1PERL581DELTA(1) Perl Programmers Reference Guide PERL581DELTA(1)
2
6 perl581delta - what is new for perl v5.8.1
7
9 This document describes differences between the 5.8.0 release and the
10 5.8.1 release.
11 If you are upgrading from an earlier release such as 5.6.1, first read
13 the perl58delta, which describes differences between 5.6.0 and 5.8.0.
14 In case you are wondering about 5.6.1, it was bug-fix-wise rather
16 identical to the development release 5.7.1. Confused? This timeline
17 hopefully helps a bit: it lists the new major releases, their
18 maintenance releases, and the development releases.
19 New Maintenance Development
21 5.6.0 2000-Mar-22
23 5.7.0 2000-Sep-02
24 5.6.1 2001-Apr-08
25 5.7.1 2001-Apr-09
26 5.7.2 2001-Jul-13
27 5.7.3 2002-Mar-05
28 5.8.0 2002-Jul-18
29 5.8.1 2003-Sep-25
30
32 Hash Randomisation
33 Mainly due to security reasons, the "random ordering" of hashes has
34 been made even more random. Previously while the order of hash
35 elements from keys(), values(), and each() was essentially random, it
36 was still repeatable. Now, however, the order varies between different
37 runs of Perl.
38 Perl has never guaranteed any ordering of the hash keys, and the
40 ordering has already changed several times during the lifetime of Perl
41 5. Also, the ordering of hash keys has always been, and continues to
42 be, affected by the insertion order.
43 The added randomness may affect applications.
45 One possible scenario is when output of an application has included
47 hash data. For example, if you have used the Data::Dumper module to
48 dump data into different files, and then compared the files to see
49 whether the data has changed, now you will have false positives since
50 the order in which hashes are dumped will vary. In general the cure is
51 to sort the keys (or the values); in particular for Data::Dumper to use
52 the "Sortkeys" option. If some particular order is really important,
53 use tied hashes: for example the Tie::IxHash module which by default
54 preserves the order in which the hash elements were added.
55 More subtle problem is reliance on the order of "global destruction".
57 That is what happens at the end of execution: Perl destroys all data
58 structures, including user data. If your destructors (the DESTROY
59 subroutines) have assumed any particular ordering to the global
60 destruction, there might be problems ahead. For example, in a
61 destructor of one object you cannot assume that objects of any other
62 class are still available, unless you hold a reference to them. If the
63 environment variable PERL_DESTRUCT_LEVEL is set to a non-zero value, or
64 if Perl is exiting a spawned thread, it will also destruct the ordinary
65 references and the symbol tables that are no longer in use. You can't
66 call a class method or an ordinary function on a class that has been
67 collected that way.
68 The hash randomisation is certain to reveal hidden assumptions about
70 some particular ordering of hash elements, and outright bugs: it
71 revealed a few bugs in the Perl core and core modules.
72 To disable the hash randomisation in runtime, set the environment
74 variable PERL_HASH_SEED to 0 (zero) before running Perl (for more
75 information see "PERL_HASH_SEED" in perlrun), or to disable the feature
76 completely in compile time, compile with "-DNO_HASH_SEED" (see
77 INSTALL).
78 See "Algorithmic Complexity Attacks" in perlsec for the original
80 rationale behind this change.
81 UTF-8 On Filehandles No Longer Activated By Locale
83 In Perl 5.8.0 all filehandles, including the standard filehandles, were
84 implicitly set to be in Unicode UTF-8 if the locale settings indicated
85 the use of UTF-8. This feature caused too many problems, so the
86 feature was turned off and redesigned: see "Core Enhancements".
87 Single-number v-strings are no longer v-strings before "=>"
89 The version strings or v-strings (see "Version Strings" in perldata)
90 feature introduced in Perl 5.6.0 has been a source of some confusion--
91 especially when the user did not want to use it, but Perl thought it
92 knew better. Especially troublesome has been the feature that before a
93 "=>" a version string (a "v" followed by digits) has been interpreted
94 as a v-string instead of a string literal. In other words:
95 %h = ( v65 => 42 );
97 has meant since Perl 5.6.0
99 %h = ( 'A' => 42 );
101 (at least in platforms of ASCII progeny) Perl 5.8.1 restores the more
103 natural interpretation
104 %h = ( 'v65' => 42 );
106 The multi-number v-strings like v65.66 and 65.66.67 still continue to
108 be v-strings in Perl 5.8.
