephen McCamant, is a new compiler backend for
793 walking the Perl syntax tree, printing concise info about ops.
794 The output is highly customisable. See L<B::Concise>. [561+]
798 The new bignum, bigint, and bigrat pragmas, by Tels, implement
799 transparent bignum support (using the Math::BigInt, Math::BigFloat,
800 and Math::BigRat backends).
804 C<Class::ISA>, by Sean Burke, is a module for reporting the search
805 path for a class's ISA tree. See L<Class::ISA>.
809 C<Cwd> now has a split personality: if possible, an XS extension is
810 used, (this will hopefully be faster, more secure, and more robust)
811 but if not possible, the familiar Perl implementation is used.
815 C<Devel::PPPort>, originally by Kenneth Albanowski and now
816 maintained by Paul Marquess, has been added. It is primarily used
817 by C<h2xs> to enhance portability of XS modules between different
818 versions of Perl. See L<Devel::PPPort>.
822 C<Digest>, frontend module for calculating digests (checksums), from
823 Gisle Aas, has been added. See L<Digest>.
827 C<Digest::MD5> for calculating MD5 digests (checksums) as defined in
828 RFC 1321, from Gisle Aas, has been added. See L<Digest::MD5>.
830 use Digest::MD5 'md5_hex';
832 $digest = md5_hex("Thirsty Camel");
834 print $digest, "\n"; # 01d19d9d2045e005c3f1b80e8b164de1
836 NOTE: the C<MD5> backward compatibility module is deliberately not
837 included since its further use is discouraged.
839 See also L<PerlIO::via::QuotedPrint>.
843 C<Encode>, originally by Nick Ing-Simmons and now maintained by Dan
844 Kogai, provides a mechanism to translate between different character
845 encodings. Support for Unicode, ISO-8859-1, and ASCII are compiled in
846 to the module. Several other encodings (like the rest of the
847 ISO-8859, CP*/Win*, Mac, KOI8-R, three variants EBCDIC, Chinese,
848 Japanese, and Korean encodings) are included and can be loaded at
849 runtime. (For space considerations, the largest Chinese encodings
850 have been separated into their own CPAN module, Encode::HanExtra,
851 which Encode will use if available). See L<Encode>.
853 Any encoding supported by Encode module is also available to the
854 ":encoding()" layer if PerlIO is used.
858 C<Hash::Util> is the interface to the new I<restricted hashes>
859 feature. (Implemented by Jeffrey Friedl, Nick Ing-Simmons, and
860 Michael Schwern.) See L<Hash::Util>.
864 C<I18N::Langinfo> can be used to query locale information.
865 See L<I18N::Langinfo>.
869 C<I18N::LangTags>, by Sean Burke, has functions for dealing with
870 RFC3066-style language tags. See L<I18N::LangTags>.
874 C<ExtUtils::Constant>, by Nicholas Clark, is a new tool for extension
875 writers for generating XS code to import C header constants.
876 See L<ExtUtils::Constant>.
880 C<Filter::Simple>, by Damian Conway, is an easy-to-use frontend to
881 Filter::Util::Call. See L<Filter::Simple>.
887 use Filter::Simple sub {
888 while (my ($from, $to) = splice @_, 0, 2) {
897 use MyFilter qr/red/ => 'green';
899 print "red\n"; # this code is filtered, will print "green\n"
900 print "bored\n"; # this code is filtered, will print "bogreen\n"
904 print "red\n"; # this code is not filtered, will print "red\n"
908 C<File::Temp>, by Tim Jenness, allows one to create temporary files
909 and directories in an easy, portable, and secure way. See L<File::Temp>.
914 C<Filter::Util::Call>, by Paul Marquess, provides you with the
915 framework to write I<source filters> in Perl. For most uses, the
916 frontend Filter::Simple is to be preferred. See L<Filter::Util::Call>.
920 C<if>, by Ilya Zakharevich, is a new pragma for conditional inclusion
925 L<libnet>, by Graham Barr, is a collection of perl5 modules related
926 to network programming. See L<Net::FTP>, L<Net::NNTP>, L<Net::Ping>
927 (not part of libnet, but related), L<Net::POP3>, L<Net::SMTP>,
930 Perl installation leaves libnet unconfigured; use F<libnetcfg>
935 C<List::Util>, by Graham Barr, is a selection of general-utility
936 list subroutines, such as sum(), min(), first(), and shuffle().
941 C<Locale::Constants>, C<Locale::Country>, C<Locale::Currency>
942 C<Locale::Language>, and L<Locale::Script>, by Neil Bowers, have
943 been added. They provide the codes for various locale standards, such
944 as "fr" for France, "usd" for US Dollar, and "ja" for Japanese.
948 $country = code2country('jp'); # $country gets 'Japan'
949 $code = country2code('Norway'); # $code gets 'no'
951 See L<Locale::Constants>, L<Locale::Country>, L<Locale::Currency>,
952 and L<Locale::Language>.
956 C<Locale::Maketext>, by Sean Burke, is a localization framework. See
957 L<Locale::Maketext>, and L<Locale::Maketext::TPJ13>. The latter is an
958 article about software localization, originally published in The Perl
959 Journal #13, and republished here with kind permission.
963 C<Math::BigRat> for big rational numbers, to accompany Math::BigInt and
964 Math::BigFloat, from Tels. See L<Math::BigRat>.
968 C<Memoize> can make your functions faster by trading space for time,
969 from Mark-Jason Dominus. See L<Memoize>.
973 C<MIME::Base64>, by Gisle Aas, allows you to encode data in base64,
974 as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail
979 $encoded = encode_base64('Aladdin:open sesame');
980 $decoded = decode_base64($encoded);
982 print $encoded, "\n"; # "QWxhZGRpbjpvcGVuIHNlc2FtZQ=="
988 C<MIME::QuotedPrint>, by Gisle Aas, allows you to encode data
989 in quoted-printable encoding, as defined in RFC 2045 - I<MIME
990 (Multipurpose Internet Mail Extensions)>.
992 use MIME::QuotedPrint;
994 $encoded = encode_qp("Smiley in Unicode: \x{263a}");
995 $decoded = decode_qp($encoded);
997 print $encoded, "\n"; # "Smiley in Unicode: =263A"
999 See also L<PerlIO::via::QuotedPrint>.
1003 C<NEXT>, by Damian Conway, is a pseudo-class for method redispatch.
1008 C<open> is a new pragma for setting the default I/O layers
1013 C<PerlIO::scalar>, by Nick Ing-Simmons, provides the implementation
1014 of IO to "in memory" Perl scalars as discussed above. It also serves
1015 as an example of a loadable PerlIO layer. Other future possibilities
1016 include PerlIO::Array and PerlIO::Code. See L<PerlIO::scalar>.
1020 C<PerlIO::via>, by Nick Ing-Simmons, acts as a PerlIO layer and wraps
1021 PerlIO layer functionality provided by a class (typically implemented
1026 C<PerlIO::via::QuotedPrint>, by Elizabeth Mattijsen, is an example
1027 of a C<PerlIO::via> class:
1029 use PerlIO::via::QuotedPrint;
1030 open($fh,">:via(QuotedPrint)",$path);
1032 This will automatically convert everything output to C<$fh> to
1033 Quoted-Printable. See L<PerlIO::via> and L<PerlIO::via::QuotedPrint>.
1037 C<Pod::ParseLink>, by Russ Allbery, has been added,
1038 to parse LZ<><> links in pods as described in the new
1043 C<Pod::Text::Overstrike>, by Joe Smith, has been added.
1044 It converts POD data to formatted overstrike text.
1045 See L<Pod::Text::Overstrike>. [561+]
1049 C<Scalar::Util> is a selection of general-utility scalar subroutines,
1050 such as blessed(), reftype(), and tainted(). See L<Scalar::Util>.
1054 C<sort> is a new pragma for controlling the behaviour of sort().
1058 C<Storable> gives persistence to Perl data structures by allowing the
1059 storage and retrieval of Perl data to and from files in a fast and
1060 compact binary format. Because in effect Storable does serialisation
1061 of Perl data structures, with it you can also clone deep, hierarchical
1062 datastructures. Storable was originally created by Raphael Manfredi,
1063 but it is now maintained by Abhijit Menon-Sen. Storable has been
1064 enhanced to understand the two new hash features, Unicode keys and
1065 restricted hashes. See L<Storable>.
1069 C<Switch>, by Damian Conway, has been added. Just by saying
1073 you have C<switch> and C<case> available in Perl.
1079 case 1 { print "number 1" }
1080 case "a" { print "string a" }
1081 case [1..10,42] { print "number in list" }
1082 case (@array) { print "number in list" }
1083 case /\w+/ { print "pattern" }
1084 case qr/\w+/ { print "pattern" }
1085 case (%hash) { print "entry in hash" }
1086 case (\%hash) { print "entry in hash" }
1087 case (\&sub) { print "arg to subroutine" }
1088 else { print "previous case not true" }
1095 C<Test::More>, by Michael Schwern, is yet another framework for writing
1096 test scripts, more extensive than Test::Simple. See L<Test::More>.
1100 C<Test::Simple>, by Michael Schwern, has basic utilities for writing
1101 tests. See L<Test::Simple>.
1105 C<Text::Balanced>, by Damian Conway, has been added, for extracting
1106 delimited text sequences from strings.
1108 use Text::Balanced 'extract_delimited';
1110 ($a, $b) = extract_delimited("'never say never', he never said", "'", '');
1112 $a will be "'never say never'", $b will be ', he never said'.
1114 In addition to extract_delimited(), there are also extract_bracketed(),
1115 extract_quotelike(), extract_codeblock(), extract_variable(),
1116 extract_tagged(), extract_multiple(), gen_delimited_pat(), and
1117 gen_extract_tagged(). With these, you can implement rather advanced
1118 parsing algorithms. See L<Text::Balanced>.
1122 C<threads>, by Arthur Bergman, is an interface to interpreter threads.
1123 Interpreter threads (ithreads) is the new thread model introduced in
1124 Perl 5.6 but only available as an internal interface for extension
1125 writers (and for Win32 Perl for C<fork()> emulation). See L<threads>,
1126 L<threads::shared>, and L<perlthrtut>.
1130 C<threads::shared>, by Arthur Bergman, allows data sharing for
1131 interpreter threads. In the ithreads model any data sharing between
1132 threads must be explicit, as opposed to the old 5.005 thread model
1133 where data sharing was implicit. See L<threads::shared>.
