3 perl561delta - what's new for perl v5.6.x
7 This document describes differences between the 5.005 release and the 5.6.1
10 =head1 Summary of changes between 5.6.0 and 5.6.1
12 This section contains a summary of the changes between the 5.6.0 release
13 and the 5.6.1 release. More details about the changes mentioned here
14 may be found in the F<Changes> files that accompany the Perl source
15 distribution. See L<perlhack> for pointers to online resources where you
16 can inspect the individual patches described by these changes.
18 =head2 Security Issues
20 suidperl will not run /bin/mail anymore, because some platforms have
21 a /bin/mail that is vulnerable to buffer overflow attacks.
23 Note that suidperl is neither built nor installed by default in
24 any recent version of perl. Use of suidperl is highly discouraged.
25 If you think you need it, try alternatives such as sudo first.
26 See http://www.courtesan.com/sudo/ .
30 This is not an exhaustive list. It is intended to cover only the
31 significant user-visible changes.
35 =item C<UNIVERSAL::isa()>
37 A bug in the caching mechanism used by C<UNIVERSAL::isa()> that affected
38 base.pm has been fixed. The bug has existed since the 5.005 releases,
39 but wasn't tickled by base.pm in those releases.
43 Various cases of memory leaks and attempts to access uninitialized memory
44 have been cured. See L</"Known Problems"> below for further issues.
46 =item Numeric conversions
48 Numeric conversions did not recognize changes in the string value
49 properly in certain circumstances.
51 In other situations, large unsigned numbers (those above 2**31) could
52 sometimes lose their unsignedness, causing bogus results in arithmetic
55 Integer modulus on large unsigned integers sometimes returned
58 Perl 5.6.0 generated "not a number" warnings on certain conversions where
59 previous versions didn't.
61 These problems have all been rectified.
63 Infinity is now recognized as a number.
67 In Perl 5.6.0, qw(a\\b) produced a string with two backslashes instead
68 of one, in a departure from the behavior in previous versions. The
69 older behavior has been reinstated.
73 caller() could cause core dumps in certain situations. Carp was sometimes
74 affected by this problem.
76 =item Bugs in regular expressions
78 Pattern matches on overloaded values are now handled correctly.
80 Perl 5.6.0 parsed m/\x{ab}/ incorrectly, leading to spurious warnings.
81 This has been corrected.
83 The RE engine found in Perl 5.6.0 accidentally pessimised certain kinds
84 of simple pattern matches. These are now handled better.
86 Regular expression debug output (whether through C<use re 'debug'>
87 or via C<-Dr>) now looks better.
89 Multi-line matches like C<"a\nxb\n" =~ /(?!\A)x/m> were flawed. The
92 Use of $& could trigger a core dump under some situations. This
95 Match variables $1 et al., weren't being unset when a pattern match
96 was backtracking, and the anomaly showed up inside C</...(?{ ... }).../>
97 etc. These variables are now tracked correctly.
99 pos() did not return the correct value within s///ge in earlier
100 versions. This is now handled correctly.
104 readline() on files opened in "slurp" mode could return an extra "" at
105 the end in certain situations. This has been corrected.
107 =item Autovivification of symbolic references to special variables
109 Autovivification of symbolic references of special variables described
110 in L<perlvar> (as in C<${$num}>) was accidentally disabled. This works
113 =item Lexical warnings
115 Lexical warnings now propagate correctly into C<eval "...">.
117 C<use warnings qw(FATAL all)> did not work as intended. This has been
120 Lexical warnings could leak into other scopes in some situations.
123 warnings::enabled() now reports the state of $^W correctly if the caller
124 isn't using lexical warnings.
126 =item Spurious warnings and errors
128 Perl 5.6.0 could emit spurious warnings about redefinition of dl_error()
129 when statically building extensions into perl. This has been corrected.
131 "our" variables could result in bogus "Variable will not stay shared"
132 warnings. This is now fixed.
134 "our" variables of the same name declared in two sibling blocks
135 resulted in bogus warnings about "redeclaration" of the variables.
136 The problem has been corrected.
140 Compatibility of the builtin glob() with old csh-based glob has been
141 improved with the addition of GLOB_ALPHASORT option. See C<File::Glob>.
143 File::Glob::glob() has been renamed to File::Glob::bsd_glob()
144 because the name clashes with the builtin glob(). The older
145 name is still available for compatibility, but is deprecated.
147 Spurious syntax errors generated in certain situations, when glob()
148 caused File::Glob to be loaded for the first time, have been fixed.
152 Some cases of inconsistent taint propagation (such as within hash
153 values) have been fixed.
155 The tainting behavior of sprintf() has been rationalized. It does
156 not taint the result of floating point formats anymore, making the
157 behavior consistent with that of string interpolation.
161 Arguments to sort() weren't being provided the right wantarray() context.
162 The comparison block is now run in scalar context, and the arguments to
163 be sorted are always provided list context.
165 sort() is also fully reentrant, in the sense that the sort function
166 can itself call sort(). This did not work reliably in previous releases.
168 =item #line directives
170 #line directives now work correctly when they appear at the very
171 beginning of C<eval "...">.
173 =item Subroutine prototypes
175 The (\&) prototype now works properly.
179 map() could get pathologically slow when the result list it generates
180 is larger than the source list. The performance has been improved for
185 Debugger exit code now reflects the script exit code.
187 Condition C<"0"> in breakpoints is now treated correctly.
189 The C<d> command now checks the line number.
191 C<$.> is no longer corrupted by the debugger.
193 All debugger output now correctly goes to the socket if RemotePort
198 PERL5OPT can be set to more than one switch group. Previously,
199 it used to be limited to one group of options only.
203 chop(@list) in list context returned the characters chopped in reverse
204 order. This has been reversed to be in the right order.
206 =item Unicode support
208 Unicode support has seen a large number of incremental improvements,
209 but continues to be highly experimental. It is not expected to be
210 fully supported in the 5.6.x maintenance releases.
212 substr(), join(), repeat(), reverse(), quotemeta() and string
213 concatenation were all handling Unicode strings incorrectly in
214 Perl 5.6.0. This has been corrected.
216 Support for C<tr///CU> and C<tr///UC> etc., have been removed since
217 we realized the interface is broken. For similar functionality,
218 see L<perlfunc/pack>.
220 The Unicode Character Database has been updated to version 3.0.1
221 with additions made available to the public as of August 30, 2000.
223 The Unicode character classes \p{Blank} and \p{SpacePerl} have been
224 added. "Blank" is like C isblank(), that is, it contains only
225 "horizontal whitespace" (the space character is, the newline isn't),
226 and the "SpacePerl" is the Unicode equivalent of C<\s> (\p{Space}
227 isn't, since that includes the vertical tabulator character, whereas
230 If you are experimenting with Unicode support in perl, the development
231 versions of Perl may have more to offer. In particular, I/O layers
232 are now available in the development track, but not in the maintenance
233 track, primarily to do backward compatibility issues. Unicode support
234 is also evolving rapidly on a daily basis in the development track--the
235 maintenance track only reflects the most conservative of these changes.
239 Support for 64-bit platforms has been improved, but continues to be
240 experimental. The level of support varies greatly among platforms.
244 The B Compiler and its various backends have had many incremental
245 improvements, but they continue to remain highly experimental. Use in
246 production environments is discouraged.
248 The perlcc tool has been rewritten so that the user interface is much
249 more like that of a C compiler.
251 The perlbc tools has been removed. Use C<perlcc -B> instead.
253 =item Lvalue subroutines
255 There have been various bugfixes to support lvalue subroutines better.
256 However, the feature still remains experimental.
260 IO::Socket::INET failed to open the specified port if the service
261 name was not known. It now correctly uses the supplied port number
266 File::Find now chdir()s correctly when chasing symbolic links.
270 xsubpp now tolerates embedded POD sections.
274 C<no Module;> does not produce an error even if Module does not have an
275 unimport() method. This parallels the behavior of C<use> vis-a-vis
280 A large number of tests have been added.
286 untie() will now call an UNTIE() hook if it exists. See L<perltie>
289 The C<-DT> command line switch outputs copious tokenizing information.
292 Arrays are now always interpolated in double-quotish strings. Previously,
293 C<"foo@bar.com"> used to be a fatal error at compile time, if an array
294 C<@bar> was not used or declared. This transitional behavior was
295 intended to help migrate perl4 code, and is deemed to be no longer useful.
296 See L</"Arrays now always interpolate into double-quoted strings">.
298 keys(), each(), pop(), push(), shift(), splice() and unshift()
299 can all be overridden now.
301 C<my __PACKAGE__ $obj> now does the expected thing.
303 =head2 Configuration issues
305 On some systems (IRIX and Solaris among them) the system malloc is demonstrably
306 better. While the defaults haven't been changed in order to retain binary
307 compatibility with earlier releases, you may be better off building perl
308 with C<Configure -Uusemymalloc ...> as discussed in the F<INSTALL> file.
310 C<Configure> has been enhanced in various ways:
316 Minimizes use of temporary files.
320 By default, does not link perl with libraries not used by it, such as
321 the various dbm libraries. SunOS 4.x hints preserve behavior on that
326 Support for pdp11-style memory models has been removed due to obsolescence.
330 Building outside the source tree is supported on systems that have
331 symbolic links. This is done by running
333 sh /path/to/source/Configure -Dmksymlinks ...
334 make all test install
336 in a directory other than the perl source directory. See F<INSTALL>.
340 C<Configure -S> can be run non-interactively.
346 README.aix, README.solaris and README.macos have been added.
347 README.posix-bc has been renamed to README.bs2000. These are
348 installed as L<perlaix>, L<perlsolaris>, L<perlmacos>, and
349 L<perlbs2000> respectively.
351 The following pod documents are brand new:
353 perlclib Internal replacements for standard C library functions
354 perldebtut Perl debugging tutorial
355 perlebcdic Considerations for running Perl on EBCDIC platforms
356 perlnewmod Perl modules: preparing a new module for distribution
357 perlrequick Perl regular expressions quick start
358 perlretut Perl regular expressions tutorial
359 perlutil utilities packaged with the Perl distribution
361 The F<INSTALL> file has been expanded to cover various issues, such as
364 A longer list of contributors has been added to the source distribution.
365 See the file C<AUTHORS>.
367 Numerous other changes have been made to the included documentation and FAQs.
369 =head2 Bundled modules
371 The following modules have been added.
377 Walks Perl syntax tree, printing concise info about ops. See L<B::Concise>.
381 Returns name and handle of a temporary file safely. See L<File::Temp>.
385 Converts Pod data to formatted LaTeX. See L<Pod::LaTeX>.
387 =item Pod::Text::Overstrike
389 Converts POD data to formatted overstrike text. See L<Pod::Text::Overstrike>.
393 The following modules have been upgraded.
399 CGI v2.752 is now included.
403 CPAN v1.59_54 is now included.
407 Various bugfixes have been added.
411 DB_File v1.75 supports newer Berkeley DB versions, among other
416 Devel::Peek has been enhanced to support dumping of memory statistics,
417 when perl is built with the included malloc().