109 (Win32) The -C Switch Has Been Repurposed
111 The -C switch has changed in an incompatible way. The old semantics of
112 this switch only made sense in Win32 and only in the "use utf8"
113 universe in 5.6.x releases, and do not make sense for the Unicode
114 implementation in 5.8.0. Since this switch could not have been used by
115 anyone, it has been repurposed. The behavior that this switch enabled
116 in 5.6.x releases may be supported in a transparent, data-dependent
117 fashion in a future release.
118 For the new life of this switch, see "UTF-8 no longer default under
120 UTF-8 locales", and "-C" in perlrun.
121 (Win32) The /d Switch Of cmd.exe
123 Perl 5.8.1 uses the /d switch when running the cmd.exe shell internally
124 for system(), backticks, and when opening pipes to external programs.
125 The extra switch disables the execution of AutoRun commands from the
126 registry, which is generally considered undesirable when running
127 external programs. If you wish to retain compatibility with the older
128 behavior, set PERL5SHELL in your environment to "cmd /x/c".
129
131 UTF-8 no longer default under UTF-8 locales
132 In Perl 5.8.0 many Unicode features were introduced. One of them was
133 found to be of more nuisance than benefit: the automagic (and silent)
134 "UTF-8-ification" of filehandles, including the standard filehandles,
135 if the user's locale settings indicated use of UTF-8.
136 For example, if you had "en_US.UTF-8" as your locale, your STDIN and
138 STDOUT were automatically "UTF-8", in other words an implicit
139 binmode(..., ":utf8") was made. This meant that trying to print, say,
140 chr(0xff), ended up printing the bytes 0xc3 0xbf. Hardly what you had
141 in mind unless you were aware of this feature of Perl 5.8.0. The
142 problem is that the vast majority of people weren't: for example in
143 RedHat releases 8 and 9 the default locale setting is UTF-8, so all
144 RedHat users got UTF-8 filehandles, whether they wanted it or not. The
145 pain was intensified by the Unicode implementation of Perl 5.8.0
146 (still) having nasty bugs, especially related to the use of s/// and
147 tr///. (Bugs that have been fixed in 5.8.1)
148 Therefore a decision was made to backtrack the feature and change it
150 from implicit silent default to explicit conscious option. The new
151 Perl command line option "-C" and its counterpart environment variable
152 PERL_UNICODE can now be used to control how Perl and Unicode interact
153 at interfaces like I/O and for example the command line arguments. See
154 "-C" in perlrun and "PERL_UNICODE" in perlrun for more information.
155 Unsafe signals again available
157 In Perl 5.8.0 the so-called "safe signals" were introduced. This means
158 that Perl no longer handles signals immediately but instead "between
159 opcodes", when it is safe to do so. The earlier immediate handling
160 easily could corrupt the internal state of Perl, resulting in
161 mysterious crashes.
162 However, the new safer model has its problems too. Because now an
164 opcode, a basic unit of Perl execution, is never interrupted but
165 instead let to run to completion, certain operations that can take a
166 long time now really do take a long time. For example, certain network
167 operations have their own blocking and timeout mechanisms, and being
168 able to interrupt them immediately would be nice.
169 Therefore perl 5.8.1 introduces a "backdoor" to restore the pre-5.8.0
171 (pre-5.7.3, really) signal behaviour. Just set the environment
172 variable PERL_SIGNALS to "unsafe", and the old immediate (and unsafe)
173 signal handling behaviour returns. See "PERL_SIGNALS" in perlrun and
174 "Deferred Signals (Safe Signals)" in perlipc.
175 In completely unrelated news, you can now use safe signals with
177 POSIX::SigAction. See "POSIX::SigAction" in POSIX.
178 Tied Arrays with Negative Array Indices
180 Formerly, the indices passed to "FETCH", "STORE", "EXISTS", and
181 "DELETE" methods in tied array class were always non-negative. If the
182 actual argument was negative, Perl would call FETCHSIZE implicitly and
183 add the result to the index before passing the result to the tied array
184 method. This behaviour is now optional. If the tied array class
185 contains a package variable named $NEGATIVE_INDICES which is set to a
186 true value, negative values will be passed to "FETCH", "STORE",
187 "EXISTS", and "DELETE" unchanged.
188 local ${$x}
190 The syntaxes
191 local ${$x}
193 local @{$x}
194 local %{$x}
195 now do localise variables, given that the $x is a valid variable name.
197 Unicode Character Database 4.0.0
199 The copy of the Unicode Character Database included in Perl 5.8 has
200 been updated to 4.0.0 from 3.2.0. This means for example that the
201 Unicode character properties are as in Unicode 4.0.0.
202 Deprecation Warnings
204 There is one new feature deprecation. Perl 5.8.0 forgot to add some
205 deprecation warnings, these warnings have now been added. Finally, a
206 reminder of an impending feature removal.