1137 C<Tie::File>, by Mark-Jason Dominus, associates a Perl array with the
1138 lines of a file. See L<Tie::File>.
1142 C<Tie::Memoize>, by Ilya Zakharevich, provides on-demand loaded hashes.
1143 See L<Tie::Memoize>.
1147 C<Tie::RefHash::Nestable>, by Edward Avis, allows storing hash
1148 references (unlike the standard Tie::RefHash) The module is contained
1149 within Tie::RefHash. See L<Tie::RefHash>.
1153 C<Time::HiRes>, by Douglas E. Wegscheid, provides high resolution
1154 timing (ualarm, usleep, and gettimeofday). See L<Time::HiRes>.
1158 C<Unicode::UCD> offers a querying interface to the Unicode Character
1159 Database. See L<Unicode::UCD>.
1163 C<Unicode::Collate>, by SADAHIRO Tomoyuki, implements the UCA
1164 (Unicode Collation Algorithm) for sorting Unicode strings.
1165 See L<Unicode::Collate>.
1169 C<Unicode::Normalize>, by SADAHIRO Tomoyuki, implements the various
1170 Unicode normalization forms. See L<Unicode::Normalize>.
1174 C<XS::APItest>, by Tim Jenness, is a test extension that exercises XS
1175 APIs. Currently only C<printf()> is tested: how to output various
1176 basic data types from XS.
1180 C<XS::Typemap>, by Tim Jenness, is a test extension that exercises
1181 XS typemaps. Nothing gets installed, but the code is worth studying
1182 for extension writers.
1186 =head2 Updated And Improved Modules and Pragmata
1192 The following independently supported modules have been updated to the
1193 newest versions from CPAN: CGI, CPAN, DB_File, File::Spec, File::Temp,
1194 Getopt::Long, Math::BigFloat, Math::BigInt, the podlators bundle
1195 (Pod::Man, Pod::Text), Pod::LaTeX [561+], Pod::Parser, Storable,
1196 Term::ANSIColor, Test, Text-Tabs+Wrap.
1200 attributes::reftype() now works on tied arguments.
1204 AutoLoader can now be disabled with C<no AutoLoader;>.
1208 B::Deparse has been significantly enhanced by Robin Houston. It can
1209 now deparse almost all of the standard test suite (so that the tests
1210 still succeed). There is a make target "test.deparse" for trying this
1215 Carp now has better interface documentation, and the @CARP_NOT
1216 interface has been added to get optional control over where errors
1217 are reported independently of @ISA, by Ben Tilly.
1221 Class::Struct can now define the classes in compile time.
1225 Class::Struct now assigns the array/hash element if the accessor
1226 is called with an array/hash element as the B<sole> argument.
1230 The return value of Cwd::fastcwd() is now tainted.
1234 Data::Dumper now has an option to sort hashes.
1238 Data::Dumper now has an option to dump code references
1243 DB_File now supports newer Berkeley DB versions, among
1248 Devel::Peek now has an interface for the Perl memory statistics
1249 (this works only if you are using perl's malloc, and if you have
1250 compiled with debugging).
1254 The English module can now be used without the infamous performance
1257 use English '-no_match_vars';
1259 (Assuming, of course, that you don't need the troublesome variables
1260 C<$`>, C<$&>, or C<$'>.) Also, introduced C<@LAST_MATCH_START> and
1261 C<@LAST_MATCH_END> English aliases for C<@-> and C<@+>.
1265 ExtUtils::MakeMaker has been significantly cleaned up and fixed.
1266 The enhanced version has also been backported to earlier releases
1267 of Perl and submitted to CPAN so that the earlier releases can
1272 The arguments of WriteMakefile() in Makefile.PL are now checked
1273 for sanity much more carefully than before. This may cause new
1274 warnings when modules are being installed. See L<ExtUtils::MakeMaker>
1279 ExtUtils::MakeMaker now uses File::Spec internally, which hopefully
1280 leads to better portability.
1284 Fcntl, Socket, and Sys::Syslog have been rewritten by Nicholas Clark
1285 to use the new-style constant dispatch section (see L<ExtUtils::Constant>).
1286 This means that they will be more robust and hopefully faster.
1290 File::Find now chdir()s correctly when chasing symbolic links. [561]
1294 File::Find now has pre- and post-processing callbacks. It also
1295 correctly changes directories when chasing symbolic links. Callbacks
1296 (naughtily) exiting with "next;" instead of "return;" now work.
1300 File::Find is now (again) reentrant. It also has been made
1305 The warnings issued by File::Find now belong to their own category.
1306 You can enable/disable them with C<use/no warnings 'File::Find';>.
1310 File::Glob::glob() has been renamed to File::Glob::bsd_glob()
1311 because the name clashes with the builtin glob(). The older
1312 name is still available for compatibility, but is deprecated. [561]
1316 File::Glob now supports C<GLOB_LIMIT> constant to limit the size of
1317 the returned list of filenames.
1321 IPC::Open3 now allows the use of numeric file descriptors.
1325 IO::Socket now has an atmark() method, which returns true if the socket
1326 is positioned at the out-of-band mark. The method is also exportable
1327 as a sockatmark() function.
1331 IO::Socket::INET failed to open the specified port if the service name
1332 was not known. It now correctly uses the supplied port number as is. [561]
1336 IO::Socket::INET has support for the ReusePort option (if your
1337 platform supports it). The Reuse option now has an alias, ReuseAddr.
1338 For clarity, you may want to prefer ReuseAddr.
1342 IO::Socket::INET now supports a value of zero for C<LocalPort>
1343 (usually meaning that the operating system will make one up.)
1347 'use lib' now works identically to @INC. Removing directories
1348 with 'no lib' now works.
1352 Math::BigFloat and Math::BigInt have undergone a full rewrite by Tels.
1353 They are now magnitudes faster, and they support various bignum
1354 libraries such as GMP and PARI as their backends.
1358 Math::Complex handles inf, NaN etc., better.
1362 Net::Ping has been considerably enhanced by Rob Brown: multihoming is
1363 now supported, Win32 functionality is better, there is now time
1364 measuring functionality (optionally high-resolution using
1365 Time::HiRes), and there is now "external" protocol which uses
1366 Net::Ping::External module which runs your external ping utility and
1367 parses the output. A version of Net::Ping::External is available in
1370 Note that some of the Net::Ping tests are disabled when running
1371 under the Perl distribution since one cannot assume one or more
1372 of the following: enabled echo port at localhost, full Internet
1373 connectivity, or sympathetic firewalls. You can set the environment
1374 variable PERL_TEST_Net_Ping to "1" (one) before running the Perl test
1375 suite to enable all the Net::Ping tests.
1379 POSIX::sigaction() is now much more flexible and robust.
1380 You can now install coderef handlers, 'DEFAULT', and 'IGNORE'
1381 handlers, installing new handlers was not atomic.
1385 In Safe, C<%INC> is now localised in a Safe compartment so that
1390 In SDBM_File on dosish platforms, some keys went missing because of
1391 lack of support for files with "holes". A workaround for the problem
1396 In Search::Dict one can now have a pre-processing hook for the
1397 lines being searched.
1401 The Shell module now has an OO interface.
1405 In Sys::Syslog there is now a failover mechanism that will go
1406 through alternative connection mechanisms until the message
1407 is successfully logged.
1411 The Test module has been significantly enhanced.
1415 Time::Local::timelocal() does not handle fractional seconds anymore.
1416 The rationale is that neither does localtime(), and timelocal() and
1417 localtime() are supposed to be inverses of each other.
1421 The vars pragma now supports declaring fully qualified variables.
1422 (Something that C<our()> does not and will not support.)
1426 The C<utf8::> name space (as in the pragma) provides various
1427 Perl-callable functions to provide low level access to Perl's
1428 internal Unicode representation. At the moment only length()
1429 has been implemented.
1433 =head1 Utility Changes
1439 Emacs perl mode (emacs/cperl-mode.el) has been updated to version
1444 F<emacs/e2ctags.pl> is now much faster.
1448 C<enc2xs> is a tool for people adding their own encodings to the
1453 C<h2ph> now supports C trigraphs.
1457 C<h2xs> now produces a template README.
1461 C<h2xs> now uses C<Devel::PPPort> for better portability between
1462 different versions of Perl.
1466 C<h2xs> uses the new L<ExtUtils::Constant|ExtUtils::Constant> module
1467 which will affect newly created extensions that define constants.
1468 Since the new code is more correct (if you have two constants where the
1469 first one is a prefix of the second one, the first constant B<never>
1470 got defined), less lossy (it uses integers for integer constant,
1471 as opposed to the old code that used floating point numbers even for
1472 integer constants), and slightly faster, you might want to consider
1473 regenerating your extension code (the new scheme makes regenerating
1474 easy). L<h2xs> now also supports C trigraphs.
1478 C<libnetcfg> has been added to configure libnet.
1482 C<perlbug> is now much more robust. It also sends the bug report to
1483 perl.org, not perl.com.
1487 C<perlcc> has been rewritten and its user interface (that is,
1488 command line) is much more like that of the UNIX C compiler, cc.
1489 (The perlbc tools has been removed. Use C<perlcc -B> instead.)
1490 B<Note that perlcc is still considered very experimental and
1495 C<perlivp> is a new Installation Verification Procedure utility
1496 for running any time after installing Perl.
1500 C<piconv> is an implementation of the character conversion utility
1501 C<iconv>, demonstrating the new Encode module.
1505 C<pod2html> now allows specifying a cache directory.
1509 C<pod2html> now produces XHTML 1.0.
1513 C<pod2html> now understands POD written using different line endings
1514 (PC-like CRLF versus UNIX-like LF versus MacClassic-like CR).
1518 C<s2p> has been completely rewritten in Perl. (It is in fact a full
1519 implementation of sed in Perl: you can use the sed functionality by
1520 using the C<psed> utility.)
1524 C<xsubpp> now understands POD documentation embedded in the *.xs
1529 C<xsubpp> now supports the OUT keyword.
1533 =head1 New Documentation
1539 perl56delta details the changes between the 5.005 release and the
1544 perlclib documents the internal replacements for standard C library
1545 functions. (Interesting only for extension writers and Perl core
1550 perldebtut is a Perl debugging tutorial. [561+]
1554 perlebcdic contains considerations for running Perl on EBCDIC
1559 perlintro is a gentle introduction to Perl.