421 File::Find now supports pre and post-processing of the files in order
426 Getopt::Long v2.25 is included.
430 Various bug fixes have been included.
434 IPC::Open3 allows use of numeric file descriptors.
438 The fmod() function supports modulus operations. Various bug fixes
439 have also been included.
443 Math::Complex handles inf, NaN etc., better.
447 ping() could fail on odd number of data bytes, and when the echo service
448 isn't running. This has been corrected.
452 A memory leak has been fixed.
456 Version 1.13 of the Pod::Parser suite is included.
460 Pod::Text and related modules have been upgraded to the versions
461 in podlators suite v2.08.
465 On dosish platforms, some keys went missing because of lack of support for
466 files with "holes". A workaround for the problem has been added.
470 Various bug fixes have been included.
474 Now supports Tie::RefHash::Nestable to automagically tie hashref values.
476 =item Tie::SubstrHash
478 Various bug fixes have been included.
482 =head2 Platform-specific improvements
484 The following new ports are now available.
494 Perl now builds under Amdahl UTS.
496 Perl has also been verified to build under Amiga OS.
498 Support for EPOC has been much improved. See README.epoc.
500 Building perl with -Duseithreads or -Duse5005threads now works
501 under HP-UX 10.20 (previously it only worked under 10.30 or later).
502 You will need a thread library package installed. See README.hpux.
504 Long doubles should now work under Linux.
506 Mac OS Classic is now supported in the mainstream source package.
509 Support for MPE/iX has been updated. See README.mpeix.
511 Support for OS/2 has been improved. See C<os2/Changes> and README.os2.
513 Dynamic loading on z/OS (formerly OS/390) has been improved. See
516 Support for VMS has seen many incremental improvements, including
517 better support for operators like backticks and system(), and better
518 %ENV handling. See C<README.vms> and L<perlvms>.
520 Support for Stratus VOS has been improved. See C<vos/Changes> and README.vos.
522 Support for Windows has been improved.
528 fork() emulation has been improved in various ways, but still continues
529 to be experimental. See L<perlfork> for known bugs and caveats.
533 %SIG has been enabled under USE_ITHREADS, but its use is completely
534 unsupported under all configurations.
538 Borland C++ v5.5 is now a supported compiler that can build Perl.
539 However, the generated binaries continue to be incompatible with those
540 generated by the other supported compilers (GCC and Visual C++).
544 Non-blocking waits for child processes (or pseudo-processes) are
545 supported via C<waitpid($pid, &POSIX::WNOHANG)>.
549 A memory leak in accept() has been fixed.
553 wait(), waitpid() and backticks now return the correct exit status under
558 Trailing new %ENV entries weren't propagated to child processes. This
563 Current directory entries in %ENV are now correctly propagated to child
568 Duping socket handles with open(F, ">&MYSOCK") now works under Windows 9x.
572 The makefiles now provide a single switch to bulk-enable all the features
573 enabled in ActiveState ActivePerl (a popular binary distribution).
577 Win32::GetCwd() correctly returns C:\ instead of C: when at the drive root.
578 Other bugs in chdir() and Cwd::cwd() have also been fixed.
582 fork() correctly returns undef and sets EAGAIN when it runs out of
583 pseudo-process handles.
587 ExtUtils::MakeMaker now uses $ENV{LIB} to search for libraries.
591 UNC path handling is better when perl is built to support fork().
595 A handle leak in socket handling has been fixed.
599 send() works from within a pseudo-process.
603 Unless specifically qualified otherwise, the remainder of this document
604 covers changes between the 5.005 and 5.6.0 releases.
606 =head1 Core Enhancements
608 =head2 Interpreter cloning, threads, and concurrency
610 Perl 5.6.0 introduces the beginnings of support for running multiple
611 interpreters concurrently in different threads. In conjunction with
612 the perl_clone() API call, which can be used to selectively duplicate
613 the state of any given interpreter, it is possible to compile a
614 piece of code once in an interpreter, clone that interpreter
615 one or more times, and run all the resulting interpreters in distinct
618 On the Windows platform, this feature is used to emulate fork() at the
619 interpreter level. See L<perlfork> for details about that.
621 This feature is still in evolution. It is eventually meant to be used
622 to selectively clone a subroutine and data reachable from that
623 subroutine in a separate interpreter and run the cloned subroutine
624 in a separate thread. Since there is no shared data between the
625 interpreters, little or no locking will be needed (unless parts of
626 the symbol table are explicitly shared). This is obviously intended
627 to be an easy-to-use replacement for the existing threads support.
629 Support for cloning interpreters and interpreter concurrency can be
630 enabled using the -Dusethreads Configure option (see win32/Makefile for
631 how to enable it on Windows.) The resulting perl executable will be
632 functionally identical to one that was built with -Dmultiplicity, but
633 the perl_clone() API call will only be available in the former.
635 -Dusethreads enables the cpp macro USE_ITHREADS by default, which in turn
636 enables Perl source code changes that provide a clear separation between
637 the op tree and the data it operates with. The former is immutable, and
638 can therefore be shared between an interpreter and all of its clones,
639 while the latter is considered local to each interpreter, and is therefore
640 copied for each clone.
642 Note that building Perl with the -Dusemultiplicity Configure option
643 is adequate if you wish to run multiple B<independent> interpreters
644 concurrently in different threads. -Dusethreads only provides the
645 additional functionality of the perl_clone() API call and other
646 support for running B<cloned> interpreters concurrently.
648 NOTE: This is an experimental feature. Implementation details are
651 =head2 Lexically scoped warning categories
653 You can now control the granularity of warnings emitted by perl at a finer
654 level using the C<use warnings> pragma. L<warnings> and L<perllexwarn>
655 have copious documentation on this feature.
657 =head2 Unicode and UTF-8 support
659 Perl now uses UTF-8 as its internal representation for character
660 strings. The C<utf8> and C<bytes> pragmas are used to control this support
661 in the current lexical scope. See L<perlunicode>, L<utf8> and L<bytes> for
664 This feature is expected to evolve quickly to support some form of I/O
665 disciplines that can be used to specify the kind of input and output data
666 (bytes or characters). Until that happens, additional modules from CPAN
667 will be needed to complete the toolkit for dealing with Unicode.
669 NOTE: This should be considered an experimental feature. Implementation
670 details are subject to change.
672 =head2 Support for interpolating named characters
674 The new C<\N> escape interpolates named characters within strings.
675 For example, C<"Hi! \N{WHITE SMILING FACE}"> evaluates to a string
676 with a Unicode smiley face at the end.
678 =head2 "our" declarations
680 An "our" declaration introduces a value that can be best understood
681 as a lexically scoped symbolic alias to a global variable in the
682 package that was current where the variable was declared. This is
683 mostly useful as an alternative to the C<vars> pragma, but also provides
684 the opportunity to introduce typing and other attributes for such
685 variables. See L<perlfunc/our>.
687 =head2 Support for strings represented as a vector of ordinals
689 Literals of the form C<v1.2.3.4> are now parsed as a string composed
690 of characters with the specified ordinals. This is an alternative, more
691 readable way to construct (possibly Unicode) strings instead of
692 interpolating characters, as in C<"\x{1}\x{2}\x{3}\x{4}">. The leading
693 C<v> may be omitted if there are more than two ordinals, so C<1.2.3> is
694 parsed the same as C<v1.2.3>.
696 Strings written in this form are also useful to represent version "numbers".
697 It is easy to compare such version "numbers" (which are really just plain
698 strings) using any of the usual string comparison operators C<eq>, C<ne>,
699 C<lt>, C<gt>, etc., or perform bitwise string operations on them using C<|>,
702 In conjunction with the new C<$^V> magic variable (which contains
703 the perl version as a string), such literals can be used as a readable way
704 to check if you're running a particular version of Perl:
706 # this will parse in older versions of Perl also
707 if ($^V and $^V gt v5.6.0) {
708 # new features supported
711 C<require> and C<use> also have some special magic to support such literals.
712 They will be interpreted as a version rather than as a module name:
714 require v5.6.0; # croak if $^V lt v5.6.0
715 use v5.6.0; # same, but croaks at compile-time
717 Alternatively, the C<v> may be omitted if there is more than one dot:
722 Also, C<sprintf> and C<printf> support the Perl-specific format flag C<%v>
723 to print ordinals of characters in arbitrary strings:
725 printf "v%vd", $^V; # prints current version, such as "v5.5.650"
726 printf "%*vX", ":", $addr; # formats IPv6 address
727 printf "%*vb", " ", $bits; # displays bitstring
729 See L<perldata/"Scalar value constructors"> for additional information.
731 =head2 Improved Perl version numbering system
733 Beginning with Perl version 5.6.0, the version number convention has been
734 changed to a "dotted integer" scheme that is more commonly found in open
737 Maintenance versions of v5.6.0 will be released as v5.6.1, v5.6.2 etc.
738 The next development series following v5.6.0 will be numbered v5.7.x,
739 beginning with v5.7.0, and the next major production release following
740 v5.6.0 will be v5.8.0.
742 The English module now sets $PERL_VERSION to $^V (a string value) rather
743 than C<$]> (a numeric value). (This is a potential incompatibility.
744 Send us a report via perlbug if you are affected by this.)
746 The v1.2.3 syntax is also now legal in Perl.
747 See L<Support for strings represented as a vector of ordinals> for more on that.
749 To cope with the new versioning system's use of at least three significant
750 digits for each version component, the method used for incrementing the
751 subversion number has also changed slightly. We assume that versions older
752 than v5.6.0 have been incrementing the subversion component in multiples of
753 10. Versions after v5.6.0 will increment them by 1. Thus, using the new
754 notation, 5.005_03 is the "same" as v5.5.30, and the first maintenance
755 version following v5.6.0 will be v5.6.1 (which should be read as being
756 equivalent to a floating point value of 5.006_001 in the older format,
759 =head2 New syntax for declaring subroutine attributes
761 Formerly, if you wanted to mark a subroutine as being a method call or
762 as requiring an automatic lock() when it is entered, you had to declare
763 that with a C<use attrs> pragma in the body of the subroutine.
764 That can now be accomplished with declaration syntax, like this:
766 sub mymethod : locked method ;
768 sub mymethod : locked method {
772 sub othermethod :locked :method ;
774 sub othermethod :locked :method {
779 (Note how only the first C<:> is mandatory, and whitespace surrounding
780 the C<:> is optional.)
782 F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes
783 with the stubs they provide. See L<attributes>.