207 (Reminder) Pseudo-hashes are deprecated (really)
209 Pseudo-hashes were deprecated in Perl 5.8.0 and will be removed in Perl
211 5.10.0, see perl58delta for details. Each attempt to access pseudo-
212 hashes will trigger the warning "Pseudo-hashes are deprecated". If you
213 really want to continue using pseudo-hashes but not to see the
214 deprecation warnings, use:
215 no warnings 'deprecated';
217 Or you can continue to use the fields pragma, but please don't expect
219 the data structures to be pseudohashes any more.
220 (Reminder) 5.005-style threads are deprecated (really)
222 5.005-style threads (activated by "use Thread;") were deprecated in
224 Perl 5.8.0 and will be removed after Perl 5.8, see perl58delta for
225 details. Each 5.005-style thread creation will trigger the warning
226 "5.005 threads are deprecated". If you really want to continue using
227 the 5.005 threads but not to see the deprecation warnings, use:
228 no warnings 'deprecated';
230 (Reminder) The $* variable is deprecated (really)
232 The $* variable controlling multi-line matching has been deprecated and
234 will be removed after 5.8. The variable has been deprecated for a long
235 time, and a deprecation warning "Use of $* is deprecated" is given, now
236 the variable will just finally be removed. The functionality has been
237 supplanted by the "/s" and "/m" modifiers on pattern matching. If you
238 really want to continue using the $*-variable but not to see the
239 deprecation warnings, use:
240 no warnings 'deprecated';
242 Miscellaneous Enhancements
244 "map" in void context is no longer expensive. "map" is now context
245 aware, and will not construct a list if called in void context.
246 If a socket gets closed by the server while printing to it, the client
248 now gets a SIGPIPE. While this new feature was not planned, it fell
249 naturally out of PerlIO changes, and is to be considered an accidental
250 feature.
251 PerlIO::get_layers(FH) returns the names of the PerlIO layers active on
253 a filehandle.
254 PerlIO::via layers can now have an optional UTF8 method to indicate
256 whether the layer wants to "auto-:utf8" the stream.
257 utf8::is_utf8() has been added as a quick way to test whether a scalar
259 is encoded internally in UTF-8 (Unicode).
260
262 Updated Modules And Pragmata
263 The following modules and pragmata have been updated since Perl 5.8.0:
264 base
266 B::Bytecode
267 In much better shape than it used to be. Still far from perfect,
268 but maybe worth a try.
269 B::Concise
271 B::Deparse
272 Benchmark
273 An optional feature, ":hireswallclock", now allows for high
274 resolution wall clock times (uses Time::HiRes).
275 ByteLoader
277 See B::Bytecode.
278 bytes
280 Now has bytes::substr.
281 CGI
283 charnames
284 One can now have custom character name aliases.
285 CPAN
287 There is now a simple command line frontend to the CPAN.pm module
288 called cpan.
289 Data::Dumper
291 A new option, Pair, allows choosing the separator between hash keys
292 and values.
293 DB_File
295 Devel::PPPort
296 Digest::MD5
297 Encode
298 Significant updates on the encoding pragma functionality (tr/// and
299 the DATA filehandle, formats).
300 If a filehandle has been marked as to have an encoding, unmappable
302 characters are detected already during input, not later (when the
303 corrupted data is being used).
304 The ISO 8859-6 conversion table has been corrected (the 0x30..0x39
306 erroneously mapped to U+0660..U+0669, instead of U+0030..U+0039).
307 The GSM 03.38 conversion did not handle escape sequences correctly.
308 The UTF-7 encoding has been added (making Encode feature-complete
309 with Unicode::String).
310 fields
312 libnet
313 Math::BigInt
314 A lot of bugs have been fixed since v1.60, the version included in
315 Perl v5.8.0. Especially noteworthy are the bug in Calc that caused
316 div and mod to fail for some large values, and the fixes to the
317 handling of bad inputs.
318 Some new features were added, e.g. the broot() method, you can now
320 pass parameters to config() to change some settings at runtime, and
321 it is now possible to trap the creation of NaN and infinity.
322 As usual, some optimizations took place and made the math overall a
324 tad faster. In some cases, quite a lot faster, actually. Especially
325 alternative libraries like Math::BigInt::GMP benefit from this. In
326 addition, a lot of the quite clunky routines like fsqrt() and
327 flog() are now much much faster.
328 MIME::Base64
330 NEXT
331 Diamond inheritance now works.
332 Net::Ping
334 PerlIO::scalar
335 Reading from non-string scalars (like the special variables, see
336 perlvar) now works.