1563 perliol documents the internals of PerlIO with layers.
1567 perlmodstyle is a style guide for writing modules.
1571 perlnewmod tells about writing and submitting a new module. [561+]
1575 perlpacktut is a pack() tutorial.
1579 perlpod has been rewritten to be clearer and to record the best
1580 practices gathered over the years.
1584 perlpodspec is a more formal specification of the pod format,
1585 mainly of interest for writers of pod applications, not to
1586 people writing in pod.
1590 perlretut is a regular expression tutorial. [561+]
1594 perlrequick is a regular expressions quick-start guide.
1595 Yes, much quicker than perlretut. [561]
1599 perltodo has been updated.
1603 perltootc has been renamed as perltooc (to not to conflict
1604 with perltoot in filesystems restricted to "8.3" names).
1608 perluniintro is an introduction to using Unicode in Perl.
1609 (perlunicode is more of a detailed reference and background
1614 perlutil explains the command line utilities packaged with the Perl
1615 distribution. [561+]
1619 The following platform-specific documents are available before
1620 the installation as README.I<platform>, and after the installation
1623 perlaix perlamiga perlapollo perlbeos perlbs2000
1624 perlce perlcygwin perldgux perldos perlepoc perlfreebsd perlhpux
1625 perlhurd perlirix perlmachten perlmacos perlmint perlmpeix
1626 perlnetware perlos2 perlos390 perlplan9 perlqnx perlsolaris
1627 perltru64 perluts perlvmesa perlvms perlvos perlwin32
1629 These documents usually detail one or more of the following subjects:
1630 configuring, building, testing, installing, and sometimes also using
1631 Perl on the said platform.
1633 Eastern Asian Perl users are now welcomed in their own languages:
1634 README.jp (Japanese), README.ko (Korean), README.cn (simplified
1635 Chinese) and README.tw (traditional Chinese), which are written in
1636 normal pod but encoded in EUC-JP, EUC-KR, EUC-CN and Big5. These
1637 will get installed as
1639 perljp perlko perlcn perltw
1645 The documentation for the POSIX-BC platform is called "BS2000", to avoid
1646 confusion with the Perl POSIX module.
1650 The documentation for the WinCE platform is called perlce (README.ce
1651 in the source code kit), to avoid confusion with the perlwin32
1652 documentation on 8.3-restricted filesystems.
1656 =head1 Performance Enhancements
1662 map() could get pathologically slow when the result list it generates
1663 is larger than the source list. The performance has been improved for
1664 common scenarios. [561]
1668 sort() is also fully reentrant, in the sense that the sort function
1669 can itself call sort(). This did not work reliably in previous
1674 sort() has been changed to use primarily mergesort internally as
1675 opposed to the earlier quicksort. For very small lists this may
1676 result in slightly slower sorting times, but in general the speedup
1677 should be at least 20%. Additional bonuses are that the worst case
1678 behaviour of sort() is now better (in computer science terms it now
1679 runs in time O(N log N), as opposed to quicksort's Theta(N**2)
1680 worst-case run time behaviour), and that sort() is now stable
1681 (meaning that elements with identical keys will stay ordered as they
1682 were before the sort). See the C<sort> pragma for information.
1684 The story in more detail: suppose you want to serve yourself a little
1687 @digits = ( 3,1,4,1,5,9 );
1689 A numerical sort of the digits will yield (1,1,3,4,5,9), as expected.
1690 Which C<1> comes first is hard to know, since one C<1> looks pretty
1691 much like any other. You can regard this as totally trivial,
1692 or somewhat profound. However, if you just want to sort the even
1693 digits ahead of the odd ones, then what will
1695 sort { ($a % 2) <=> ($b % 2) } @digits;
1697 yield? The only even digit, C<4>, will come first. But how about
1698 the odd numbers, which all compare equal? With the quicksort algorithm
1699 used to implement Perl 5.6 and earlier, the order of ties is left up
1700 to the sort. So, as you add more and more digits of Pi, the order
1701 in which the sorted even and odd digits appear will change.
1702 and, for sufficiently large slices of Pi, the quicksort algorithm
1703 in Perl 5.8 won't return the same results even if reinvoked with the
1704 same input. The justification for this rests with quicksort's
1705 worst case behavior. If you run
1707 sort { $a <=> $b } ( 1 .. $N , 1 .. $N );
1709 (something you might approximate if you wanted to merge two sorted
1710 arrays using sort), doubling $N doesn't just double the quicksort time,
1711 it I<quadruples> it. Quicksort has a worst case run time that can
1712 grow like N**2, so-called I<quadratic> behaviour, and it can happen
1713 on patterns that may well arise in normal use. You won't notice this
1714 for small arrays, but you I<will> notice it with larger arrays,
1715 and you may not live long enough for the sort to complete on arrays
1716 of a million elements. So the 5.8 quicksort scrambles large arrays
1717 before sorting them, as a statistical defence against quadratic behaviour.
1718 But that means if you sort the same large array twice, ties may be
1719 broken in different ways.
1721 Because of the unpredictability of tie-breaking order, and the quadratic
1722 worst-case behaviour, quicksort was I<almost> replaced completely with
1723 a stable mergesort. I<Stable> means that ties are broken to preserve
1724 the original order of appearance in the input array. So
1726 sort { ($a % 2) <=> ($b % 2) } (3,1,4,1,5,9);
1728 will yield (4,3,1,1,5,9), guaranteed. The even and odd numbers
1729 appear in the output in the same order they appeared in the input.
1730 Mergesort has worst case O(N log N) behaviour, the best value
1731 attainable. And, ironically, this mergesort does particularly
1732 well where quicksort goes quadratic: mergesort sorts (1..$N, 1..$N)
1733 in O(N) time. But quicksort was rescued at the last moment because
1734 it is faster than mergesort on certain inputs and platforms.
1735 For example, if you really I<don't> care about the order of even
1736 and odd digits, quicksort will run in O(N) time; it's very good
1737 at sorting many repetitions of a small number of distinct elements.
1738 The quicksort divide and conquer strategy works well on platforms
1739 with relatively small, very fast, caches. Eventually, the problem gets
1740 whittled down to one that fits in the cache, from which point it
1741 benefits from the increased memory speed.
1743 Quicksort was rescued by implementing a sort pragma to control aspects
1744 of the sort. The B<stable> subpragma forces stable behaviour,
1745 regardless of algorithm. The B<_quicksort> and B<_mergesort>
1746 subpragmas are heavy-handed ways to select the underlying implementation.
1747 The leading C<_> is a reminder that these subpragmas may not survive
1748 beyond 5.8. More appropriate mechanisms for selecting the implementation
1749 exist, but they wouldn't have arrived in time to save quicksort.
1753 Hashes now use Bob Jenkins "One-at-a-Time" hashing key algorithm
1754 ( http://burtleburtle.net/bob/hash/doobs.html ). This algorithm is
1755 reasonably fast while producing a much better spread of values than
1756 the old hashing algorithm (originally by Chris Torek, later tweaked by
1757 Ilya Zakharevich). Hash values output from the algorithm on a hash of
1758 all 3-char printable ASCII keys comes much closer to passing the
1759 DIEHARD random number generation tests. According to perlbench, this
1760 change has not affected the overall speed of Perl.
1764 unshift() should now be noticeably faster.
1768 =head1 Installation and Configuration Improvements
1770 =head2 Generic Improvements
1776 INSTALL now explains how you can configure Perl to use 64-bit
1777 integers even on non-64-bit platforms.
1781 Policy.sh policy change: if you are reusing a Policy.sh file
1782 (see INSTALL) and you use Configure -Dprefix=/foo/bar and in the old
1783 Policy $prefix eq $siteprefix and $prefix eq $vendorprefix, all of
1784 them will now be changed to the new prefix, /foo/bar. (Previously
1785 only $prefix changed.) If you do not like this new behaviour,
1786 specify prefix, siteprefix, and vendorprefix explicitly.
1790 A new optional location for Perl libraries, otherlibdirs, is available.
1791 It can be used for example for vendor add-ons without disturbing Perl's
1792 own library directories.
1796 In many platforms, the vendor-supplied 'cc' is too stripped-down to
1797 build Perl (basically, 'cc' doesn't do ANSI C). If this seems
1798 to be the case and 'cc' does not seem to be the GNU C compiler
1799 'gcc', an automatic attempt is made to find and use 'gcc' instead.
1803 gcc needs to closely track the operating system release to avoid
1804 build problems. If Configure finds that gcc was built for a different
1805 operating system release than is running, it now gives a clearly visible
1806 warning that there may be trouble ahead.
1810 Since Perl 5.8 is not binary-compatible with previous releases
1811 of Perl, Configure no longer suggests including the 5.005
1816 Configure C<-S> can now run non-interactively. [561]
1820 Configure support for pdp11-style memory models has been removed due
1821 to obsolescence. [561]
1825 configure.gnu now works with options with whitespace in them.
1829 installperl now outputs everything to STDERR.
1833 Because PerlIO is now the default on most platforms, "-perlio" doesn't
1834 get appended to the $Config{archname} (also known as $^O) anymore.
1835 Instead, if you explicitly choose not to use perlio (Configure command
1836 line option -Uuseperlio), you will get "-stdio" appended.
1840 Another change related to the architecture name is that "-64all"
1841 (-Duse64bitall, or "maximally 64-bit") is appended only if your
1842 pointers are 64 bits wide. (To be exact, the use64bitall is ignored.)
1846 In AFS installations, one can configure the root of the AFS to be
1847 somewhere else than the default F</afs> by using the Configure
1848 parameter C<-Dafsroot=/some/where/else>.
1852 APPLLIB_EXP, a lesser-known configuration-time definition, has been
1853 documented. It can be used to prepend site-specific directories
1854 to Perl's default search path (@INC); see INSTALL for information.
1858 The version of Berkeley DB used when the Perl (and, presumably, the
1859 DB_File extension) was built is now available as
1860 C<@Config{qw(db_version_major db_version_minor db_version_patch)}>
1861 from Perl and as C<DB_VERSION_MAJOR_CFG DB_VERSION_MINOR_CFG
1862 DB_VERSION_PATCH_CFG> from C.
1866 Building Berkeley DB3 for compatibility modes for DB, NDBM, and ODBM
1867 has been documented in INSTALL.