785 =head2 File and directory handles can be autovivified
787 Similar to how constructs such as C<< $x->[0] >> autovivify a reference,
788 handle constructors (open(), opendir(), pipe(), socketpair(), sysopen(),
789 socket(), and accept()) now autovivify a file or directory handle
790 if the handle passed to them is an uninitialized scalar variable. This
791 allows the constructs such as C<open(my $fh, ...)> and C<open(local $fh,...)>
792 to be used to create filehandles that will conveniently be closed
793 automatically when the scope ends, provided there are no other references
794 to them. This largely eliminates the need for typeglobs when opening
795 filehandles that must be passed around, as in the following example:
799 or die "Can't open '@_': $!";
804 my $f = myopen("</etc/motd");
806 # $f implicitly closed here
809 =head2 open() with more than two arguments
811 If open() is passed three arguments instead of two, the second argument
812 is used as the mode and the third argument is taken to be the file name.
813 This is primarily useful for protecting against unintended magic behavior
814 of the traditional two-argument form. See L<perlfunc/open>.
816 =head2 64-bit support
818 Any platform that has 64-bit integers either
820 (1) natively as longs or ints
821 (2) via special compiler flags
822 (3) using long long or int64_t
824 is able to use "quads" (64-bit integers) as follows:
830 constants (decimal, hexadecimal, octal, binary) in the code
834 arguments to oct() and hex()
838 arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)
846 pack() and unpack() "q" and "Q" formats
850 in basic arithmetics: + - * / % (NOTE: operating close to the limits
851 of the integer values may produce surprising results)
855 in bit arithmetics: & | ^ ~ << >> (NOTE: these used to be forced
856 to be 32 bits wide but now operate on the full native width.)
864 Note that unless you have the case (a) you will have to configure
865 and compile Perl using the -Duse64bitint Configure flag.
867 NOTE: The Configure flags -Duselonglong and -Duse64bits have been
868 deprecated. Use -Duse64bitint instead.
870 There are actually two modes of 64-bitness: the first one is achieved
871 using Configure -Duse64bitint and the second one using Configure
872 -Duse64bitall. The difference is that the first one is minimal and
873 the second one maximal. The first works in more places than the second.
875 The C<use64bitint> does only as much as is required to get 64-bit
876 integers into Perl (this may mean, for example, using "long longs")
877 while your memory may still be limited to 2 gigabytes (because your
878 pointers could still be 32-bit). Note that the name C<64bitint> does
879 not imply that your C compiler will be using 64-bit C<int>s (it might,
880 but it doesn't have to): the C<use64bitint> means that you will be
881 able to have 64 bits wide scalar values.
883 The C<use64bitall> goes all the way by attempting to switch also
884 integers (if it can), longs (and pointers) to being 64-bit. This may
885 create an even more binary incompatible Perl than -Duse64bitint: the
886 resulting executable may not run at all in a 32-bit box, or you may
887 have to reboot/reconfigure/rebuild your operating system to be 64-bit
890 Natively 64-bit systems like Alpha and Cray need neither -Duse64bitint
893 Last but not least: note that due to Perl's habit of always using
894 floating point numbers, the quads are still not true integers.
895 When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned,
896 -9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they
897 are silently promoted to floating point numbers, after which they will
898 start losing precision (in their lower digits).
900 NOTE: 64-bit support is still experimental on most platforms.
901 Existing support only covers the LP64 data model. In particular, the
902 LLP64 data model is not yet supported. 64-bit libraries and system
903 APIs on many platforms have not stabilized--your mileage may vary.
905 =head2 Large file support
907 If you have filesystems that support "large files" (files larger than
908 2 gigabytes), you may now also be able to create and access them from
911 NOTE: The default action is to enable large file support, if
912 available on the platform.
914 If the large file support is on, and you have a Fcntl constant
915 O_LARGEFILE, the O_LARGEFILE is automatically added to the flags
918 Beware that unless your filesystem also supports "sparse files" seeking
919 to umpteen petabytes may be inadvisable.
921 Note that in addition to requiring a proper file system to do large
922 files you may also need to adjust your per-process (or your
923 per-system, or per-process-group, or per-user-group) maximum filesize
924 limits before running Perl scripts that try to handle large files,
925 especially if you intend to write such files.
927 Finally, in addition to your process/process group maximum filesize
928 limits, you may have quota limits on your filesystems that stop you
929 (your user id or your user group id) from using large files.
931 Adjusting your process/user/group/file system/operating system limits
932 is outside the scope of Perl core language. For process limits, you
933 may try increasing the limits using your shell's limits/limit/ulimit
934 command before running Perl. The BSD::Resource extension (not
935 included with the standard Perl distribution) may also be of use, it
936 offers the getrlimit/setrlimit interface that can be used to adjust
937 process resource usage limits, including the maximum filesize limit.
941 In some systems you may be able to use long doubles to enhance the
942 range and precision of your double precision floating point numbers
943 (that is, Perl's numbers). Use Configure -Duselongdouble to enable
944 this support (if it is available).
948 You can "Configure -Dusemorebits" to turn on both the 64-bit support
949 and the long double support.
951 =head2 Enhanced support for sort() subroutines
953 Perl subroutines with a prototype of C<($$)>, and XSUBs in general, can
954 now be used as sort subroutines. In either case, the two elements to
955 be compared are passed as normal parameters in @_. See L<perlfunc/sort>.
957 For unprototyped sort subroutines, the historical behavior of passing
958 the elements to be compared as the global variables $a and $b remains
961 =head2 C<sort $coderef @foo> allowed
963 sort() did not accept a subroutine reference as the comparison
964 function in earlier versions. This is now permitted.
966 =head2 File globbing implemented internally
968 Perl now uses the File::Glob implementation of the glob() operator
969 automatically. This avoids using an external csh process and the
970 problems associated with it.
972 NOTE: This is currently an experimental feature. Interfaces and
973 implementation are subject to change.
975 =head2 Support for CHECK blocks
977 In addition to C<BEGIN>, C<INIT>, C<END>, C<DESTROY> and C<AUTOLOAD>,
978 subroutines named C<CHECK> are now special. These are queued up during
979 compilation and behave similar to END blocks, except they are called at
980 the end of compilation rather than at the end of execution. They cannot
983 =head2 POSIX character class syntax [: :] supported
985 For example to match alphabetic characters use /[[:alpha:]]/.
986 See L<perlre> for details.
988 =head2 Better pseudo-random number generator
990 In 5.005_0x and earlier, perl's rand() function used the C library
991 rand(3) function. As of 5.005_52, Configure tests for drand48(),
992 random(), and rand() (in that order) and picks the first one it finds.
994 These changes should result in better random numbers from rand().
996 =head2 Improved C<qw//> operator
998 The C<qw//> operator is now evaluated at compile time into a true list
999 instead of being replaced with a run time call to C<split()>. This
1000 removes the confusing misbehaviour of C<qw//> in scalar context, which
1001 had inherited that behaviour from split().
1005 $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
1007 now correctly prints "3|a", instead of "2|a".
1009 =head2 Better worst-case behavior of hashes
1011 Small changes in the hashing algorithm have been implemented in
1012 order to improve the distribution of lower order bits in the
1013 hashed value. This is expected to yield better performance on
1014 keys that are repeated sequences.
1016 =head2 pack() format 'Z' supported
1018 The new format type 'Z' is useful for packing and unpacking null-terminated
1019 strings. See L<perlfunc/"pack">.
1021 =head2 pack() format modifier '!' supported
1023 The new format type modifier '!' is useful for packing and unpacking
1024 native shorts, ints, and longs. See L<perlfunc/"pack">.
1026 =head2 pack() and unpack() support counted strings
1028 The template character '/' can be used to specify a counted string
1029 type to be packed or unpacked. See L<perlfunc/"pack">.
1031 =head2 Comments in pack() templates
1033 The '#' character in a template introduces a comment up to
1034 end of the line. This facilitates documentation of pack()
1037 =head2 Weak references
1039 In previous versions of Perl, you couldn't cache objects so as
1040 to allow them to be deleted if the last reference from outside
1041 the cache is deleted. The reference in the cache would hold a
1042 reference count on the object and the objects would never be
1045 Another familiar problem is with circular references. When an
1046 object references itself, its reference count would never go
1047 down to zero, and it would not get destroyed until the program
1050 Weak references solve this by allowing you to "weaken" any
1051 reference, that is, make it not count towards the reference count.
1052 When the last non-weak reference to an object is deleted, the object
1053 is destroyed and all the weak references to the object are
1054 automatically undef-ed.
1056 To use this feature, you need the Devel::WeakRef package from CPAN, which
1057 contains additional documentation.
1059 NOTE: This is an experimental feature. Details are subject to change.
1061 =head2 Binary numbers supported
1063 Binary numbers are now supported as literals, in s?printf formats, and
1067 printf "The answer is: %b\n", oct("0b101010");
1069 =head2 Lvalue subroutines
1071 Subroutines can now return modifiable lvalues.
1072 See L<perlsub/"Lvalue subroutines">.
1074 NOTE: This is an experimental feature. Details are subject to change.
1076 =head2 Some arrows may be omitted in calls through references
1078 Perl now allows the arrow to be omitted in many constructs
1079 involving subroutine calls through references. For example,
1080 C<< $foo[10]->('foo') >> may now be written C<$foo[10]('foo')>.
1081 This is rather similar to how the arrow may be omitted from
1082 C<< $foo[10]->{'foo'} >>. Note however, that the arrow is still
1083 required for C<< foo(10)->('bar') >>.
1085 =head2 Boolean assignment operators are legal lvalues
1087 Constructs such as C<($a ||= 2) += 1> are now allowed.
1089 =head2 exists() is supported on subroutine names
1091 The exists() builtin now works on subroutine names. A subroutine
1092 is considered to exist if it has been declared (even if implicitly).
1093 See L<perlfunc/exists> for examples.
1095 =head2 exists() and delete() are supported on array elements
1097 The exists() and delete() builtins now work on simple arrays as well.
1098 The behavior is similar to that on hash elements.
1100 exists() can be used to check whether an array element has been
1101 initialized. This avoids autovivifying array elements that don't exist.
1102 If the array is tied, the EXISTS() method in the corresponding tied
1103 package will be invoked.
1105 delete() may be used to remove an element from the array and return
1106 it. The array element at that position returns to its uninitialized
1107 state, so that testing for the same element with exists() will return
1108 false. If the element happens to be the one at the end, the size of
1109 the array also shrinks up to the highest element that tests true for
1110 exists(), or 0 if none such is found. If the array is tied, the DELETE()
1111 method in the corresponding tied package will be invoked.
1113 See L<perlfunc/exists> and L<perlfunc/delete> for examples.
1115 =head2 Pseudo-hashes work better
1117 Dereferencing some types of reference values in a pseudo-hash,
1118 such as C<< $ph->{foo}[1] >>, was accidentally disallowed. This has
1121 When applied to a pseudo-hash element, exists() now reports whether
1122 the specified value exists, not merely if the key is valid.
1124 delete() now works on pseudo-hashes. When given a pseudo-hash element
1125 or slice it deletes the values corresponding to the keys (but not the keys
1126 themselves). See L<perlref/"Pseudo-hashes: Using an array as a hash">.