337 podlators
339 Pod::LaTeX
340 PodParsers
341 Pod::Perldoc
342 Complete rewrite. As a side-effect, no longer refuses to startup
343 when run by root.
344 Scalar::Util
346 New utilities: refaddr, isvstring, looks_like_number,
347 set_prototype.
348 Storable
350 Can now store code references (via B::Deparse, so not foolproof).
351 strict
353 Earlier versions of the strict pragma did not check the parameters
354 implicitly passed to its "import" (use) and "unimport" (no)
355 routine. This caused the false idiom such as:
356 use strict qw(@ISA);
358 @ISA = qw(Foo);
359 This however (probably) raised the false expectation that the
361 strict refs, vars and subs were being enforced (and that @ISA was
362 somehow "declared"). But the strict refs, vars, and subs are not
363 enforced when using this false idiom.
364 Starting from Perl 5.8.1, the above will cause an error to be
366 raised. This may cause programs which used to execute seemingly
367 correctly without warnings and errors to fail when run under 5.8.1.
368 This happens because
369 use strict qw(@ISA);
371 will now fail with the error:
373 Unknown 'strict' tag(s) '@ISA'
375 The remedy to this problem is to replace this code with the correct
377 idiom:
378 use strict;
380 use vars qw(@ISA);
381 @ISA = qw(Foo);
382 Term::ANSIcolor
384 Test::Harness
385 Now much more picky about extra or missing output from test
386 scripts.
387 Test::More
389 Test::Simple
390 Text::Balanced
391 Time::HiRes
392 Use of nanosleep(), if available, allows mixing subsecond sleeps
393 with alarms.
394 threads
396 Several fixes, for example for join() problems and memory leaks.
397 In some platforms (like Linux) that use glibc the minimum memory
398 footprint of one ithread has been reduced by several hundred
399 kilobytes.
400 threads::shared
402 Many memory leaks have been fixed.
403 Unicode::Collate
405 Unicode::Normalize
406 Win32::GetFolderPath
407 Win32::GetOSVersion
408 Now returns extra information.
409
411 The "h2xs" utility now produces a more modern layout:
412 Foo-Bar/lib/Foo/Bar.pm instead of Foo/Bar/Bar.pm. Also, the
413 boilerplate test is now called t/Foo-Bar.t instead of t/1.t.
414 The Perl debugger (lib/perl5db.pl) has now been extensively documented
416 and bugs found while documenting have been fixed.
417 "perldoc" has been rewritten from scratch to be more robust and feature
419 rich.
420 "perlcc -B" works now at least somewhat better, while "perlcc -c" is
422 rather more broken. (The Perl compiler suite as a whole continues to
423 be experimental.)
424
426 perl573delta has been added to list the differences between the (now
427 quite obsolete) development releases 5.7.2 and 5.7.3.
428 perl58delta has been added: it is the perldelta of 5.8.0, detailing the
430 differences between 5.6.0 and 5.8.0.
431 perlartistic has been added: it is the Artistic License in pod format,
433 making it easier for modules to refer to it.
434 perlcheat has been added: it is a Perl cheat sheet.
436 perlgpl has been added: it is the GNU General Public License in pod
438 format, making it easier for modules to refer to it.
439 perlmacosx has been added to tell about the installation and use of
441 Perl in Mac OS X.
442 perlos400 has been added to tell about the installation and use of Perl
444 in OS/400 PASE.
445 perlreref has been added: it is a regular expressions quick reference.
447
449 The Unix standard Perl location, /usr/bin/perl, is no longer
450 overwritten by default if it exists. This change was very prudent
451 because so many Unix vendors already provide a /usr/bin/perl, but
452 simultaneously many system utilities may depend on that exact version
453 of Perl, so better not to overwrite it.
454 One can now specify installation directories for site and vendor man
456 and HTML pages, and site and vendor scripts. See INSTALL.
457 One can now specify a destination directory for Perl installation by
459 specifying the DESTDIR variable for "make install". (This feature is
460 slightly different from the previous "Configure -Dinstallprefix=...".)
461 See INSTALL.
462 gcc versions 3.x introduced a new warning that caused a lot of noise
464 during Perl compilation: "gcc -Ialreadyknowndirectory (warning:
465 changing search order)". This warning has now been avoided by
466 Configure weeding out such directories before the compilation.
467 One can now build subsets of Perl core modules by using the Configure
469 flags "-Dnoextensions=..." and "-Donlyextensions=...", see INSTALL.
470 Platform-specific enhancements
472 In Cygwin Perl can now be built with threads ("Configure
473 -Duseithreads"). This works with both Cygwin 1.3.22 and Cygwin 1.5.3.