1871 If you have CPAN access (either network or a local copy such as a
1872 CD-ROM) you can during specify extra modules to Configure to build and
1873 install with Perl using the -Dextras=... option. See INSTALL for
1878 In addition to config.over, a new override file, config.arch, is
1879 available. This file is supposed to be used by hints file writers
1880 for architecture-wide changes (as opposed to config.over which is
1881 for site-wide changes).
1885 If your file system supports symbolic links, you can build Perl outside
1886 of the source directory by
1888 mkdir /tmp/perl/build/directory
1889 cd /tmp/perl/build/directory
1890 sh /path/to/perl/source/Configure -Dmksymlinks ...
1892 This will create in /tmp/perl/build/directory a tree of symbolic links
1893 pointing to files in /path/to/perl/source. The original files are left
1894 unaffected. After Configure has finished, you can just say
1898 and Perl will be built and tested, all in /tmp/perl/build/directory.
1903 For Perl developers, several new make targets for profiling
1904 and debugging have been added; see L<perlhack>.
1910 Use of the F<gprof> tool to profile Perl has been documented in
1911 L<perlhack>. There is a make target called "perl.gprof" for
1912 generating a gprofiled Perl executable.
1916 If you have GCC 3, there is a make target called "perl.gcov" for
1917 creating a gcoved Perl executable for coverage analysis. See
1922 If you are on IRIX or Tru64 platforms, new profiling/debugging options
1923 have been added; see L<perlhack> for more information about pixie and
1930 Guidelines of how to construct minimal Perl installations have
1931 been added to INSTALL.
1935 The Thread extension is now not built at all under ithreads
1936 (C<Configure -Duseithreads>) because it wouldn't work anyway (the
1937 Thread extension requires being Configured with C<-Duse5005threads>).
1939 B<Note that the 5.005 threads are unsupported and deprecated: if you
1940 have code written for the old threads you should migrate it to the
1941 new ithreads model.>
1945 The Gconvert macro ($Config{d_Gconvert}) used by perl for stringifying
1946 floating-point numbers is now more picky about using sprintf %.*g
1947 rules for the conversion. Some platforms that used to use gcvt may
1948 now resort to the slower sprintf.
1952 The obsolete method of making a special (e.g., debugging) flavor
1955 make LIBPERL=libperld.a
1957 has been removed. Use -DDEBUGGING instead.
1961 =head2 New Or Improved Platforms
1963 For the list of platforms known to support Perl,
1964 see L<perlport/"Supported Platforms">.
1970 AIX dynamic loading should be now better supported.
1974 AIX should now work better with gcc, threads, and 64-bitness. Also the
1975 long doubles support in AIX should be better now. See L<perlaix>.
1979 AtheOS ( http://www.atheos.cx/ ) is a new platform.
1983 BeOS has been reclaimed.
1987 The DG/UX platform now supports 5.005-style threads.
1992 The DYNIX/ptx platform (also known as dynixptx) is supported at or
1997 EBCDIC platforms (z/OS (also known as OS/390), POSIX-BC, and VM/ESA)
1998 have been regained. Many test suite tests still fail and the
1999 co-existence of Unicode and EBCDIC isn't quite settled, but the
2000 situation is much better than with Perl 5.6. See L<perlos390>,
2001 L<perlbs2000> (for POSIX-BC), and L<perlvmesa> for more information.
2005 Building perl with -Duseithreads or -Duse5005threads now works under
2006 HP-UX 10.20 (previously it only worked under 10.30 or later). You will
2007 need a thread library package installed. See README.hpux. [561]
2011 Mac OS Classic is now supported in the mainstream source package
2012 (MacPerl has of course been available since perl 5.004 but now the
2013 source code bases of standard Perl and MacPerl have been synchronised)
2018 Mac OS X (or Darwin) should now be able to build Perl even on HFS+
2019 filesystems. (The case-insensitivity used to confuse the Perl build
2024 NCR MP-RAS is now supported. [561]
2028 All the NetBSD specific patches (except for the installation
2029 specific ones) have been merged back to the main distribution.
2033 NetWare from Novell is now supported. See L<perlnetware>.
2037 NonStop-UX is now supported. [561]
2041 NEC SUPER-UX is now supported.
2045 All the OpenBSD specific patches (except for the installation
2046 specific ones) have been merged back to the main distribution.
2050 Perl has been tested with the GNU pth userlevel thread package
2051 ( http://www.gnu.org/software/pth/pth.html ). All thread tests
2052 of Perl now work, but not without adding some yield()s to the tests,
2053 so while pth (and other userlevel thread implementations) can be
2054 considered to be "working" with Perl ithreads, keep in mind the
2055 possible non-preemptability of the underlying thread implementation.
2059 Stratus VOS is now supported using Perl's native build method
2060 (Configure). This is the recommended method to build Perl on
2061 VOS. The older methods, which build miniperl, are still
2062 available. See L<perlvos>. [561+]
2066 The Amdahl UTS UNIX mainframe platform is now supported. [561]
2070 WinCE is now supported. See L<perlce>.
2074 z/OS (formerly known as OS/390, formerly known as MVS OE) now has
2075 support for dynamic loading. This is not selected by default,
2076 however, you must specify -Dusedl in the arguments of Configure. [561]
2080 =head1 Selected Bug Fixes
2082 Numerous memory leaks and uninitialized memory accesses have been
2083 hunted down. Most importantly, anonymous subs used to leak quite
2090 The autouse pragma didn't work for Multi::Part::Function::Names.
2094 caller() could cause core dumps in certain situations. Carp was
2095 sometimes affected by this problem. In particular, caller() now
2096 returns a subroutine name of C<(unknown)> for subroutines that have
2097 been removed from the symbol table.
2101 chop(@list) in list context returned the characters chopped in
2102 reverse order. This has been reversed to be in the right order. [561]
2106 Configure no longer includes the DBM libraries (dbm, gdbm, db, ndbm)
2107 when building the Perl binary. The only exception to this is SunOS 4.x,
2108 which needs them. [561]
2112 The behaviour of non-decimal but numeric string constants such as
2113 "0x23" was platform-dependent: in some platforms that was seen as 35,
2114 in some as 0, in some as a floating point number (don't ask). This
2115 was caused by Perl's using the operating system libraries in a situation
2116 where the result of the string to number conversion is undefined: now
2117 Perl consistently handles such strings as zero in numeric contexts.
2121 Several debugger fixes: exit code now reflects the script exit code,
2122 condition C<"0"> now treated correctly, the C<d> command now checks
2123 line number, C<$.> no longer gets corrupted, and all debugger output
2124 now goes correctly to the socket if RemotePort is set. [561]
2128 The debugger (perl5db.pl) has been modified to present a more
2129 consistent commands interface, via (CommandSet=580). perl5db.t was
2130 also added to test the changes, and as a placeholder for further tests.
2136 The debugger has a new C<dumpDepth> option to control the maximum
2137 depth to which nested structures are dumped. The C<x> command has
2138 been extended so that C<x N EXPR> dumps out the value of I<EXPR> to a
2139 depth of at most I<N> levels.
2143 The debugger can now show lexical variables if you have the CPAN
2144 module PadWalker installed.
2148 The order of DESTROYs has been made more predictable.
2152 Perl 5.6.0 could emit spurious warnings about redefinition of
2153 dl_error() when statically building extensions into perl.
2154 This has been corrected. [561]
2158 L<dprofpp> -R didn't work.
2162 C<*foo{FORMAT}> now works.
2166 Infinity is now recognized as a number.
2170 UNIVERSAL::isa no longer caches methods incorrectly. (This broke
2171 the Tk extension with 5.6.0.) [561]
2175 Lexicals I: lexicals outside an eval "" weren't resolved
2176 correctly inside a subroutine definition inside the eval "" if they
2177 were not already referenced in the top level of the eval""ed code.
2181 Lexicals II: lexicals leaked at file scope into subroutines that
2182 were declared before the lexicals.
2186 Lexical warnings now propagating correctly between scopes
2187 and into C<eval "...">.
2191 C<use warnings qw(FATAL all)> did not work as intended. This has been
2196 warnings::enabled() now reports the state of $^W correctly if the caller
2197 isn't using lexical warnings. [561]
2201 Line renumbering with eval and C<#line> now works. [561]
2205 Fixed numerous memory leaks, especially in eval "".
2209 Localised tied variables no longer leak memory
2212 tie my %tied_hash => 'Tie::StdHash';
2216 # Used to leak memory every time local() was called;
2217 # in a loop, this added up.
2218 local($tied_hash{Foo}) = 1;
2222 Localised hash elements (and %ENV) are correctly unlocalised to not
2223 exist, if they didn't before they were localised.
2227 tie my %tied_hash => 'Tie::StdHash';
2231 # Nothing has set the FOO element so far
2233 { local $tied_hash{FOO} = 'Bar' }
2235 # This used to print, but not now.
2236 print "exists!\n" if exists $tied_hash{FOO};
2238 As a side effect of this fix, tied hash interfaces B<must> define
2239 the EXISTS and DELETE methods.
2243 mkdir() now ignores trailing slashes in the directory name,
2244 as mandated by POSIX.
2248 Some versions of glibc have a broken modfl(). This affects builds
2249 with C<-Duselongdouble>. This version of Perl detects this brokenness
2250 and has a workaround for it. The glibc release 2.2.2 is known to have
2251 fixed the modfl() bug.
2255 Modulus of unsigned numbers now works (4063328477 % 65535 used to
2256 return 27406, instead of 27047). [561]
2260 Some "not a number" warnings introduced in 5.6.0 eliminated to be
2261 more compatible with 5.005. Infinity is now recognised as a number. [561]
2265 Numeric conversions did not recognize changes in the string value
2266 properly in certain circumstances. [561]
2270 Attributes (such as :shared) didn't work with our().
2274 our() variables will not cause bogus "Variable will not stay shared"
2279 "our" variables of the same name declared in two sibling blocks
2280 resulted in bogus warnings about "redeclaration" of the variables.
2281 The problem has been corrected. [561]
2285 pack "Z" now correctly terminates the string with "\0".