1128 Pseudo-hash slices with constant keys are now optimized to array lookups
1131 List assignments to pseudo-hash slices are now supported.
1133 The C<fields> pragma now provides ways to create pseudo-hashes, via
1134 fields::new() and fields::phash(). See L<fields>.
1136 NOTE: The pseudo-hash data type continues to be experimental.
1137 Limiting oneself to the interface elements provided by the
1138 fields pragma will provide protection from any future changes.
1140 =head2 Automatic flushing of output buffers
1142 fork(), exec(), system(), qx//, and pipe open()s now flush buffers
1143 of all files opened for output when the operation was attempted. This
1144 mostly eliminates confusing buffering mishaps suffered by users unaware
1145 of how Perl internally handles I/O.
1147 This is not supported on some platforms like Solaris where a suitably
1148 correct implementation of fflush(NULL) isn't available.
1150 =head2 Better diagnostics on meaningless filehandle operations
1152 Constructs such as C<< open(<FH>) >> and C<< close(<FH>) >>
1153 are compile time errors. Attempting to read from filehandles that
1154 were opened only for writing will now produce warnings (just as
1155 writing to read-only filehandles does).
1157 =head2 Where possible, buffered data discarded from duped input filehandle
1159 C<< open(NEW, "<&OLD") >> now attempts to discard any data that
1160 was previously read and buffered in C<OLD> before duping the handle.
1161 On platforms where doing this is allowed, the next read operation
1162 on C<NEW> will return the same data as the corresponding operation
1163 on C<OLD>. Formerly, it would have returned the data from the start
1164 of the following disk block instead.
1166 =head2 eof() has the same old magic as <>
1168 C<eof()> would return true if no attempt to read from C<< <> >> had
1169 yet been made. C<eof()> has been changed to have a little magic of its
1170 own, it now opens the C<< <> >> files.
1172 =head2 binmode() can be used to set :crlf and :raw modes
1174 binmode() now accepts a second argument that specifies a discipline
1175 for the handle in question. The two pseudo-disciplines ":raw" and
1176 ":crlf" are currently supported on DOS-derivative platforms.
1177 See L<perlfunc/"binmode"> and L<open>.
1179 =head2 C<-T> filetest recognizes UTF-8 encoded files as "text"
1181 The algorithm used for the C<-T> filetest has been enhanced to
1182 correctly identify UTF-8 content as "text".
1184 =head2 system(), backticks and pipe open now reflect exec() failure
1186 On Unix and similar platforms, system(), qx() and open(FOO, "cmd |")
1187 etc., are implemented via fork() and exec(). When the underlying
1188 exec() fails, earlier versions did not report the error properly,
1189 since the exec() happened to be in a different process.
1191 The child process now communicates with the parent about the
1192 error in launching the external command, which allows these
1193 constructs to return with their usual error value and set $!.
1195 =head2 Improved diagnostics
1197 Line numbers are no longer suppressed (under most likely circumstances)
1198 during the global destruction phase.
1200 Diagnostics emitted from code running in threads other than the main
1201 thread are now accompanied by the thread ID.
1203 Embedded null characters in diagnostics now actually show up. They
1204 used to truncate the message in prior versions.
1206 $foo::a and $foo::b are now exempt from "possible typo" warnings only
1207 if sort() is encountered in package C<foo>.
1209 Unrecognized alphabetic escapes encountered when parsing quote
1210 constructs now generate a warning, since they may take on new
1211 semantics in later versions of Perl.
1213 Many diagnostics now report the internal operation in which the warning
1214 was provoked, like so:
1216 Use of uninitialized value in concatenation (.) at (eval 1) line 1.
1217 Use of uninitialized value in print at (eval 1) line 1.
1219 Diagnostics that occur within eval may also report the file and line
1220 number where the eval is located, in addition to the eval sequence
1221 number and the line number within the evaluated text itself. For
1224 Not enough arguments for scalar at (eval 4)[newlib/perl5db.pl:1411] line 2, at EOF
1226 =head2 Diagnostics follow STDERR
1228 Diagnostic output now goes to whichever file the C<STDERR> handle
1229 is pointing at, instead of always going to the underlying C runtime
1230 library's C<stderr>.
1232 =head2 More consistent close-on-exec behavior
1234 On systems that support a close-on-exec flag on filehandles, the
1235 flag is now set for any handles created by pipe(), socketpair(),
1236 socket(), and accept(), if that is warranted by the value of $^F
1237 that may be in effect. Earlier versions neglected to set the flag
1238 for handles created with these operators. See L<perlfunc/pipe>,
1239 L<perlfunc/socketpair>, L<perlfunc/socket>, L<perlfunc/accept>,
1242 =head2 syswrite() ease-of-use
1244 The length argument of C<syswrite()> has become optional.
1246 =head2 Better syntax checks on parenthesized unary operators
1248 Expressions such as:
1250 print defined(&foo,&bar,&baz);
1251 print uc("foo","bar","baz");
1254 used to be accidentally allowed in earlier versions, and produced
1255 unpredictable behaviour. Some produced ancillary warnings
1256 when used in this way; others silently did the wrong thing.
1258 The parenthesized forms of most unary operators that expect a single
1259 argument now ensure that they are not called with more than one
1260 argument, making the cases shown above syntax errors. The usual
1263 print defined &foo, &bar, &baz;
1264 print uc "foo", "bar", "baz";
1267 remains unchanged. See L<perlop>.
1269 =head2 Bit operators support full native integer width
1271 The bit operators (& | ^ ~ << >>) now operate on the full native
1272 integral width (the exact size of which is available in $Config{ivsize}).
1273 For example, if your platform is either natively 64-bit or if Perl
1274 has been configured to use 64-bit integers, these operations apply
1275 to 8 bytes (as opposed to 4 bytes on 32-bit platforms).
1276 For portability, be sure to mask off the excess bits in the result of
1277 unary C<~>, e.g., C<~$x & 0xffffffff>.
1279 =head2 Improved security features
1281 More potentially unsafe operations taint their results for improved
1284 The C<passwd> and C<shell> fields returned by the getpwent(), getpwnam(),
1285 and getpwuid() are now tainted, because the user can affect their own
1286 encrypted password and login shell.
1288 The variable modified by shmread(), and messages returned by msgrcv()
1289 (and its object-oriented interface IPC::SysV::Msg::rcv) are also tainted,
1290 because other untrusted processes can modify messages and shared memory
1291 segments for their own nefarious purposes.
1293 =head2 More functional bareword prototype (*)
1295 Bareword prototypes have been rationalized to enable them to be used
1296 to override builtins that accept barewords and interpret them in
1297 a special way, such as C<require> or C<do>.
1299 Arguments prototyped as C<*> will now be visible within the subroutine
1300 as either a simple scalar or as a reference to a typeglob.
1301 See L<perlsub/Prototypes>.
1303 =head2 C<require> and C<do> may be overridden
1305 C<require> and C<do 'file'> operations may be overridden locally
1306 by importing subroutines of the same name into the current package
1307 (or globally by importing them into the CORE::GLOBAL:: namespace).
1308 Overriding C<require> will also affect C<use>, provided the override
1309 is visible at compile-time.
1310 See L<perlsub/"Overriding Built-in Functions">.
1312 =head2 $^X variables may now have names longer than one character
1314 Formerly, $^X was synonymous with ${"\cX"}, but $^XY was a syntax
1315 error. Now variable names that begin with a control character may be
1316 arbitrarily long. However, for compatibility reasons, these variables
1317 I<must> be written with explicit braces, as C<${^XY}> for example.
1318 C<${^XYZ}> is synonymous with ${"\cXYZ"}. Variable names with more
1319 than one control character, such as C<${^XY^Z}>, are illegal.
1321 The old syntax has not changed. As before, `^X' may be either a
1322 literal control-X character or the two-character sequence `caret' plus
1323 `X'. When braces are omitted, the variable name stops after the
1324 control character. Thus C<"$^XYZ"> continues to be synonymous with
1325 C<$^X . "YZ"> as before.
1327 As before, lexical variables may not have names beginning with control
1328 characters. As before, variables whose names begin with a control
1329 character are always forced to be in package `main'. All such variables
1330 are reserved for future extensions, except those that begin with
1331 C<^_>, which may be used by user programs and are guaranteed not to
1332 acquire special meaning in any future version of Perl.
1334 =head2 New variable $^C reflects C<-c> switch
1336 C<$^C> has a boolean value that reflects whether perl is being run
1337 in compile-only mode (i.e. via the C<-c> switch). Since
1338 BEGIN blocks are executed under such conditions, this variable
1339 enables perl code to determine whether actions that make sense
1340 only during normal running are warranted. See L<perlvar>.
1342 =head2 New variable $^V contains Perl version as a string
1344 C<$^V> contains the Perl version number as a string composed of
1345 characters whose ordinals match the version numbers, i.e. v5.6.0.
1346 This may be used in string comparisons.
1348 See C<Support for strings represented as a vector of ordinals> for an
1351 =head2 Optional Y2K warnings
1353 If Perl is built with the cpp macro C<PERL_Y2KWARN> defined,
1354 it emits optional warnings when concatenating the number 19
1355 with another number.
1357 This behavior must be specifically enabled when running Configure.
1358 See F<INSTALL> and F<README.Y2K>.
1360 =head2 Arrays now always interpolate into double-quoted strings
1362 In double-quoted strings, arrays now interpolate, no matter what. The
1363 behavior in earlier versions of perl 5 was that arrays would interpolate
1364 into strings if the array had been mentioned before the string was
1365 compiled, and otherwise Perl would raise a fatal compile-time error.
1366 In versions 5.000 through 5.003, the error was
1368 Literal @example now requires backslash
1370 In versions 5.004_01 through 5.6.0, the error was
1372 In string, @example now must be written as \@example
1374 The idea here was to get people into the habit of writing
1375 C<"fred\@example.com"> when they wanted a literal C<@> sign, just as
1376 they have always written C<"Give me back my \$5"> when they wanted a
1379 Starting with 5.6.1, when Perl now sees an C<@> sign in a
1380 double-quoted string, it I<always> attempts to interpolate an array,
1381 regardless of whether or not the array has been used or declared
1382 already. The fatal error has been downgraded to an optional warning:
1384 Possible unintended interpolation of @example in string
1386 This warns you that C<"fred@example.com"> is going to turn into
1387 C<fred.com> if you don't backslash the C<@>.
1388 See http://www.plover.com/~mjd/perl/at-error.html for more details
1389 about the history here.
1391 =head1 Modules and Pragmata
1399 While used internally by Perl as a pragma, this module also
1400 provides a way to fetch subroutine and variable attributes.
1405 The Perl Compiler suite has been extensively reworked for this
1406 release. More of the standard Perl testsuite passes when run
1407 under the Compiler, but there is still a significant way to
1408 go to achieve production quality compiled executables.