474 In newer FreeBSD releases Perl 5.8.0 compilation failed because of
476 trying to use malloc.h, which in FreeBSD is just a dummy file, and a
477 fatal error to even try to use. Now malloc.h is not used.
478 Perl is now known to build also in Hitachi HI-UXMPP.
480 Perl is now known to build again in LynxOS.
482 Mac OS X now installs with Perl version number embedded in installation
484 directory names for easier upgrading of user-compiled Perl, and the
485 installation directories in general are more standard. In other words,
486 the default installation no longer breaks the Apple-provided Perl. On
487 the other hand, with "Configure -Dprefix=/usr" you can now really
488 replace the Apple-supplied Perl (please be careful).
489 Mac OS X now builds Perl statically by default. This change was done
491 mainly for faster startup times. The Apple-provided Perl is still
492 dynamically linked and shared, and you can enable the sharedness for
493 your own Perl builds by "Configure -Duseshrplib".
494 Perl has been ported to IBM's OS/400 PASE environment. The best way to
496 build a Perl for PASE is to use an AIX host as a cross-compilation
497 environment. See README.os400.
498 Yet another cross-compilation option has been added: now Perl builds on
500 OpenZaurus, a Linux distribution based on Mandrake + Embedix for the
501 Sharp Zaurus PDA. See the Cross/README file.
502 Tru64 when using gcc 3 drops the optimisation for toke.c to "-O2"
504 because of gigantic memory use with the default "-O3".
505 Tru64 can now build Perl with the newer Berkeley DBs.
507 Building Perl on WinCE has been much enhanced, see README.ce and
509 README.perlce.
510
512 Closures, eval and lexicals
513 There have been many fixes in the area of anonymous subs, lexicals and
514 closures. Although this means that Perl is now more "correct", it is
515 possible that some existing code will break that happens to rely on the
516 faulty behaviour. In practice this is unlikely unless your code
517 contains a very complex nesting of anonymous subs, evals and lexicals.
518 Generic fixes
520 If an input filehandle is marked ":utf8" and Perl sees illegal UTF-8
521 coming in when doing "<FH>", if warnings are enabled a warning is
522 immediately given - instead of being silent about it and Perl being
523 unhappy about the broken data later. (The :encoding(utf8) layer also
524 works the same way.)
525 binmode(SOCKET, ":utf8") only worked on the input side, not on the
527 output side of the socket. Now it works both ways.
528 For threaded Perls certain system database functions like getpwent()
530 and getgrent() now grow their result buffer dynamically, instead of
531 failing. This means that at sites with lots of users and groups the
532 functions no longer fail by returning only partial results.
533 Perl 5.8.0 had accidentally broken the capability for users to define
535 their own uppercase<->lowercase Unicode mappings (as advertised by the
536 Camel). This feature has been fixed and is also documented better.
537 In 5.8.0 this
539 $some_unicode .= <FH>;
541 didn't work correctly but instead corrupted the data. This has now
543 been fixed.
544 Tied methods like FETCH etc. may now safely access tied values, i.e.
546 resulting in a recursive call to FETCH etc. Remember to break the
547 recursion, though.
548 At startup Perl blocks the SIGFPE signal away since there isn't much
550 Perl can do about it. Previously this blocking was in effect also for
551 programs executed from within Perl. Now Perl restores the original
552 SIGFPE handling routine, whatever it was, before running external
553 programs.
554 Linenumbers in Perl scripts may now be greater than 65536, or 2**16.
556 (Perl scripts have always been able to be larger than that, it's just
557 that the linenumber for reported errors and warnings have "wrapped
558 around".) While scripts that large usually indicate a need to rethink
559 your code a bit, such Perl scripts do exist, for example as results
560 from generated code. Now linenumbers can go all the way to 4294967296,
561 or 2**32.
562 Platform-specific fixes
564 Linux
565 • Setting $0 works again (with certain limitations that Perl cannot
567 do much about: see "$0" in perlvar)
568 HP-UX
570 • Setting $0 now works.
572 VMS
574 • Configuration now tests for the presence of poll(), and IO::Poll
576 now uses the vendor-supplied function if detected.
577 • A rare access violation at Perl start-up could occur if the Perl
579 image was installed with privileges or if there was an identifier
580 with the subsystem attribute set in the process's rightslist.
581 Either of these circumstances triggered tainting code that
582 contained a pointer bug. The faulty pointer arithmetic has been
583 fixed.