2289 Fix password routines which in some shadow password platforms
2290 (e.g. HP-UX) caused getpwent() to return every other entry.
2294 The PERL5OPT environment variable (for passing command line arguments
2295 to Perl) didn't work for more than a single group of options. [561]
2299 PERL5OPT with embedded spaces didn't work.
2303 printf() no longer resets the numeric locale to "C".
2307 C<qw(a\\b)> now parses correctly as C<'a\\b'>: that is, as three
2308 characters, not four. [561]
2312 pos() did not return the correct value within s///ge in earlier
2313 versions. This is now handled correctly. [561]
2317 Printing quads (64-bit integers) with printf/sprintf now works
2318 without the q L ll prefixes (assuming you are on a quad-capable platform).
2322 Regular expressions on references and overloaded scalars now work. [561+]
2326 Right-hand side magic (GMAGIC) could in many cases such as string
2327 concatenation be invoked too many times.
2331 scalar() now forces scalar context even when used in void context.
2335 SOCKS support is now much more robust.
2339 sort() arguments are now compiled in the right wantarray context
2340 (they were accidentally using the context of the sort() itself).
2341 The comparison block is now run in scalar context, and the arguments
2342 to be sorted are always provided list context. [561]
2346 Changed the POSIX character class C<[[:space:]]> to include the (very
2347 rarely used) vertical tab character. Added a new POSIX-ish character
2348 class C<[[:blank:]]> which stands for horizontal whitespace
2349 (currently, the space and the tab).
2353 The tainting behaviour of sprintf() has been rationalized. It does
2354 not taint the result of floating point formats anymore, making the
2355 behaviour consistent with that of string interpolation. [561]
2359 Some cases of inconsistent taint propagation (such as within hash
2360 values) have been fixed.
2364 The RE engine found in Perl 5.6.0 accidentally pessimised certain kinds
2365 of simple pattern matches. These are now handled better. [561]
2369 Regular expression debug output (whether through C<use re 'debug'>
2370 or via C<-Dr>) now looks better. [561]
2374 Multi-line matches like C<"a\nxb\n" =~ /(?!\A)x/m> were flawed. The
2375 bug has been fixed. [561]
2379 Use of $& could trigger a core dump under some situations. This
2380 is now avoided. [561]
2384 The regular expression captured submatches ($1, $2, ...) are now
2385 more consistently unset if the match fails, instead of leaving false
2386 data lying around in them. [561]
2390 readline() on files opened in "slurp" mode could return an extra
2391 "" (blank line) at the end in certain situations. This has been
2396 Autovivification of symbolic references of special variables described
2397 in L<perlvar> (as in C<${$num}>) was accidentally disabled. This works
2402 Sys::Syslog ignored the C<LOG_AUTH> constant.
2406 $AUTOLOAD, sort(), lock(), and spawning subprocesses
2407 in multiple threads simultaneously are now thread-safe.
2411 Tie::Array's SPLICE method was broken.
2415 Allow a read-only string on the left-hand side of a non-modifying tr///.
2419 If C<STDERR> is tied, warnings caused by C<warn> and C<die> now
2420 correctly pass to it.
2424 Several Unicode fixes.
2430 BOMs (byte order marks) at the beginning of Perl files
2431 (scripts, modules) should now be transparently skipped.
2432 UTF-16 and UCS-2 encoded Perl files should now be read correctly.
2436 The character tables have been updated to Unicode 3.2.0.
2440 Comparing with utf8 data does not magically upgrade non-utf8 data
2441 into utf8. (This was a problem for example if you were mixing data
2442 from I/O and Unicode data: your output might have got magically encoded
2447 Generating illegal Unicode code points such as U+FFFE, or the UTF-16
2448 surrogates, now also generates an optional warning.
2452 C<IsAlnum>, C<IsAlpha>, and C<IsWord> now match titlecase.
2456 Concatenation with the C<.> operator or via variable interpolation,
2457 C<eq>, C<substr>, C<reverse>, C<quotemeta>, the C<x> operator,
2458 substitution with C<s///>, single-quoted UTF8, should now work.
2462 The C<tr///> operator now works. Note that the C<tr///CU>
2463 functionality has been removed (but see pack('U0', ...)).
2467 C<eval "v200"> now works.
2471 Perl 5.6.0 parsed m/\x{ab}/ incorrectly, leading to spurious warnings.
2472 This has been corrected. [561]
2476 Zero entries were missing from the Unicode classes such as C<IsDigit>.
2482 Large unsigned numbers (those above 2**31) could sometimes lose their
2483 unsignedness, causing bogus results in arithmetic operations. [561]
2487 The Perl parser has been stress tested using both random input and
2488 Markov chain input and the few found crashes and lockups have been
2493 =head2 Platform Specific Changes and Fixes
2501 Perl now works on post-4.0 BSD/OSes.
2507 Setting C<$0> now works (as much as possible; see L<perlvar> for details).
2513 Numerous updates; currently synchronised with Cygwin 1.3.10.
2517 Previously DYNIX/ptx had problems in its Configure probe for non-blocking I/O.
2523 EPOC now better supported. See README.epoc. [561]
2529 Perl now works on post-3.0 FreeBSDs.
2535 README.hpux updated; C<Configure -Duse64bitall> now works;
2536 now uses HP-UX malloc instead of Perl malloc.
2542 Numerous compilation flag and hint enhancements; accidental mixing
2543 of 32-bit and 64-bit libraries (a doomed attempt) made much harder.
2553 Long doubles should now work (see INSTALL). [561]
2557 Linux previously had problems related to sockaddrlen when using
2558 accept(), recvfrom() (in Perl: recv()), getpeername(), and
2567 Compilation of the standard Perl distribution in Mac OS Classic should
2568 now work if you have the Metrowerks development environment and the
2569 missing Mac-specific toolkit bits. Contact the macperl mailing list
2576 MPE/iX update after Perl 5.6.0. See README.mpeix. [561]
2580 NetBSD/threads: try installing the GNU pth (should be in the
2581 packages collection, or http://www.gnu.org/software/pth/),
2582 and Configure with -Duseithreads.
2588 Perl now works on NetBSD/sparc.
2594 Now works with usethreads (see INSTALL). [561]
2600 64-bitness using the Sun Workshop compiler now works.
2606 The native build method requires at least VOS Release 14.5.0
2607 and GNU C++/GNU Tools 2.0.1 or later. The Perl pack function
2608 now maps overflowed values to +infinity and underflowed values
2613 Tru64 (aka Digital UNIX, aka DEC OSF/1)
2615 The operating system version letter now recorded in $Config{osvers}.
2616 Allow compiling with gcc (previously explicitly forbidden). Compiling
2617 with gcc still not recommended because buggy code results, even with
2624 Fixed various alignment problems that lead into core dumps either
2625 during build or later; no longer dies on math errors at runtime;
2626 now using full quad integers (64 bits), previously was using
2627 only 46 bit integers for speed.
2633 See L</"Socket Extension Dynamic in VMS"> and L</"IEEE-format Floating Point
2634 Default on OpenVMS Alpha"> for important changes not otherwise listed here.
2636 chdir() now works better despite a CRT bug; now works with MULTIPLICITY
2637 (see INSTALL); now works with Perl's malloc.
2639 The tainting of C<%ENV> elements via C<keys> or C<values> was previously
2640 unimplemented. It now works as documented.
2642 The C<waitpid> emulation has been improved. The worst bug (now fixed)
2643 was that a pid of -1 would cause a wildcard search of all processes on
2646 POSIX-style signals are now emulated much better on VMS versions prior
2649 The C<system> function and backticks operator have improved
2650 functionality and better error handling. [561]
2652 File access tests now use current process privileges rather than the
2653 user's default privileges, which could sometimes result in a mismatch
2654 between reported access and actual access. This improvement is only
2655 available on VMS v6.0 and later.
2657 There is a new C<kill> implementation based on C<sys$sigprc> that allows
2658 older VMS systems (pre-7.0) to use C<kill> to send signals rather than
2659 simply force exit. This implementation also allows later systems to
2660 call C<kill> from within a signal handler.
2662 Iterative logical name translations are now limited to 10 iterations in
2663 imitation of SHOW LOGICAL and other OpenVMS facilities.
2673 Signal handling now works better than it used to. It is now implemented
2674 using a Windows message loop, and is therefore less prone to random
2679 fork() emulation is now more robust, but still continues to have a few
2680 esoteric bugs and caveats. See L<perlfork> for details. [561+]
2684 A failed (pseudo)fork now returns undef and sets errno to EAGAIN. [561]
2688 The following modules now work on Windows:
2690 ExtUtils::Embed [561]
2697 IO::File::new_tmpfile() is no longer limited to 32767 invocations
2702 Better chdir() return value for a non-existent directory.
2706 Compiling perl using the 64-bit Platform SDK tools is now supported.
2710 The Win32::SetChildShowWindow() builtin can be used to control the
2711 visibility of windows created by child processes. See L<Win32> for
2716 Non-blocking waits for child processes (or pseudo-processes) are
2717 supported via C<waitpid($pid, &POSIX::WNOHANG)>.
2721 The behavior of system() with multiple arguments has been rationalized.
2722 Each unquoted argument will be automatically quoted to protect whitespace,
2723 and any existing whitespace in the arguments will be preserved. This
2724 improves the portability of system(@args) by avoiding the need for
2725 Windows C<cmd> shell specific quoting in perl programs.
2727 Note that this means that some scripts that may have relied on earlier
2728 buggy behavior may no longer work correctly. For example,
2729 C<system("nmake /nologo", @args)> will now attempt to run the file
2730 C<nmake /nologo> and will fail when such a file isn't found.
2731 On the other hand, perl will now execute code such as
2732 C<system("c:/Program Files/MyApp/foo.exe", @args)> correctly.
2736 The perl header files no longer suppress common warnings from the
2737 Microsoft Visual C++ compiler. This means that additional warnings may
2738 now show up when compiling XS code.