1410 NOTE: The Compiler suite remains highly experimental. The
1411 generated code may not be correct, even when it manages to execute
1416 Overall, Benchmark results exhibit lower average error and better timing
1419 You can now run tests for I<n> seconds instead of guessing the right
1420 number of tests to run: e.g., timethese(-5, ...) will run each
1421 code for at least 5 CPU seconds. Zero as the "number of repetitions"
1422 means "for at least 3 CPU seconds". The output format has also
1423 changed. For example:
1425 use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
1427 will now output something like this:
1429 Benchmark: running a, b, each for at least 5 CPU seconds...
1430 a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
1431 b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
1433 New features: "each for at least N CPU seconds...", "wallclock secs",
1434 and the "@ operations/CPU second (n=operations)".
1436 timethese() now returns a reference to a hash of Benchmark objects containing
1437 the test results, keyed on the names of the tests.
1439 timethis() now returns the iterations field in the Benchmark result object
1442 timethese(), timethis(), and the new cmpthese() (see below) can also take
1443 a format specifier of 'none' to suppress output.
1445 A new function countit() is just like timeit() except that it takes a
1446 TIME instead of a COUNT.
1448 A new function cmpthese() prints a chart comparing the results of each test
1449 returned from a timethese() call. For each possible pair of tests, the
1450 percentage speed difference (iters/sec or seconds/iter) is shown.
1452 For other details, see L<Benchmark>.
1456 The ByteLoader is a dedicated extension to generate and run
1457 Perl bytecode. See L<ByteLoader>.
1461 References can now be used.
1463 The new version also allows a leading underscore in constant names, but
1464 disallows a double leading underscore (as in "__LINE__"). Some other names
1465 are disallowed or warned against, including BEGIN, END, etc. Some names
1466 which were forced into main:: used to fail silently in some cases; now they're
1467 fatal (outside of main::) and an optional warning (inside of main::).
1468 The ability to detect whether a constant had been set with a given name has
1475 This pragma implements the C<\N> string escape. See L<charnames>.
1479 A C<Maxdepth> setting can be specified to avoid venturing
1480 too deeply into deep data structures. See L<Data::Dumper>.
1482 The XSUB implementation of Dump() is now automatically called if the
1483 C<Useqq> setting is not in use.
1485 Dumping C<qr//> objects works correctly.
1489 C<DB> is an experimental module that exposes a clean abstraction
1490 to Perl's debugging API.
1494 DB_File can now be built with Berkeley DB versions 1, 2 or 3.
1495 See C<ext/DB_File/Changes>.
1499 Devel::DProf, a Perl source code profiler has been added. See
1500 L<Devel::DProf> and L<dprofpp>.
1504 The Devel::Peek module provides access to the internal representation
1505 of Perl variables and data. It is a data debugging tool for the XS programmer.
1509 The Dumpvalue module provides screen dumps of Perl data.
1513 DynaLoader now supports a dl_unload_file() function on platforms that
1514 support unloading shared objects using dlclose().
1516 Perl can also optionally arrange to unload all extension shared objects
1517 loaded by Perl. To enable this, build Perl with the Configure option
1518 C<-Accflags=-DDL_UNLOAD_ALL_AT_EXIT>. (This maybe useful if you are
1519 using Apache with mod_perl.)
1523 $PERL_VERSION now stands for C<$^V> (a string value) rather than for C<$]>
1528 Env now supports accessing environment variables like PATH as array
1533 More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
1534 large file (more than 4GB) access (NOTE: the O_LARGEFILE is
1535 automatically added to sysopen() flags if large file support has been
1536 configured, as is the default), Free/Net/OpenBSD locking behaviour
1537 flags F_FLOCK, F_POSIX, Linux F_SHLCK, and O_ACCMODE: the combined
1538 mask of O_RDONLY, O_WRONLY, and O_RDWR. The seek()/sysseek()
1539 constants SEEK_SET, SEEK_CUR, and SEEK_END are available via the
1540 C<:seek> tag. The chmod()/stat() S_IF* constants and S_IS* functions
1541 are available via the C<:mode> tag.
1545 A compare_text() function has been added, which allows custom
1546 comparison functions. See L<File::Compare>.
1550 File::Find now works correctly when the wanted() function is either
1551 autoloaded or is a symbolic reference.
1553 A bug that caused File::Find to lose track of the working directory
1554 when pruning top-level directories has been fixed.
1556 File::Find now also supports several other options to control its
1557 behavior. It can follow symbolic links if the C<follow> option is
1558 specified. Enabling the C<no_chdir> option will make File::Find skip
1559 changing the current directory when walking directories. The C<untaint>
1560 flag can be useful when running with taint checks enabled.
1566 This extension implements BSD-style file globbing. By default,
1567 it will also be used for the internal implementation of the glob()
1568 operator. See L<File::Glob>.
1572 New methods have been added to the File::Spec module: devnull() returns
1573 the name of the null device (/dev/null on Unix) and tmpdir() the name of
1574 the temp directory (normally /tmp on Unix). There are now also methods
1575 to convert between absolute and relative filenames: abs2rel() and
1576 rel2abs(). For compatibility with operating systems that specify volume
1577 names in file paths, the splitpath(), splitdir(), and catdir() methods
1580 =item File::Spec::Functions
1582 The new File::Spec::Functions modules provides a function interface
1583 to the File::Spec module. Allows shorthand
1585 $fullname = catfile($dir1, $dir2, $file);
1589 $fullname = File::Spec->catfile($dir1, $dir2, $file);
1593 Getopt::Long licensing has changed to allow the Perl Artistic License
1594 as well as the GPL. It used to be GPL only, which got in the way of
1595 non-GPL applications that wanted to use Getopt::Long.
1597 Getopt::Long encourages the use of Pod::Usage to produce help
1598 messages. For example:
1604 GetOptions('help|?' => \$help, man => \$man) or pod2usage(2);
1605 pod2usage(1) if $help;
1606 pod2usage(-exitstatus => 0, -verbose => 2) if $man;
1612 sample - Using Getopt::Long and Pod::Usage
1616 sample [options] [file ...]
1619 -help brief help message
1620 -man full documentation
1628 Print a brief help message and exits.
1632 Prints the manual page and exits.
1638 B<This program> will read the given input file(s) and do something
1639 useful with the contents thereof.
1643 See L<Pod::Usage> for details.
1645 A bug that prevented the non-option call-back <> from being
1646 specified as the first argument has been fixed.
1648 To specify the characters < and > as option starters, use ><. Note,
1649 however, that changing option starters is strongly deprecated.
1653 write() and syswrite() will now accept a single-argument
1654 form of the call, for consistency with Perl's syswrite().
1656 You can now create a TCP-based IO::Socket::INET without forcing
1657 a connect attempt. This allows you to configure its options
1658 (like making it non-blocking) and then call connect() manually.
1660 A bug that prevented the IO::Socket::protocol() accessor
1661 from ever returning the correct value has been corrected.
1663 IO::Socket::connect now uses non-blocking IO instead of alarm()
1664 to do connect timeouts.
1666 IO::Socket::accept now uses select() instead of alarm() for doing
1669 IO::Socket::INET->new now sets $! correctly on failure. $@ is
1670 still set for backwards compatibility.
1674 Java Perl Lingo is now distributed with Perl. See jpl/README
1675 for more information.
1679 C<use lib> now weeds out any trailing duplicate entries.
1680 C<no lib> removes all named entries.
1684 The bitwise operations C<<< << >>>, C<<< >> >>>, C<&>, C<|>,
1685 and C<~> are now supported on bigints.
1689 The accessor methods Re, Im, arg, abs, rho, and theta can now also
1690 act as mutators (accessor $z->Re(), mutator $z->Re(3)).
1692 The class method C<display_format> and the corresponding object method
1693 C<display_format>, in addition to accepting just one argument, now can
1694 also accept a parameter hash. Recognized keys of a parameter hash are
1695 C<"style">, which corresponds to the old one parameter case, and two
1696 new parameters: C<"format">, which is a printf()-style format string
1697 (defaults usually to C<"%.15g">, you can revert to the default by
1698 setting the format string to C<undef>) used for both parts of a
1699 complex number, and C<"polar_pretty_print"> (defaults to true),
1700 which controls whether an attempt is made to try to recognize small
1701 multiples and rationals of pi (2pi, pi/2) at the argument (angle) of a
1702 polar complex number.
1704 The potentially disruptive change is that in list context both methods
1705 now I<return the parameter hash>, instead of only the value of the
1706 C<"style"> parameter.
1710 A little bit of radial trigonometry (cylindrical and spherical),
1711 radial coordinate conversions, and the great circle distance were added.
1713 =item Pod::Parser, Pod::InputObjects
1715 Pod::Parser is a base class for parsing and selecting sections of
1716 pod documentation from an input stream. This module takes care of
1717 identifying pod paragraphs and commands in the input and hands off the
1718 parsed paragraphs and commands to user-defined methods which are free
1719 to interpret or translate them as they see fit.
1721 Pod::InputObjects defines some input objects needed by Pod::Parser, and
1722 for advanced users of Pod::Parser that need more about a command besides
1725 As of release 5.6.0 of Perl, Pod::Parser is now the officially sanctioned
1726 "base parser code" recommended for use by all pod2xxx translators.
1727 Pod::Text (pod2text) and Pod::Man (pod2man) have already been converted
1728 to use Pod::Parser and efforts to convert Pod::HTML (pod2html) are already
1729 underway. For any questions or comments about pod parsing and translating
1730 issues and utilities, please use the pod-people@perl.org mailing list.
1732 For further information, please see L<Pod::Parser> and L<Pod::InputObjects>.
1734 =item Pod::Checker, podchecker
1736 This utility checks pod files for correct syntax, according to
1737 L<perlpod>. Obvious errors are flagged as such, while warnings are
1738 printed for mistakes that can be handled gracefully. The checklist is
1739 not complete yet. See L<Pod::Checker>.
1741 =item Pod::ParseUtils, Pod::Find
1743 These modules provide a set of gizmos that are useful mainly for pod
1744 translators. L<Pod::Find|Pod::Find> traverses directory structures and
1745 returns found pod files, along with their canonical names (like
1746 C<File::Spec::Unix>). L<Pod::ParseUtils|Pod::ParseUtils> contains
1747 B<Pod::List> (useful for storing pod list information), B<Pod::Hyperlink>
1748 (for parsing the contents of C<LE<lt>E<gt>> sequences) and B<Pod::Cache>
1749 (for caching information about pod files, e.g., link nodes).
1751 =item Pod::Select, podselect
1753 Pod::Select is a subclass of Pod::Parser which provides a function
1754 named "podselect()" to filter out user-specified sections of raw pod
1755 documentation from an input stream. podselect is a script that provides
1756 access to Pod::Select from other scripts to be used as a filter.
1759 =item Pod::Usage, pod2usage
1761 Pod::Usage provides the function "pod2usage()" to print usage messages for
1762 a Perl script based on its embedded pod documentation. The pod2usage()
1763 function is generally useful to all script authors since it lets them
1764 write and maintain a single source (the pods) for documentation, thus
1765 removing the need to create and maintain redundant usage message text
1766 consisting of information already in the pods.