584 • The length limit on values (not keys) in the %ENV hash has been
586 raised from 255 bytes to 32640 bytes (except when the
587 PERL_ENV_TABLES setting overrides the default use of logical names
588 for %ENV). If it is necessary to access these long values from
589 outside Perl, be aware that they are implemented using search list
590 logical names that store the value in pieces, each 255-byte piece
591 (up to 128 of them) being an element in the search list. When doing
592 a lookup in %ENV from within Perl, the elements are combined into a
593 single value. The existing VMS-specific ability to access
594 individual elements of a search list logical name via the
595 $ENV{'foo;N'} syntax (where N is the search list index) is
596 unimpaired.
597 • The piping implementation now uses local rather than global DCL
599 symbols for inter-process communication.
600 • File::Find could become confused when navigating to a relative
602 directory whose name collided with a logical name. This problem
603 has been corrected by adding directory syntax to relative path
604 names, thus preventing logical name translation.
605 Win32
607 • A memory leak in the fork() emulation has been fixed.
609 • The return value of the ioctl() built-in function was accidentally
611 broken in 5.8.0. This has been corrected.
612 • The internal message loop executed by perl during blocking
614 operations sometimes interfered with messages that were external to
615 Perl. This often resulted in blocking operations terminating
616 prematurely or returning incorrect results, when Perl was executing
617 under environments that could generate Windows messages. This has
618 been corrected.
619 • Pipes and sockets are now automatically in binary mode.
621 • The four-argument form of select() did not preserve $! (errno)
623 properly when there were errors in the underlying call. This is
624 now fixed.
625 • The "CR CR LF" problem of has been fixed, binmode(FH, ":crlf") is
627 now effectively a no-op.
628
630 All the warnings related to pack() and unpack() were made more
631 informative and consistent.
632 Changed "A thread exited while %d threads were running"
634 The old version
635 A thread exited while %d other threads were still running
637 was misleading because the "other" included also the thread giving the
639 warning.
640 Removed "Attempt to clear a restricted hash"
642 It is not illegal to clear a restricted hash, so the warning was
643 removed.
644 New "Illegal declaration of anonymous subroutine"
646 You must specify the block of code for "sub".
647 Changed "Invalid range "%s" in transliteration operator"
649 The old version
650 Invalid [] range "%s" in transliteration operator
652 was simply wrong because there are no "[] ranges" in tr///.
654 New "Missing control char name in \c"
656 Self-explanatory.
657 New "Newline in left-justified string for %s"
659 The padding spaces would appear after the newline, which is probably
660 not what you had in mind.
661 New "Possible precedence problem on bitwise %c operator"
663 If you think this
664 $x & $y == 0
666 tests whether the bitwise AND of $x and $y is zero, you will like this
668 warning.
669 New "Pseudo-hashes are deprecated"
671 This warning should have been already in 5.8.0, since they are.
672 New "read() on %s filehandle %s"
674 You cannot read() (or sysread()) from a closed or unopened filehandle.
675 New "5.005 threads are deprecated"
677 This warning should have been already in 5.8.0, since they are.
678 New "Tied variable freed while still in use"
680 Something pulled the plug on a live tied variable, Perl plays safe by
681 bailing out.
682 New "To%s: illegal mapping '%s'"
684 An illegal user-defined Unicode casemapping was specified.
685 New "Use of freed value in iteration"
687 Something modified the values being iterated over. This is not good.
688
690 These news matter to you only if you either write XS code or like to
691 know about or hack Perl internals (using Devel::Peek or any of the
692 "B::" modules counts), or like to run Perl with the "-D" option.
693 The embedding examples of perlembed have been reviewed to be up to date
695 and consistent: for example, the correct use of PERL_SYS_INIT3() and
696 PERL_SYS_TERM().
697 Extensive reworking of the pad code (the code responsible for lexical
699 variables) has been conducted by Dave Mitchell.
700 Extensive work on the v-strings by John Peacock.
702 UTF-8 length and position cache: to speed up the handling of Unicode
704 (UTF-8) scalars, a cache was introduced. Potential problems exist if
705 an extension bypasses the official APIs and directly modifies the PV of
706 an SV: the UTF-8 cache does not get cleared as it should.
707 APIs obsoleted in Perl 5.8.0, like sv_2pv, sv_catpvn, sv_catsv,
709 sv_setsv, are again available.
710 Certain Perl core C APIs like cxinc and regatom are no longer available
712 at all to code outside the Perl core of the Perl core extensions. This
713 is intentional. They never should have been available with the shorter
714 names, and if you application depends on them, you should (be ashamed
715 and) contact perl5-porters to discuss what are the proper APIs.