2742 Borland C++ v5.5 is now a supported compiler that can build Perl.
2743 However, the generated binaries continue to be incompatible with those
2744 generated by the other supported compilers (GCC and Visual C++). [561]
2748 Duping socket handles with open(F, ">&MYSOCK") now works under Windows 9x.
2753 Current directory entries in %ENV are now correctly propagated to child
2758 New %ENV entries now propagate to subprocesses. [561]
2762 Win32::GetCwd() correctly returns C:\ instead of C: when at the drive root.
2763 Other bugs in chdir() and Cwd::cwd() have also been fixed. [561]
2767 The makefiles now default to the features enabled in ActiveState ActivePerl
2768 (a popular Win32 binary distribution). [561]
2772 HTML files will now be installed in c:\perl\html instead of
2773 c:\perl\lib\pod\html
2777 REG_EXPAND_SZ keys are now allowed in registry settings used by perl. [561]
2781 Can now send() from all threads, not just the first one. [561]
2785 ExtUtils::MakeMaker now uses $ENV{LIB} to search for libraries. [561]
2789 Less stack reserved per thread so that more threads can run
2790 concurrently. (Still 16M per thread.) [561]
2794 C<< File::Spec->tmpdir() >> now prefers C:/temp over /tmp
2795 (works better when perl is running as service).
2799 Better UNC path handling under ithreads. [561]
2803 wait(), waitpid(), and backticks now return the correct exit status
2804 under Windows 9x. [561]
2808 A socket handle leak in accept() has been fixed. [561]
2814 =head1 New or Changed Diagnostics
2816 Please see L<perldiag> for more details.
2822 Ambiguous range in the transliteration operator (like a-z-9) now
2827 Two new debugging options have been added: if you have compiled your
2828 Perl with debugging, you can use the -DT [561] and -DR options to trace
2829 tokenising and to add reference counts to displaying variables,
2834 The lexical warnings category "deprecated" is no longer a sub-category
2835 of the "syntax" category. It is now a top-level category in its own
2840 Unadorned dump() will now give a warning suggesting to
2841 use explicit CORE::dump() if that's what really is meant.
2845 The "Unrecognized escape" warning has been extended to include C<\8>,
2846 C<\9>, and C<\_>. There is no need to escape any of the C<\w> characters.
2850 All regular expression compilation error messages are now hopefully
2851 easier to understand both because the error message now comes before
2852 the failed regex and because the point of failure is now clearly
2853 marked by a C<E<lt>-- HERE> marker.
2857 Various I/O (and socket) functions like binmode(), close(), and so
2858 forth now more consistently warn if they are used illogically either
2859 on a yet unopened or on an already closed filehandle (or socket).
2863 Using lstat() on a filehandle now gives a warning. (It's a non-sensical
2868 The C<-M> and C<-m> options now warn if you didn't supply the module name.
2872 If you in C<use> specify a required minimum version, modules matching
2873 the name and but not defining a $VERSION will cause a fatal failure.
2877 Using negative offset for vec() in lvalue context is now a warnable offense.
2881 Odd number of arguments to oveload::constant now elicits a warning.
2885 Odd number of elements to in anonymous hash now elicits a warning.
2889 The various "opened only for", "on closed", "never opened" warnings
2890 drop the C<main::> prefix for filehandles in the C<main> package,
2891 for example C<STDIN> instead of C<main::STDIN>.
2895 Subroutine prototypes are now checked more carefully, you may
2896 get warnings for example if you have used non-prototype characters.
2900 If an attempt to use a (non-blessed) reference as an array index
2901 is made, a warning is given.
2905 C<push @a;> and C<unshift @a;> (with no values to push or unshift)
2906 now give a warning. This may be a problem for generated and evaled
2911 If you try to L<perlfunc/pack> a number less than 0 or larger than 255
2912 using the C<"C"> format you will get an optional warning. Similarly
2913 for the C<"c"> format and a number less than -128 or more than 127.
2917 pack C<P> format now demands an explicit size.
2921 unpack C<w> now warns of unterminated compressed integers.
2925 Warnings relating to the use of PerlIO have been added.
2929 Certain regex modifiers such as C<(?o)> make sense only if applied to
2930 the entire regex. You will get an optional warning if you try to do
2935 Variable length lookbehind has not yet been implemented, trying to
2936 use it will tell that.
2940 Using arrays or hashes as references (e.g. C<< %foo->{bar} >>
2941 has been deprecated for a while. Now you will get an optional warning.
2945 Warnings relating to the use of the new restricted hashes feature
2950 Self-ties of arrays and hashes are not supported and fatal errors
2951 will happen even at an attempt to do so.
2955 Using C<sort> in scalar context now issues an optional warning.
2956 This didn't do anything useful, as the sort was not performed.
2960 Using the /g modifier in split() is meaningless and will cause a warning.
2964 Using splice() past the end of an array now causes a warning.
2968 Malformed Unicode encodings (UTF-8 and UTF-16) cause a lot of warnings,
2969 ad doestrying to use UTF-16 surrogates (which are unimplemented).
2973 Trying to use Unicode characters on an I/O stream without marking the
2974 stream's encoding (using open() or binmode()) will cause "Wide character"
2979 Use of v-strings in use/require causes a (backward) portability warning.
2983 Warnings relating to the use interpreter threads and their shared data
2988 =head1 Changed Internals
2994 PerlIO is now the default.
2998 perlapi.pod (a companion to perlguts) now attempts to document the
3003 You can now build a really minimal perl called microperl.
3004 Building microperl does not require even running Configure;
3005 C<make -f Makefile.micro> should be enough. Beware: microperl makes
3006 many assumptions, some of which may be too bold; the resulting
3007 executable may crash or otherwise misbehave in wondrous ways.
3008 For careful hackers only.
3012 Added rsignal(), whichsig(), do_join(), op_clear, op_null,
3013 ptr_table_clear(), ptr_table_free(), sv_setref_uv(), and several UTF-8
3014 interfaces to the publicised API. For the full list of the available
3015 APIs see L<perlapi>.
3019 Made possible to propagate customised exceptions via croak()ing.
3023 Now xsubs can have attributes just like subs. (Well, at least the
3024 built-in attributes.)
3028 dTHR and djSP have been obsoleted; the former removed (because it's
3029 a no-op) and the latter replaced with dSP.
3033 PERL_OBJECT has been completely removed.
3037 The MAGIC constants (e.g. C<'P'>) have been macrofied
3038 (e.g. C<PERL_MAGIC_TIED>) for better source code readability
3039 and maintainability.
3043 The regex compiler now maintains a structure that identifies nodes in
3044 the compiled bytecode with the corresponding syntactic features of the
3045 original regex expression. The information is attached to the new
3046 C<offsets> member of the C<struct regexp>. See L<perldebguts> for more
3047 complete information.
3051 The C code has been made much more C<gcc -Wall> clean. Some warning
3052 messages still remain in some platforms, so if you are compiling with
3053 gcc you may see some warnings about dubious practices. The warnings
3054 are being worked on.
3058 F<perly.c>, F<sv.c>, and F<sv.h> have now been extensively commented.
3062 Documentation on how to use the Perl source repository has been added
3063 to F<Porting/repository.pod>.
3067 There are now several profiling make targets.
3071 =head1 Security Vulnerability Closed [561]
3073 (This change was already made in 5.7.0 but bears repeating here.)
3074 (5.7.0 came out before 5.6.1: the development branch 5.7 released
3075 earlier than the maintenance branch 5.6)
3077 A potential security vulnerability in the optional suidperl component
3078 of Perl was identified in August 2000. suidperl is neither built nor
3079 installed by default. As of November 2001 the only known vulnerable
3080 platform is Linux, most likely all Linux distributions. CERT and
3081 various vendors and distributors have been alerted about the vulnerability.
3082 See http://www.cpan.org/src/5.0/sperl-2000-08-05/sperl-2000-08-05.txt
3083 for more information.
3085 The problem was caused by Perl trying to report a suspected security
3086 exploit attempt using an external program, /bin/mail. On Linux
3087 platforms the /bin/mail program had an undocumented feature which
3088 when combined with suidperl gave access to a root shell, resulting in
3089 a serious compromise instead of reporting the exploit attempt. If you
3090 don't have /bin/mail, or if you have 'safe setuid scripts', or if
3091 suidperl is not installed, you are safe.
3093 The exploit attempt reporting feature has been completely removed from
3094 Perl 5.8.0 (and the maintenance release 5.6.1, and it was removed also
3095 from all the Perl 5.7 releases), so that particular vulnerability
3096 isn't there anymore. However, further security vulnerabilities are,
3097 unfortunately, always possible. The suidperl functionality is most
3098 probably going to be removed in Perl 5.10. In any case, suidperl
3099 should only be used by security experts who know exactly what they are
3100 doing and why they are using suidperl instead of some other solution
3101 such as sudo ( see http://www.courtesan.com/sudo/ ).
3105 Several new tests have been added, especially for the F<lib> and
3106 F<ext> subsections. There are now about 69 000 individual tests
3107 (spread over about 700 test scripts), in the regression suite (5.6.1
3108 has about 11 700 tests, in 258 test scripts) The exact numbers depend
3109 on the platform and Perl configuration used. Many of the new tests
3110 are of course introduced by the new modules, but still in general Perl
3111 is now more thoroughly tested.
3113 Because of the large number of tests, running the regression suite
3114 will take considerably longer time than it used to: expect the suite
3115 to take up to 4-5 times longer to run than in perl 5.6. On a really
3116 fast machine you can hope to finish the suite in about 6-8 minutes
3119 The tests are now reported in a different order than in earlier Perls.
3120 (This happens because the test scripts from under t/lib have been moved
3121 to be closer to the library/extension they are testing.)
3123 =head1 Known Problems
3131 If using the AIX native make command, instead of just "make" issue
3132 "make all". In some setups the former has been known to spuriously
3133 also try to run "make install". Alternatively, you may want to use
3138 In AIX 4.2, Perl extensions that use C++ functions that use statics
3139 may have problems in that the statics are not getting initialized.
3140 In newer AIX releases, this has been solved by linking Perl with
3141 the libC_r library, but unfortunately in AIX 4.2 the said library
3142 has an obscure bug where the various functions related to time
3143 (such as time() and gettimeofday()) return broken values, and
3144 therefore in AIX 4.2 Perl is not linked against libC_r.
3148 vac 5.0.0.0 May Produce Buggy Code For Perl
3150 The AIX C compiler vac version 5.0.0.0 may produce buggy code,
3151 resulting in a few random tests failing when run as part of "make
3152 test", but when the failing tests are run by hand, they succeed.