1768 There is also a pod2usage script which can be used from other kinds of
1769 scripts to print usage messages from pods (even for non-Perl scripts
1770 with pods embedded in comments).
1772 For details and examples, please see L<Pod::Usage>.
1774 =item Pod::Text and Pod::Man
1776 Pod::Text has been rewritten to use Pod::Parser. While pod2text() is
1777 still available for backwards compatibility, the module now has a new
1778 preferred interface. See L<Pod::Text> for the details. The new Pod::Text
1779 module is easily subclassed for tweaks to the output, and two such
1780 subclasses (Pod::Text::Termcap for man-page-style bold and underlining
1781 using termcap information, and Pod::Text::Color for markup with ANSI color
1782 sequences) are now standard.
1784 pod2man has been turned into a module, Pod::Man, which also uses
1785 Pod::Parser. In the process, several outstanding bugs related to quotes
1786 in section headers, quoting of code escapes, and nested lists have been
1787 fixed. pod2man is now a wrapper script around this module.
1791 An EXISTS method has been added to this module (and sdbm_exists() has
1792 been added to the underlying sdbm library), so one can now call exists
1793 on an SDBM_File tied hash and get the correct result, rather than a
1796 A bug that may have caused data loss when more than one disk block
1797 happens to be read from the database in a single FETCH() has been
1802 Sys::Syslog now uses XSUBs to access facilities from syslog.h so it
1803 no longer requires syslog.ph to exist.
1807 Sys::Hostname now uses XSUBs to call the C library's gethostname() or
1808 uname() if they exist.
1810 =item Term::ANSIColor
1812 Term::ANSIColor is a very simple module to provide easy and readable
1813 access to the ANSI color and highlighting escape sequences, supported by
1814 most ANSI terminal emulators. It is now included standard.
1818 The timelocal() and timegm() functions used to silently return bogus
1819 results when the date fell outside the machine's integer range. They
1820 now consistently croak() if the date falls in an unsupported range.
1824 The error return value in list context has been changed for all functions
1825 that return a list of values. Previously these functions returned a list
1826 with a single element C<undef> if an error occurred. Now these functions
1827 return the empty list in these situations. This applies to the following
1833 The remaining functions are unchanged and continue to return C<undef> on
1834 error even in list context.
1836 The Win32::SetLastError(ERROR) function has been added as a complement
1837 to the Win32::GetLastError() function.
1839 The new Win32::GetFullPathName(FILENAME) returns the full absolute
1840 pathname for FILENAME in scalar context. In list context it returns
1841 a two-element list containing the fully qualified directory name and
1842 the filename. See L<Win32>.
1846 The XSLoader extension is a simpler alternative to DynaLoader.
1851 A new feature called "DBM Filters" has been added to all the
1852 DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
1853 DBM Filters add four new methods to each DBM module:
1860 These can be used to filter key-value pairs before the pairs are
1861 written to the database or just after they are read from the database.
1862 See L<perldbmfilter> for further information.
1868 C<use attrs> is now obsolete, and is only provided for
1869 backward-compatibility. It's been replaced by the C<sub : attributes>
1870 syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>.
1872 Lexical warnings pragma, C<use warnings;>, to control optional warnings.
1875 C<use filetest> to control the behaviour of filetests (C<-r> C<-w>
1876 ...). Currently only one subpragma implemented, "use filetest
1877 'access';", that uses access(2) or equivalent to check permissions
1878 instead of using stat(2) as usual. This matters in filesystems
1879 where there are ACLs (access control lists): the stat(2) might lie,
1880 but access(2) knows better.
1882 The C<open> pragma can be used to specify default disciplines for
1883 handle constructors (e.g. open()) and for qx//. The two
1884 pseudo-disciplines C<:raw> and C<:crlf> are currently supported on
1885 DOS-derivative platforms (i.e. where binmode is not a no-op).
1886 See also L</"binmode() can be used to set :crlf and :raw modes">.
1888 =head1 Utility Changes
1892 C<dprofpp> is used to display profile data generated using C<Devel::DProf>.
1897 The C<find2perl> utility now uses the enhanced features of the File::Find
1898 module. The -depth and -follow options are supported. Pod documentation
1899 is also included in the script.
1903 The C<h2xs> tool can now work in conjunction with C<C::Scan> (available
1904 from CPAN) to automatically parse real-life header files. The C<-M>,
1905 C<-a>, C<-k>, and C<-o> options are new.
1909 C<perlcc> now supports the C and Bytecode backends. By default,
1910 it generates output from the simple C backend rather than the
1911 optimized C backend.
1913 Support for non-Unix platforms has been improved.
1917 C<perldoc> has been reworked to avoid possible security holes.
1918 It will not by default let itself be run as the superuser, but you
1919 may still use the B<-U> switch to try to make it drop privileges
1922 =head2 The Perl Debugger
1924 Many bug fixes and enhancements were added to F<perl5db.pl>, the
1925 Perl debugger. The help documentation was rearranged. New commands
1926 include C<< < ? >>, C<< > ? >>, and C<< { ? >> to list out current
1927 actions, C<man I<docpage>> to run your doc viewer on some perl
1928 docset, and support for quoted options. The help information was
1929 rearranged, and should be viewable once again if you're using B<less>
1930 as your pager. A serious security hole was plugged--you should
1931 immediately remove all older versions of the Perl debugger as
1932 installed in previous releases, all the way back to perl3, from
1933 your system to avoid being bitten by this.
1935 =head1 Improved Documentation
1937 Many of the platform-specific README files are now part of the perl
1938 installation. See L<perl> for the complete list.
1944 The official list of public Perl API functions.
1948 A tutorial for beginners on object-oriented Perl.
1950 =item perlcompile.pod
1952 An introduction to using the Perl Compiler suite.
1954 =item perldbmfilter.pod
1956 A howto document on using the DBM filter facility.
1960 All material unrelated to running the Perl debugger, plus all
1961 low-level guts-like details that risked crushing the casual user
1962 of the debugger, have been relocated from the old manpage to the
1965 =item perldebguts.pod
1967 This new manpage contains excessively low-level material not related
1968 to the Perl debugger, but slightly related to debugging Perl itself.
1969 It also contains some arcane internal details of how the debugging
1970 process works that may only be of interest to developers of Perl
1975 Notes on the fork() emulation currently available for the Windows platform.
1977 =item perlfilter.pod
1979 An introduction to writing Perl source filters.
1983 Some guidelines for hacking the Perl source code.
1985 =item perlintern.pod
1987 A list of internal functions in the Perl source code.
1988 (List is currently empty.)
1990 =item perllexwarn.pod
1992 Introduction and reference information about lexically scoped
1995 =item perlnumber.pod
1997 Detailed information about numbers as they are represented in Perl.
1999 =item perlopentut.pod
2001 A tutorial on using open() effectively.
2003 =item perlreftut.pod
2005 A tutorial that introduces the essentials of references.
2009 A tutorial on managing class data for object modules.
2013 Discussion of the most often wanted features that may someday be
2016 =item perlunicode.pod
2018 An introduction to Unicode support features in Perl.
2022 =head1 Performance enhancements
2024 =head2 Simple sort() using { $a <=> $b } and the like are optimized
2026 Many common sort() operations using a simple inlined block are now
2027 optimized for faster performance.
2029 =head2 Optimized assignments to lexical variables
2031 Certain operations in the RHS of assignment statements have been
2032 optimized to directly set the lexical variable on the LHS,
2033 eliminating redundant copying overheads.
2035 =head2 Faster subroutine calls
2037 Minor changes in how subroutine calls are handled internally
2038 provide marginal improvements in performance.
2040 =head2 delete(), each(), values() and hash iteration are faster
2042 The hash values returned by delete(), each(), values() and hashes in a
2043 list context are the actual values in the hash, instead of copies.
2044 This results in significantly better performance, because it eliminates
2045 needless copying in most situations.
2047 =head1 Installation and Configuration Improvements
2049 =head2 -Dusethreads means something different
2051 The -Dusethreads flag now enables the experimental interpreter-based thread
2052 support by default. To get the flavor of experimental threads that was in
2053 5.005 instead, you need to run Configure with "-Dusethreads -Duse5005threads".
2055 As of v5.6.0, interpreter-threads support is still lacking a way to
2056 create new threads from Perl (i.e., C<use Thread;> will not work with
2057 interpreter threads). C<use Thread;> continues to be available when you
2058 specify the -Duse5005threads option to Configure, bugs and all.
2060 NOTE: Support for threads continues to be an experimental feature.
2061 Interfaces and implementation are subject to sudden and drastic changes.
2063 =head2 New Configure flags
2065 The following new flags may be enabled on the Configure command line
2066 by running Configure with C<-Dflag>.
2069 usethreads useithreads (new interpreter threads: no Perl API yet)
2070 usethreads use5005threads (threads as they were in 5.005)
2072 use64bitint (equal to now deprecated 'use64bits')
2078 usesocks (only SOCKS v5 supported)
2080 =head2 Threadedness and 64-bitness now more daring
2082 The Configure options enabling the use of threads and the use of
2083 64-bitness are now more daring in the sense that they no more have an
2084 explicit list of operating systems of known threads/64-bit
2085 capabilities. In other words: if your operating system has the
2086 necessary APIs and datatypes, you should be able just to go ahead and
2087 use them, for threads by Configure -Dusethreads, and for 64 bits
2088 either explicitly by Configure -Duse64bitint or implicitly if your
2089 system has 64-bit wide datatypes. See also L<"64-bit support">.
2093 Some platforms have "long doubles", floating point numbers of even
2094 larger range than ordinary "doubles". To enable using long doubles for
2095 Perl's scalars, use -Duselongdouble.
2097 =head2 -Dusemorebits
2099 You can enable both -Duse64bitint and -Duselongdouble with -Dusemorebits.
2100 See also L<"64-bit support">.
2102 =head2 -Duselargefiles
2104 Some platforms support system APIs that are capable of handling large files
2105 (typically, files larger than two gigabytes). Perl will try to use these
2106 APIs if you ask for -Duselargefiles.
2108 See L<"Large file support"> for more information.
2110 =head2 installusrbinperl
2112 You can use "Configure -Uinstallusrbinperl" which causes installperl
2113 to skip installing perl also as /usr/bin/perl. This is useful if you
2114 prefer not to modify /usr/bin for some reason or another but harmful
2115 because many scripts assume to find Perl in /usr/bin/perl.
2117 =head2 SOCKS support
2119 You can use "Configure -Dusesocks" which causes Perl to probe
2120 for the SOCKS proxy protocol library (v5, not v4). For more information
2123 http://www.socks.nec.com/
2127 You can "post-edit" the Configure variables using the Configure C<-A>
2128 switch. The editing happens immediately after the platform specific
2129 hints files have been processed but before the actual configuration
2130 process starts. Run C<Configure -h> to find out the full C<-A> syntax.