716 Certain Perl core C APIs like "Perl_list" are no longer available
718 without their "Perl_" prefix. If your XS module stops working because
719 some functions cannot be found, in many cases a simple fix is to add
720 the "Perl_" prefix to the function and the thread context "aTHX_" as
721 the first argument of the function call. This is also how it should
722 always have been done: letting the Perl_-less forms to leak from the
723 core was an accident. For cleaner embedding you can also force this
724 for all APIs by defining at compile time the cpp define
725 PERL_NO_SHORT_NAMES.
726 Perl_save_bool() has been added.
728 Regexp objects (those created with "qr") now have S-magic rather than
730 R-magic. This fixed regexps of the form /...(??{...;$x})/ to no longer
731 ignore changes made to $x. The S-magic avoids dropping the caching
732 optimization and making (??{...}) constructs obscenely slow (and
733 consequently useless). See also "Magic Variables" in perlguts.
734 Regexp::Copy was affected by this change.
735 The Perl internal debugging macros DEBUG() and DEB() have been renamed
737 to PERL_DEBUG() and PERL_DEB() to avoid namespace conflicts.
738 "-DL" removed (the leaktest had been broken and unsupported for years,
740 use alternative debugging mallocs or tools like valgrind and Purify).
741 Verbose modifier "v" added for "-DXv" and "-Dsv", see perlrun.
743
745 In Perl 5.8.0 there were about 69000 separate tests in about 700 test
746 files, in Perl 5.8.1 there are about 77000 separate tests in about 780
747 test files. The exact numbers depend on the Perl configuration and on
748 the operating system platform.
749
751 The hash randomisation mentioned in "Incompatible Changes" is
752 definitely problematic: it will wake dormant bugs and shake out bad
753 assumptions.
754 If you want to use mod_perl 2.x with Perl 5.8.1, you will need
756 mod_perl-1.99_10 or higher. Earlier versions of mod_perl 2.x do not
757 work with the randomised hashes. (mod_perl 1.x works fine.) You will
758 also need Apache::Test 1.04 or higher.
759 Many of the rarer platforms that worked 100% or pretty close to it with
761 perl 5.8.0 have been left a little bit untended since their maintainers
762 have been otherwise busy lately, and therefore there will be more
763 failures on those platforms. Such platforms include Mac OS Classic,
764 IBM z/OS (and other EBCDIC platforms), and NetWare. The most common
765 Perl platforms (Unix and Unix-like, Microsoft platforms, and VMS) have
766 large enough testing and expert population that they are doing well.
767 Tied hashes in scalar context
769 Tied hashes do not currently return anything useful in scalar context,
770 for example when used as boolean tests:
771 if (%tied_hash) { ... }
773 The current nonsensical behaviour is always to return false, regardless
775 of whether the hash is empty or has elements.
776 The root cause is that there is no interface for the implementors of
778 tied hashes to implement the behaviour of a hash in scalar context.
779 Net::Ping 450_service and 510_ping_udp failures
781 The subtests 9 and 18 of lib/Net/Ping/t/450_service.t, and the subtest
782 2 of lib/Net/Ping/t/510_ping_udp.t might fail if you have an unusual
783 networking setup. For example in the latter case the test is trying to
784 send a UDP ping to the IP address 127.0.0.1.
785 B::C
787 The C-generating compiler backend B::C (the frontend being "perlcc -c")
788 is even more broken than it used to be because of the extensive lexical
789 variable changes. (The good news is that B::Bytecode and ByteLoader
790 are better than they used to be.)
791
793 EBCDIC Platforms
794 IBM z/OS and other EBCDIC platforms continue to be problematic
795 regarding Unicode support. Many Unicode tests are skipped when they
796 really should be fixed.
797 Cygwin 1.5 problems
799 In Cygwin 1.5 the io/tell and op/sysio tests have failures for some yet
800 unknown reason. In 1.5.5 the threads tests stress_cv, stress_re, and
801 stress_string are failing unless the environment variable PERLIO is set
802 to "perlio" (which makes also the io/tell failure go away).
803 Perl 5.8.1 does build and work well with Cygwin 1.3: with (uname -a)
805 "CYGWIN_NT-5.0 ... 1.3.22(0.78/3/2) 2003-03-18 09:20 i686 ..." a 100%
806 "make test" was achieved with "Configure -des -Duseithreads".
807 HP-UX: HP cc warnings about sendfile and sendpath
809 With certain HP C compiler releases (e.g. B.11.11.02) you will get many
810 warnings like this (lines wrapped for easier reading):
811 cc: "/usr/include/sys/socket.h", line 504: warning 562:
813 Redeclaration of "sendfile" with a different storage class specifier:
814 "sendfile" will have internal linkage.
815 cc: "/usr/include/sys/socket.h", line 505: warning 562:
816 Redeclaration of "sendpath" with a different storage class specifier:
817 "sendpath" will have internal linkage.