3153 We suggest upgrading to at least vac version 5.0.1.0, that has been
3154 known to compile Perl correctly. "lslpp -L|grep vac.C" will tell
3155 you the vac version. See README.aix.
3159 If building threaded Perl, you may get compilation warning from pp_sys.c:
3161 "pp_sys.c", line 4651.39: 1506-280 (W) Function argument assignment between types "unsigned char*" and "const void*" is not allowed.
3163 This is harmless; it is caused by the getnetbyaddr() and getnetbyaddr_r()
3164 having slightly different types for their first argument.
3168 =head2 Alpha systems with old gccs fail several tests
3170 If you see op/pack, op/pat, op/regexp, or ext/Storable tests failing
3171 in a Linux/alpha or *BSD/Alpha, it's probably time to upgrade your gcc.
3172 gccs prior to 2.95.3 are definitely not good enough, and gcc 3.1 may
3173 be even better. (RedHat Linux/alpha with gcc 3.1 reported no problems,
3174 as did Linux 2.4.18 with gcc 2.95.4.) (In Tru64, it is preferable to
3175 use the bundled C compiler.)
3179 Perl 5.8.0 doesn't build in AmigaOS. It broke at some point during
3180 the ithreads work and we could not find Amiga experts to unbreak the
3181 problems. Perl 5.6.1 still works for AmigaOS (as does the the 5.7.2
3182 development release).
3186 The following tests fail on 5.8.0 Perl in BeOS Personal 5.03:
3188 t/op/lfs............................FAILED at test 17
3189 t/op/magic..........................FAILED at test 24
3190 ext/Fcntl/t/syslfs..................FAILED at test 17
3191 ext/File/Glob/t/basic...............FAILED at test 3
3192 ext/POSIX/t/sigaction...............FAILED at test 13
3193 ext/POSIX/t/waitpid.................FAILED at test 1
3195 See L<perlbeos> (README.beos) for more details.
3197 =head2 Cygwin "unable to remap"
3199 For example when building the Tk extension for Cygwin,
3200 you may get an error message saying "unable to remap".
3201 This is known problem with Cygwin, and a workaround is
3202 detailed in here: http://sources.redhat.com/ml/cygwin/2001-12/msg00894.html
3204 =head2 Cygwin ndbm tests fail on FAT
3206 One can build but not install (or test the build of) the NDBM_File
3207 on FAT filesystems. Installation (or build) on NTFS works fine.
3209 =head2 ext/threads/t/libc
3211 If this test fails, it indicates that your libc (C library) is not
3212 threadsafe. This particular test stress tests the localtime() call to
3213 find out whether it is threadsafe. See L<perlthrtut> for more information.
3215 =head2 FreeBSD built with ithreads coredumps reading large directories
3217 This is a known bug in FreeBSD's readdir_r() (see L<perlfreebsd>
3218 (README.freebsd)), which hopefully will be fixed in FreeBSD 4.6.
3220 =head2 FreeBSD Failing locale Test 117 For ISO 8859-15 Locales
3222 The ISO 8859-15 locales may fail the locale test 117 in FreeBSD.
3223 This is caused by the characters \xFF (y with diaeresis) and \xBE
3224 (Y with diaeresis) not behaving correctly when being matched
3225 case-insensitively. Apparently this problem has been fixed in
3226 the latest FreeBSD releases.
3227 ( http://www.freebsd.org/cgi/query-pr.cgi?pr=34308 )
3229 =head2 IRIX fails ext/List/Util/t/shuffle.t
3231 IRIX with MIPSpro 7.3.1.3m compiler may fail the said List::Util test
3232 by dumping core. This seems to be a compiler error since if compiled
3233 with gcc no core dump ensues, and no failures on the said test on any
3236 =head2 Modifying $_ Inside for(..)
3240 works without complaint. It shouldn't. (You should be able to
3241 modify only lvalue elements inside the loops.) You can see the
3242 correct behaviour by replacing the 1..5 with 1, 2, 3, 4, 5.
3244 =head2 mod_perl 1.26 Doesn't Build With Threaded Perl
3246 Use mod_perl 1.27 or higher.
3248 =head2 lib/ftmp-security tests warn 'system possibly insecure'
3250 Don't panic. Read the 'make test' section of INSTALL instead.
3252 =head2 HP-UX lib/posix Subtest 9 Fails When LP64-Configured
3254 If perl is configured with -Duse64bitall, the successful result of the
3255 subtest 10 of lib/posix may arrive before the successful result of the
3256 subtest 9, which confuses the test harness so much that it thinks the
3259 =head2 Linux with glibc 2.2.5 fails t/op/int subtest #6 with -Duse64bitint
3261 This is a known bug in the glibc 2.2.5 with long long integers.
3262 ( http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=65612 )
3264 =head2 Linux With Sfio Fails op/misc Test 48
3268 =head2 libwww-perl (LWP) fails base/date #51
3270 Use libwww-perl 5.65 or later.
3274 Please remember to set your environment variable LC_ALL to "C"
3275 (setenv LC_ALL C) before running "make test" to avoid a lot of
3276 warnings about the broken locales of Mac OS X.
3278 The following tests are known to fail in Mac OS X 10.1.5 because of
3279 buggy (old) implementations of Berkeley DB included in Mac OS X:
3281 Failed Test Stat Wstat Total Fail Failed List of Failed
3282 -------------------------------------------------------------------------
3283 ../ext/DB_File/t/db-btree.t 0 11 ?? ?? % ??
3284 ../ext/DB_File/t/db-recno.t 149 3 2.01% 61 63 65
3286 If you are building on a UFS partition, you will also probably see
3287 t/op/stat.t subtest #9 fail. This is caused by Darwin's UFS not
3288 supporting inode change time.
3290 Also the ext/POSIX/t/posix.t subtest #10 fails but it is skipped for
3291 now because the failure is Apple's fault, not Perl's (blocked signals
3294 If you Configure with ithreads, ext/threads/t/libc.t will fail. Again,
3295 this is not Perl's fault-- the libc of Mac OS X is not threadsafe
3296 (in this particular test, the localtime() call is found to be
3299 =head2 OS/2 Test Failures
3301 The following tests are known to fail on OS/2 (for clarity
3302 only the failures are shown, not the full error messages):
3304 t/io/utf8............................FAILED at test 19
3305 t/op/grent...........................FAILED at test 2
3306 t/op/pwent...........................FAILED at test 1
3307 t/lib/os2_base.......................FAILED at test 13
3308 t/lib/os2_process....................FAILED at test 10
3309 t/lib/os2_process_kid................FAILED at test 10
3310 t/lib/rx_cmprt.......................FAILED at test 16
3311 ext/DB_File/t/db-btree...............FAILED at test 0
3312 ext/DB_File/t/db-hash................FAILED at test 0
3313 ext/DB_File/t/db-recno...............FAILED at test 0
3314 lib/ExtUtils/t/basic.................FAILED at test 14
3315 lib/ExtUtils/t/Constant..............FAILED at test 4
3316 lib/Memoize/t/errors.................FAILED at test 4
3318 =head2 op/sprintf tests 91, 129, and 130
3320 The op/sprintf tests 91, 129, and 130 are known to fail on some platforms.
3321 Examples include any platform using sfio, and Compaq/Tandem's NonStop-UX.
3323 Test 91 is known to fail on QNX6 (nto), because C<sprintf '%e',0>
3324 incorrectly produces C<0.000000e+0> instead of C<0.000000e+00>.
3326 For tests 129 and 130, the failing platforms do not comply with
3327 the ANSI C Standard: lines 19ff on page 134 of ANSI X3.159 1989, to
3328 be exact. (They produce something other than "1" and "-1" when
3329 formatting 0.6 and -0.6 using the printf format "%.0f"; most often,
3330 they produce "0" and "-0".)
3334 In case you are still using Solaris 2.5 (aka SunOS 5.5), you may
3335 experience failures (the test core dumping) in lib/locale.t.
3336 The suggested cure is to upgrade your Solaris.
3338 =head2 Solaris x86 Fails Tests With -Duse64bitint
3340 The following tests are known to fail in Solaris x86 with Perl
3341 configured to use 64 bit integers:
3343 ext/Data/Dumper/t/dumper.............FAILED at test 268
3344 ext/Devel/Peek/Peek..................FAILED at test 7
3346 =head2 SUPER-UX (NEC SX)
3348 The following tests are known to fail on SUPER-UX:
3350 op/64bitint...........................FAILED tests 29-30, 32-33, 35-36
3351 op/arith..............................FAILED tests 128-130
3352 op/pack...............................FAILED tests 25-5625
3353 op/pow................................
3354 op/taint..............................# msgsnd failed
3355 ../ext/IO/lib/IO/t/io_poll............FAILED tests 3-4
3356 ../ext/IPC/SysV/ipcsysv...............FAILED tests 2, 5-6
3357 ../ext/IPC/SysV/t/msg.................FAILED tests 2, 4-6
3358 ../ext/Socket/socketpair..............FAILED tests 12
3359 ../lib/IPC/SysV.......................FAILED tests 2, 5-6
3360 ../lib/warnings.......................FAILED tests 115-116, 118-119
3362 The op/pack failure ("Cannot compress negative numbers at op/pack.t line 126")
3363 is serious but as of yet unsolved. It points at some problems with the
3364 signedness handling of the C compiler, as do the 64bitint, arith, and pow
3365 failures. Most of the rest point at problems with SysV IPC.
3367 =head2 PDL failing some tests
3369 Use PDL 2.3.4 or later.
3371 =head2 Term::ReadKey not working on Win32
3373 Use Term::ReadKey 2.20 or later.
3375 =head2 Failure of Thread (5.005-style) tests
3377 B<Note that support for 5.005-style threading is deprecated,
3378 experimental and practically unsupported. In 5.10, it is expected
3379 to be removed. You should migrate your code to ithreads.>
3381 The following tests are known to fail due to fundamental problems in
3382 the 5.005 threading implementation. These are not new failures--Perl
3383 5.005_0x has the same bugs, but didn't have these tests.