2132 =head2 Enhanced Installation Directories
2134 The installation structure has been enriched to improve the support
2135 for maintaining multiple versions of perl, to provide locations for
2136 vendor-supplied modules, scripts, and manpages, and to ease maintenance
2137 of locally-added modules, scripts, and manpages. See the section on
2138 Installation Directories in the INSTALL file for complete details.
2139 For most users building and installing from source, the defaults should
2142 If you previously used C<Configure -Dsitelib> or C<-Dsitearch> to set
2143 special values for library directories, you might wish to consider using
2144 the new C<-Dsiteprefix> setting instead. Also, if you wish to re-use a
2145 config.sh file from an earlier version of perl, you should be sure to
2146 check that Configure makes sensible choices for the new directories.
2147 See INSTALL for complete details.
2149 =head2 gcc automatically tried if 'cc' does not seem to be working
2151 In many platforms the vendor-supplied 'cc' is too stripped-down to
2152 build Perl (basically, the 'cc' doesn't do ANSI C). If this seems
2153 to be the case and the 'cc' does not seem to be the GNU C compiler
2154 'gcc', an automatic attempt is made to find and use 'gcc' instead.
2156 =head1 Platform specific changes
2158 =head2 Supported platforms
2164 The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
2169 GNU/Hurd is now supported.
2173 Rhapsody/Darwin is now supported.
2177 EPOC is now supported (on Psion 5).
2181 The cygwin port (formerly cygwin32) has been greatly improved.
2191 Perl now works with djgpp 2.02 (and 2.03 alpha).
2195 Environment variable names are not converted to uppercase any more.
2199 Incorrect exit codes from backticks have been fixed.
2203 This port continues to use its own builtin globbing (not File::Glob).
2207 =head2 OS390 (OpenEdition MVS)
2209 Support for this EBCDIC platform has not been renewed in this release.
2210 There are difficulties in reconciling Perl's standardization on UTF-8
2211 as its internal representation for characters with the EBCDIC character
2212 set, because the two are incompatible.
2214 It is unclear whether future versions will renew support for this
2215 platform, but the possibility exists.
2219 Numerous revisions and extensions to configuration, build, testing, and
2220 installation process to accommodate core changes and VMS-specific options.
2222 Expand %ENV-handling code to allow runtime mapping to logical names,
2223 CLI symbols, and CRTL environ array.
2225 Extension of subprocess invocation code to accept filespecs as command
2228 Add to Perl command line processing the ability to use default file types and
2229 to recognize Unix-style C<2E<gt>&1>.
2231 Expansion of File::Spec::VMS routines, and integration into ExtUtils::MM_VMS.
2233 Extension of ExtUtils::MM_VMS to handle complex extensions more flexibly.
2235 Barewords at start of Unix-syntax paths may be treated as text rather than
2236 only as logical names.
2238 Optional secure translation of several logical names used internally by Perl.
2240 Miscellaneous bugfixing and porting of new core code to VMS.
2242 Thanks are gladly extended to the many people who have contributed VMS
2243 patches, testing, and ideas.
2247 Perl can now emulate fork() internally, using multiple interpreters running
2248 in different concurrent threads. This support must be enabled at build
2249 time. See L<perlfork> for detailed information.
2251 When given a pathname that consists only of a drivename, such as C<A:>,
2252 opendir() and stat() now use the current working directory for the drive
2253 rather than the drive root.
2255 The builtin XSUB functions in the Win32:: namespace are documented. See
2258 $^X now contains the full path name of the running executable.
2260 A Win32::GetLongPathName() function is provided to complement
2261 Win32::GetFullPathName() and Win32::GetShortPathName(). See L<Win32>.
2263 POSIX::uname() is supported.
2265 system(1,...) now returns true process IDs rather than process
2266 handles. kill() accepts any real process id, rather than strictly
2267 return values from system(1,...).
2269 For better compatibility with Unix, C<kill(0, $pid)> can now be used to
2270 test whether a process exists.
2272 The C<Shell> module is supported.
2274 Better support for building Perl under command.com in Windows 95
2277 Scripts are read in binary mode by default to allow ByteLoader (and
2278 the filter mechanism in general) to work properly. For compatibility,
2279 the DATA filehandle will be set to text mode if a carriage return is
2280 detected at the end of the line containing the __END__ or __DATA__
2281 token; if not, the DATA filehandle will be left open in binary mode.
2282 Earlier versions always opened the DATA filehandle in text mode.
2284 The glob() operator is implemented via the C<File::Glob> extension,
2285 which supports glob syntax of the C shell. This increases the flexibility
2286 of the glob() operator, but there may be compatibility issues for
2287 programs that relied on the older globbing syntax. If you want to
2288 preserve compatibility with the older syntax, you might want to run
2289 perl with C<-MFile::DosGlob>. For details and compatibility information,
2292 =head1 Significant bug fixes
2294 =head2 <HANDLE> on empty files
2296 With C<$/> set to C<undef>, "slurping" an empty file returns a string of
2297 zero length (instead of C<undef>, as it used to) the first time the
2298 HANDLE is read after C<$/> is set to C<undef>. Further reads yield
2301 This means that the following will append "foo" to an empty file (it used
2304 perl -0777 -pi -e 's/^/foo/' empty_file
2308 perl -pi -e 's/^/foo/' empty_file
2310 is unchanged (it continues to leave the file empty).
2312 =head2 C<eval '...'> improvements
2314 Line numbers (as reflected by caller() and most diagnostics) within
2315 C<eval '...'> were often incorrect where here documents were involved.
2316 This has been corrected.
2318 Lexical lookups for variables appearing in C<eval '...'> within
2319 functions that were themselves called within an C<eval '...'> were
2320 searching the wrong place for lexicals. The lexical search now
2321 correctly ends at the subroutine's block boundary.
2323 The use of C<return> within C<eval {...}> caused $@ not to be reset
2324 correctly when no exception occurred within the eval. This has
2327 Parsing of here documents used to be flawed when they appeared as
2328 the replacement expression in C<eval 's/.../.../e'>. This has
2331 =head2 All compilation errors are true errors
2333 Some "errors" encountered at compile time were by necessity
2334 generated as warnings followed by eventual termination of the
2335 program. This enabled more such errors to be reported in a
2336 single run, rather than causing a hard stop at the first error
2337 that was encountered.
2339 The mechanism for reporting such errors has been reimplemented
2340 to queue compile-time errors and report them at the end of the
2341 compilation as true errors rather than as warnings. This fixes
2342 cases where error messages leaked through in the form of warnings
2343 when code was compiled at run time using C<eval STRING>, and
2344 also allows such errors to be reliably trapped using C<eval "...">.
2346 =head2 Implicitly closed filehandles are safer
2348 Sometimes implicitly closed filehandles (as when they are localized,
2349 and Perl automatically closes them on exiting the scope) could
2350 inadvertently set $? or $!. This has been corrected.
2353 =head2 Behavior of list slices is more consistent
2355 When taking a slice of a literal list (as opposed to a slice of
2356 an array or hash), Perl used to return an empty list if the
2357 result happened to be composed of all undef values.
2359 The new behavior is to produce an empty list if (and only if)
2360 the original list was empty. Consider the following example:
2362 @a = (1,undef,undef,2)[2,1,2];
2364 The old behavior would have resulted in @a having no elements.
2365 The new behavior ensures it has three undefined elements.
2367 Note in particular that the behavior of slices of the following
2368 cases remains unchanged:
2371 @a = (getpwent)[7,0];
2372 @a = (anything_returning_empty_list())[2,1,2];
2374 @a = @c{'a','b','c'};
2378 =head2 C<(\$)> prototype and C<$foo{a}>
2380 A scalar reference prototype now correctly allows a hash or
2381 array element in that slot.
2383 =head2 C<goto &sub> and AUTOLOAD
2385 The C<goto &sub> construct works correctly when C<&sub> happens
2388 =head2 C<-bareword> allowed under C<use integer>
2390 The autoquoting of barewords preceded by C<-> did not work
2391 in prior versions when the C<integer> pragma was enabled.
2392 This has been fixed.
2394 =head2 Failures in DESTROY()
2396 When code in a destructor threw an exception, it went unnoticed
2397 in earlier versions of Perl, unless someone happened to be
2398 looking in $@ just after the point the destructor happened to
2399 run. Such failures are now visible as warnings when warnings are
2402 =head2 Locale bugs fixed
2404 printf() and sprintf() previously reset the numeric locale
2405 back to the default "C" locale. This has been fixed.
2407 Numbers formatted according to the local numeric locale
2408 (such as using a decimal comma instead of a decimal dot) caused
2409 "isn't numeric" warnings, even while the operations accessing
2410 those numbers produced correct results. These warnings have been
2415 The C<eval 'return sub {...}'> construct could sometimes leak
2416 memory. This has been fixed.
2418 Operations that aren't filehandle constructors used to leak memory
2419 when used on invalid filehandles. This has been fixed.
2421 Constructs that modified C<@_> could fail to deallocate values
2422 in C<@_> and thus leak memory. This has been corrected.
2424 =head2 Spurious subroutine stubs after failed subroutine calls
2426 Perl could sometimes create empty subroutine stubs when a
2427 subroutine was not found in the package. Such cases stopped
2428 later method lookups from progressing into base packages.
2429 This has been corrected.
2431 =head2 Taint failures under C<-U>
2433 When running in unsafe mode, taint violations could sometimes
2434 cause silent failures. This has been fixed.
2436 =head2 END blocks and the C<-c> switch
2438 Prior versions used to run BEGIN B<and> END blocks when Perl was
2439 run in compile-only mode. Since this is typically not the expected
2440 behavior, END blocks are not executed anymore when the C<-c> switch
2441 is used, or if compilation fails.
2443 See L</"Support for CHECK blocks"> for how to run things when the compile
2446 =head2 Potential to leak DATA filehandles
2448 Using the C<__DATA__> token creates an implicit filehandle to
2449 the file that contains the token. It is the program's
2450 responsibility to close it when it is done reading from it.
2452 This caveat is now better explained in the documentation.
2455 =head1 New or Changed Diagnostics
2459 =item "%s" variable %s masks earlier declaration in same %s
2461 (W misc) A "my" or "our" variable has been redeclared in the current scope or statement,
2462 effectively eliminating all access to the previous instance. This is almost
2463 always a typographical error. Note that the earlier variable will still exist
2464 until the end of the scope or until all closure referents to it are
2467 =item "my sub" not yet implemented
2469 (F) Lexically scoped subroutines are not yet implemented. Don't try that
2472 =item "our" variable %s redeclared
2474 (W misc) You seem to have already declared the same global once before in the
2475 current lexical scope.
2477 =item '!' allowed only after types %s
2479 (F) The '!' is allowed in pack() and unpack() only after certain types.
2480 See L<perlfunc/pack>.
2482 =item / cannot take a count
2484 (F) You had an unpack template indicating a counted-length string,
2485 but you have also specified an explicit size for the string.