818 The warnings show up both during the build of Perl and during certain
820 lib/ExtUtils tests that invoke the C compiler. The warning, however,
821 is not serious and can be ignored.
822 IRIX: t/uni/tr_7jis.t falsely failing
824 The test t/uni/tr_7jis.t is known to report failure under 'make test'
825 or the test harness with certain releases of IRIX (at least IRIX 6.5
826 and MIPSpro Compilers Version 7.3.1.1m), but if run manually the test
827 fully passes.
828 Mac OS X: no usemymalloc
830 The Perl malloc ("-Dusemymalloc") does not work at all in Mac OS X.
831 This is not that serious, though, since the native malloc works just
832 fine.
833 Tru64: No threaded builds with GNU cc (gcc)
835 In the latest Tru64 releases (e.g. v5.1B or later) gcc cannot be used
836 to compile a threaded Perl (-Duseithreads) because the system
837 "<pthread.h>" file doesn't know about gcc.
838 Win32: sysopen, sysread, syswrite
840 As of the 5.8.0 release, sysopen()/sysread()/syswrite() do not behave
841 like they used to in 5.6.1 and earlier with respect to "text" mode.
842 These built-ins now always operate in "binary" mode (even if sysopen()
843 was passed the O_TEXT flag, or if binmode() was used on the file
844 handle). Note that this issue should only make a difference for disk
845 files, as sockets and pipes have always been in "binary" mode in the
846 Windows port. As this behavior is currently considered a bug,
847 compatible behavior may be re-introduced in a future release. Until
848 then, the use of sysopen(), sysread() and syswrite() is not supported
849 for "text" mode operations.
850
852 The following things might happen in future. The first publicly
853 available releases having these characteristics will be the developer
854 releases Perl 5.9.x, culminating in the Perl 5.10.0 release. These are
855 our best guesses at the moment: we reserve the right to rethink.
856 • PerlIO will become The Default. Currently (in Perl 5.8.x) the
858 stdio library is still used if Perl thinks it can use certain
859 tricks to make stdio go really fast. For future releases our goal
860 is to make PerlIO go even faster.
861 • A new feature called assertions will be available. This means that
863 one can have code called assertions sprinkled in the code: usually
864 they are optimised away, but they can be enabled with the "-A"
865 option.
866 • A new operator "//" (defined-or) will be available. This means
868 that one will be able to say
869 $a // $b
871 instead of
873 defined $a ? $a : $b
875 and
877 $c //= $d;
879 instead of
881 $c = $d unless defined $c;
883 The operator will have the same precedence and associativity as
885 "||". A source code patch against the Perl 5.8.1 sources will be
886 available in CPAN as authors/id/H/HM/HMBRAND/dor-5.8.1.diff.
887 • unpack() will default to unpacking the $_.
889 • Various Copy-On-Write techniques will be investigated in hopes of
891 speeding up Perl.
892 • CPANPLUS, Inline, and Module::Build will become core modules.
894 • The ability to write true lexically scoped pragmas will be
896 introduced.
897 • Work will continue on the bytecompiler and byteloader.
899 • v-strings as they currently exist are scheduled to be deprecated.
901 The v-less form (1.2.3) will become a "version object" when used
902 with "use", "require", and $VERSION. $^V will also be a "version
903 object" so the printf("%vd",...) construct will no longer be
904 needed. The v-ful version (v1.2.3) will become obsolete. The
905 equivalence of strings and v-strings (e.g. that currently 5.8.0 is
906 equal to "\5\8\0") will go away. There may be no deprecation
907 warning for v-strings, though: it is quite hard to detect when
908 v-strings are being used safely, and when they are not.
909 • 5.005 Threads Will Be Removed
911 • The $* Variable Will Be Removed (it was deprecated a long time ago)
913 • Pseudohashes Will Be Removed
915
917 If you find what you think is a bug, you might check the articles
918 recently posted to the comp.lang.perl.misc newsgroup and the perl bug
919 database at http://bugs.perl.org/ . There may also be information at
920 http://www.perl.com/ , the Perl Home Page.
921 If you believe you have an unreported bug, please run the perlbug
923 program included with your release. Be sure to trim your bug down to a
924 tiny but sufficient test case. Your bug report, along with the output
925 of "perl -V", will be sent off to perlbug@perl.org to be analysed by
926 the Perl porting team. You can browse and search the Perl 5 bugs at
927 http://bugs.perl.org/
928
930 The Changes file for exhaustive details on what changed.
931 The INSTALL file for how to build Perl.
933 The README file for general stuff.
935 The Artistic and Copying files for copyright information.
937perl v5.38.2 2023-11-30 PERL581DELTA(1)