3385 ../ext/B/t/xref.t 255 65280 14 12 85.71% 3-14
3386 ../ext/List/Util/t/first.t 255 65280 7 4 57.14% 2 5-7
3387 ../lib/English.t 2 512 54 2 3.70% 2-3
3388 ../lib/FileCache.t 5 1 20.00% 5
3389 ../lib/Filter/Simple/t/data.t 6 3 50.00% 1-3
3390 ../lib/Filter/Simple/t/filter_only. 9 3 33.33% 1-2 5
3391 ../lib/Math/BigInt/t/bare_mbf.t 1627 4 0.25% 8 11 1626-1627
3392 ../lib/Math/BigInt/t/bigfltpm.t 1629 4 0.25% 10 13 1628-
3394 ../lib/Math/BigInt/t/sub_mbf.t 1633 4 0.24% 8 11 1632-1633
3395 ../lib/Math/BigInt/t/with_sub.t 1628 4 0.25% 9 12 1627-1628
3396 ../lib/Tie/File/t/31_autodefer.t 255 65280 65 32 49.23% 34-65
3397 ../lib/autouse.t 10 1 10.00% 4
3398 op/flip.t 15 1 6.67% 15
3400 These failures are unlikely to get fixed as 5.005-style threads
3401 are considered fundamentally broken. (Basically what happens is that
3402 competing threads can corrupt shared global state, one good example
3403 being regular expression engine's state.)
3405 =head2 Timing problems
3407 The following tests may fail intermittently because of timing
3408 problems, for example if the system is heavily loaded.
3411 ext/Time/HiRes/HiRes.t
3413 lib/Memoize/t/expmod_t.t
3414 lib/Memoize/t/speed.t
3416 In case of failure please try running them manually, for example
3418 ./perl -Ilib ext/Time/HiRes/HiRes.t
3420 =head2 Unicode in package/class and subroutine names does not work
3422 One can have Unicode in identifier names, but not in package/class or
3423 subroutine names. While some limited functionality towards this does
3424 exist as of Perl 5.8.0, that is more accidental than designed; use of
3425 Unicode for the said purposes is unsupported.
3427 One reason of this unfinishedness is its (currently) inherent
3428 unportability: since both package names and subroutine names may
3429 need to be mapped to file and directory names, the Unicode capability
3430 of the filesystem becomes important-- and there unfortunately aren't
3439 During Configure, the test
3441 Guessing which symbols your C compiler and preprocessor define...
3443 will probably fail with error messages like
3445 CC-20 cc: ERROR File = try.c, Line = 3
3446 The identifier "bad" is undefined.
3448 bad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K
3451 CC-65 cc: ERROR File = try.c, Line = 3
3452 A semicolon is expected at this point.
3454 This is caused by a bug in the awk utility of UNICOS/mk. You can ignore
3455 the error, but it does cause a slight problem: you cannot fully
3456 benefit from the h2ph utility (see L<h2ph>) that can be used to
3457 convert C headers to Perl libraries, mainly used to be able to access
3458 from Perl the constants defined using C preprocessor, cpp. Because of
3459 the above error, parts of the converted headers will be invisible.
3460 Luckily, these days the need for h2ph is rare.
3464 If building Perl with interpreter threads (ithreads), the
3465 getgrent(), getgrnam(), and getgrgid() functions cannot return the
3466 list of the group members due to a bug in the multithreaded support of
3467 UNICOS/mk. What this means is that in list context the functions will
3468 return only three values, not four.
3474 There are a few known test failures, see L<perluts> (README.uts).
3476 =head2 VOS (Stratus)
3478 When Perl is built using the native build process on VOS Release
3479 14.5.0 and GNU C++/GNU Tools 2.0.1, all attempted tests either
3480 pass or result in TODO (ignored) failures.
3484 There should be no reported test failures with a default configuration,
3485 though there are a number of tests marked TODO that point to areas
3486 needing further debugging and/or porting work.
3490 In multi-CPU boxes, there are some problems with the I/O buffering:
3491 some output may appear twice.
3493 =head2 XML::Parser not working
3495 Use XML::Parser 2.31 or later.
3497 =head2 z/OS (OS/390)
3499 z/OS has rather many test failures but the situation is actually
3500 better than it was in 5.6.0; it's just that so many new modules and
3501 tests have been added.
3503 Failed Test Stat Wstat Total Fail Failed List of Failed
3504 ---------------------------------------------------------------------------
3505 ../ext/Data/Dumper/t/dumper.t 357 8 2.24% 311 314 325 327
3507 ../ext/IO/lib/IO/t/io_unix.t 5 4 80.00% 2-5
3508 ../ext/Storable/t/downgrade.t 12 3072 169 12 7.10% 14-15 46-47 78-79
3510 ../lib/ExtUtils/t/Constant.t 121 30976 48 48 100.00% 1-48
3511 ../lib/ExtUtils/t/Embed.t 9 9 100.00% 1-9
3512 op/pat.t 0 11 922 283 30.69% 640-922
3513 op/sprintf.t 224 3 1.34% 98 100 136
3514 op/tr.t 97 5 5.15% 63 71-74
3515 uni/fold.t 780 6 0.77% 61 169 196 661
3518 The failures in dumper.t and downgrade.t are problems in the tests,
3519 those in io_unix and sprintf are problems in the USS (UDP sockets and
3520 printf formats). The pat, tr, and fold failures are genuine Perl
3521 problems caused by EBCDIC (and in the pat and fold cases, combining
3522 that with Unicode). The Constant and Embed are probably problems in
3523 the tests (since they test Perl's ability to build extensions, and
3524 that seems to be working reasonably well.)
3526 =head2 Localising Tied Arrays and Hashes Is Broken
3530 doesn't work as one would expect: the old value is restored
3531 incorrectly. This will be changed in a future release, but we don't
3532 know yet what the new semantics will exactly be. In any case, the
3533 change will break existing code that relies on the current
3534 (ill-defined) semantics, so just avoid doing this in general.
3536 =head2 Self-tying Problems
3538 Self-tying of arrays and hashes is broken in rather deep and
3539 hard-to-fix ways. As a stop-gap measure to avoid people from getting
3540 frustrated at the mysterious results (core dumps, most often), it is
3541 forbidden for now (you will get a fatal error even from an attempt).
3543 A change to self-tying of globs has caused them to be recursively
3544 referenced (see: L<perlobj/"Two-Phased Garbage Collection">). You
3545 will now need an explicit untie to destroy a self-tied glob. This
3546 behaviour may be fixed at a later date.
3548 Self-tying of scalars and IO thingies works.
3550 =head2 Tied/Magical Array/Hash Elements Do Not Autovivify
3552 For normal arrays C<$foo = \$bar[1]> will assign C<undef> to
3553 C<$bar[1]> (assuming that it didn't exist before), but for
3554 tied/magical arrays and hashes such autovivification does not happen
3555 because there is currently no way to catch the reference creation.
3556 The same problem affects slicing over non-existent indices/keys of
3557 a tied/magical array/hash.
3559 =head2 Building Extensions Can Fail Because Of Largefiles
3561 Some extensions like mod_perl are known to have issues with
3562 `largefiles', a change brought by Perl 5.6.0 in which file offsets
3563 default to 64 bits wide, where supported. Modules may fail to compile
3564 at all, or they may compile and work incorrectly. Currently, there
3565 is no good solution for the problem, but Configure now provides
3566 appropriate non-largefile ccflags, ldflags, libswanted, and libs
3567 in the %Config hash (e.g., $Config{ccflags_nolargefiles}) so the
3568 extensions that are having problems can try configuring themselves
3569 without the largefileness. This is admittedly not a clean solution,
3570 and the solution may not even work at all. One potential failure is
3571 whether one can (or, if one can, whether it's a good idea to) link
3572 together at all binaries with different ideas about file offsets;
3573 all this is platform-dependent.
3575 =head2 Unicode Support on EBCDIC Still Spotty
3577 Though mostly working, Unicode support still has problem spots on
3578 EBCDIC platforms. One such known spot are the C<\p{}> and C<\P{}>
3579 regular expression constructs for code points less than 256: the
3580 C<pP> are testing for Unicode code points, not knowing about EBCDIC.
3582 =head2 The Compiler Suite Is Still Very Experimental
3584 The compiler suite is slowly getting better but it continues to be
3585 highly experimental. Use in production environments is discouraged.
3587 =head2 The Long Double Support Is Still Experimental
3589 The ability to configure Perl's numbers to use "long doubles",
3590 floating point numbers of hopefully better accuracy, is still
3591 experimental. The implementations of long doubles are not yet
3592 widespread and the existing implementations are not quite mature
3593 or standardised, therefore trying to support them is a rare
3594 and moving target. The gain of more precision may also be offset
3595 by slowdown in computations (more bits to move around, and the
3596 operations are more likely to be executed by less optimised
3599 =head2 Seen In Perl 5.7 But Gone Now
3601 C<Time::Piece> (previously known as C<Time::Object>) was removed
3602 because it was felt that it didn't have enough value in it to be a
3603 core module. It is still a useful module, though, and is available
3606 Perl 5.8 unfortunately does not build anymore on AmigaOS; this broke
3607 accidentally at some point. Since there are not that many Amiga
3608 developers available, we could not get this fixed and tested in time
3609 for 5.8.0. Perl 5.6.1 still works for AmigaOS (as does the the 5.7.2
3610 development release).
3612 The C<PerlIO::Scalar> and C<PerlIO::Via> (capitalised) were renamed as
3613 C<PerlIO::scalar> and C<PerlIO::via> (all lowercase) just before 5.8.0.
3614 The main rationale was to have all core IO layers to have all
3615 lowercase names. The "plugins" are named as usual, for example
3616 C<PerlIO::via::QuotedPrint>.
3618 =head1 Reporting Bugs
3620 If you find what you think is a bug, you might check the articles
3621 recently posted to the comp.lang.perl.misc newsgroup and the perl
3622 bug database at http://bugs.perl.org/ . There may also be
3623 information at http://www.perl.com/ , the Perl Home Page.
3625 If you believe you have an unreported bug, please run the B<perlbug>
3626 program included with your release. Be sure to trim your bug down
3627 to a tiny but sufficient test case. Your bug report, along with the
3628 output of C<perl -V>, will be sent off to perlbug@perl.org to be
3629 analysed by the Perl porting team.
3633 The F<Changes> file for exhaustive details on what changed.
3635 The F<INSTALL> file for how to build Perl.
3637 The F<README> file for general stuff.
3639 The F<Artistic> and F<Copying> files for copyright information.
3643 Written by Jarkko Hietaniemi <F<jhi@iki.fi>>.