2486 See L<perlfunc/pack>.
2488 =item / must be followed by a, A or Z
2490 (F) You had an unpack template indicating a counted-length string,
2491 which must be followed by one of the letters a, A or Z
2492 to indicate what sort of string is to be unpacked.
2493 See L<perlfunc/pack>.
2495 =item / must be followed by a*, A* or Z*
2497 (F) You had a pack template indicating a counted-length string,
2498 Currently the only things that can have their length counted are a*, A* or Z*.
2499 See L<perlfunc/pack>.
2501 =item / must follow a numeric type
2503 (F) You had an unpack template that contained a '#',
2504 but this did not follow some numeric unpack specification.
2505 See L<perlfunc/pack>.
2507 =item /%s/: Unrecognized escape \\%c passed through
2509 (W regexp) You used a backslash-character combination which is not recognized
2510 by Perl. This combination appears in an interpolated variable or a
2511 C<'>-delimited regular expression. The character was understood literally.
2513 =item /%s/: Unrecognized escape \\%c in character class passed through
2515 (W regexp) You used a backslash-character combination which is not recognized
2516 by Perl inside character classes. The character was understood literally.
2518 =item /%s/ should probably be written as "%s"
2520 (W syntax) You have used a pattern where Perl expected to find a string,
2521 as in the first argument to C<join>. Perl will treat the true
2522 or false result of matching the pattern against $_ as the string,
2523 which is probably not what you had in mind.
2525 =item %s() called too early to check prototype
2527 (W prototype) You've called a function that has a prototype before the parser saw a
2528 definition or declaration for it, and Perl could not check that the call
2529 conforms to the prototype. You need to either add an early prototype
2530 declaration for the subroutine in question, or move the subroutine
2531 definition ahead of the call to get proper prototype checking. Alternatively,
2532 if you are certain that you're calling the function correctly, you may put
2533 an ampersand before the name to avoid the warning. See L<perlsub>.
2535 =item %s argument is not a HASH or ARRAY element
2537 (F) The argument to exists() must be a hash or array element, such as:
2542 =item %s argument is not a HASH or ARRAY element or slice
2544 (F) The argument to delete() must be either a hash or array element, such as:
2549 or a hash or array slice, such as:
2551 @foo[$bar, $baz, $xyzzy]
2552 @{$ref->[12]}{"susie", "queue"}
2554 =item %s argument is not a subroutine name
2556 (F) The argument to exists() for C<exists &sub> must be a subroutine
2557 name, and not a subroutine call. C<exists &sub()> will generate this error.
2559 =item %s package attribute may clash with future reserved word: %s
2561 (W reserved) A lowercase attribute name was used that had a package-specific handler.
2562 That name might have a meaning to Perl itself some day, even though it
2563 doesn't yet. Perhaps you should use a mixed-case attribute name, instead.
2566 =item (in cleanup) %s
2568 (W misc) This prefix usually indicates that a DESTROY() method raised
2569 the indicated exception. Since destructors are usually called by
2570 the system at arbitrary points during execution, and often a vast
2571 number of times, the warning is issued only once for any number
2572 of failures that would otherwise result in the same message being
2575 Failure of user callbacks dispatched using the C<G_KEEPERR> flag
2576 could also result in this warning. See L<perlcall/G_KEEPERR>.
2578 =item <> should be quotes
2580 (F) You wrote C<< require <file> >> when you should have written
2583 =item Attempt to join self
2585 (F) You tried to join a thread from within itself, which is an
2586 impossible task. You may be joining the wrong thread, or you may
2587 need to move the join() to some other thread.
2589 =item Bad evalled substitution pattern
2591 (F) You've used the /e switch to evaluate the replacement for a
2592 substitution, but perl found a syntax error in the code to evaluate,
2593 most likely an unexpected right brace '}'.
2595 =item Bad realloc() ignored
2597 (S) An internal routine called realloc() on something that had never been
2598 malloc()ed in the first place. Mandatory, but can be disabled by
2599 setting environment variable C<PERL_BADFREE> to 1.
2601 =item Bareword found in conditional
2603 (W bareword) The compiler found a bareword where it expected a conditional,
2604 which often indicates that an || or && was parsed as part of the
2605 last argument of the previous construct, for example:
2609 It may also indicate a misspelled constant that has been interpreted
2612 use constant TYPO => 1;
2613 if (TYOP) { print "foo" }
2615 The C<strict> pragma is useful in avoiding such errors.
2617 =item Binary number > 0b11111111111111111111111111111111 non-portable
2619 (W portable) The binary number you specified is larger than 2**32-1
2620 (4294967295) and therefore non-portable between systems. See
2621 L<perlport> for more on portability concerns.
2623 =item Bit vector size > 32 non-portable
2625 (W portable) Using bit vector sizes larger than 32 is non-portable.
2627 =item Buffer overflow in prime_env_iter: %s
2629 (W internal) A warning peculiar to VMS. While Perl was preparing to iterate over
2630 %ENV, it encountered a logical name or symbol definition which was too long,
2631 so it was truncated to the string shown.
2633 =item Can't check filesystem of script "%s"
2635 (P) For some reason you can't check the filesystem of the script for nosuid.
2637 =item Can't declare class for non-scalar %s in "%s"
2639 (S) Currently, only scalar variables can declared with a specific class
2640 qualifier in a "my" or "our" declaration. The semantics may be extended
2641 for other types of variables in future.
2643 =item Can't declare %s in "%s"
2645 (F) Only scalar, array, and hash variables may be declared as "my" or
2646 "our" variables. They must have ordinary identifiers as names.
2648 =item Can't ignore signal CHLD, forcing to default
2650 (W signal) Perl has detected that it is being run with the SIGCHLD signal
2651 (sometimes known as SIGCLD) disabled. Since disabling this signal
2652 will interfere with proper determination of exit status of child
2653 processes, Perl has reset the signal to its default value.
2654 This situation typically indicates that the parent program under
2655 which Perl may be running (e.g., cron) is being very careless.
2657 =item Can't modify non-lvalue subroutine call
2659 (F) Subroutines meant to be used in lvalue context should be declared as
2660 such, see L<perlsub/"Lvalue subroutines">.
2662 =item Can't read CRTL environ
2664 (S) A warning peculiar to VMS. Perl tried to read an element of %ENV
2665 from the CRTL's internal environment array and discovered the array was
2666 missing. You need to figure out where your CRTL misplaced its environ
2667 or define F<PERL_ENV_TABLES> (see L<perlvms>) so that environ is not searched.
2669 =item Can't remove %s: %s, skipping file
2671 (S) You requested an inplace edit without creating a backup file. Perl
2672 was unable to remove the original file to replace it with the modified
2673 file. The file was left unmodified.
2675 =item Can't return %s from lvalue subroutine
2677 (F) Perl detected an attempt to return illegal lvalues (such
2678 as temporary or readonly values) from a subroutine used as an lvalue.
2679 This is not allowed.
2681 =item Can't weaken a nonreference
2683 (F) You attempted to weaken something that was not a reference. Only
2684 references can be weakened.
2686 =item Character class [:%s:] unknown
2688 (F) The class in the character class [: :] syntax is unknown.
2691 =item Character class syntax [%s] belongs inside character classes
2693 (W unsafe) The character class constructs [: :], [= =], and [. .] go
2694 I<inside> character classes, the [] are part of the construct,
2695 for example: /[012[:alpha:]345]/. Note that [= =] and [. .]
2696 are not currently implemented; they are simply placeholders for
2699 =item Constant is not %s reference
2701 (F) A constant value (perhaps declared using the C<use constant> pragma)
2702 is being dereferenced, but it amounts to the wrong type of reference. The
2703 message indicates the type of reference that was expected. This usually
2704 indicates a syntax error in dereferencing the constant value.
2705 See L<perlsub/"Constant Functions"> and L<constant>.
2707 =item constant(%s): %s
2709 (F) The parser found inconsistencies either while attempting to define an
2710 overloaded constant, or when trying to find the character name specified
2711 in the C<\N{...}> escape. Perhaps you forgot to load the corresponding
2712 C<overload> or C<charnames> pragma? See L<charnames> and L<overload>.
2714 =item CORE::%s is not a keyword
2716 (F) The CORE:: namespace is reserved for Perl keywords.
2718 =item defined(@array) is deprecated
2720 (D) defined() is not usually useful on arrays because it checks for an
2721 undefined I<scalar> value. If you want to see if the array is empty,
2722 just use C<if (@array) { # not empty }> for example.
2724 =item defined(%hash) is deprecated
2726 (D) defined() is not usually useful on hashes because it checks for an
2727 undefined I<scalar> value. If you want to see if the hash is empty,
2728 just use C<if (%hash) { # not empty }> for example.
2730 =item Did not produce a valid header
2734 =item (Did you mean "local" instead of "our"?)
2736 (W misc) Remember that "our" does not localize the declared global variable.
2737 You have declared it again in the same lexical scope, which seems superfluous.
2739 =item Document contains no data
2743 =item entering effective %s failed
2745 (F) While under the C<use filetest> pragma, switching the real and
2746 effective uids or gids failed.
2748 =item false [] range "%s" in regexp
2750 (W regexp) A character class range must start and end at a literal character, not
2751 another character class like C<\d> or C<[:alpha:]>. The "-" in your false
2752 range is interpreted as a literal "-". Consider quoting the "-", "\-".
2755 =item Filehandle %s opened only for output
2757 (W io) You tried to read from a filehandle opened only for writing. If you
2758 intended it to be a read/write filehandle, you needed to open it with
2759 "+<" or "+>" or "+>>" instead of with "<" or nothing. If
2760 you intended only to read from the file, use "<". See
2763 =item flock() on closed filehandle %s
2765 (W closed) The filehandle you're attempting to flock() got itself closed some
2766 time before now. Check your logic flow. flock() operates on filehandles.
2767 Are you attempting to call flock() on a dirhandle by the same name?
2769 =item Global symbol "%s" requires explicit package name
2771 (F) You've said "use strict vars", which indicates that all variables
2772 must either be lexically scoped (using "my"), declared beforehand using
2773 "our", or explicitly qualified to say which package the global variable
2776 =item Hexadecimal number > 0xffffffff non-portable
2778 (W portable) The hexadecimal number you specified is larger than 2**32-1
2779 (4294967295) and therefore non-portable between systems. See
2780 L<perlport> for more on portability concerns.
2782 =item Ill-formed CRTL environ value "%s"
2784 (W internal) A warning peculiar to VMS. Perl tried to read the CRTL's internal
2785 environ array, and encountered an element without the C<=> delimiter
2786 used to separate keys from values. The element is ignored.
2788 =item Ill-formed message in prime_env_iter: |%s|
2790 (W internal) A warning peculiar to VMS. Perl tried to read a logical name
2791 or CLI symbol definition when preparing to iterate over %ENV, and
2792 didn't see the expected delimiter between key and value, so the
2795 =item Illegal binary digit %s
2797 (F) You used a digit other than 0 or 1 in a binary number.
2799 =item Illegal binary digit %s ignored
2801 (W digit) You may have tried to use a digit other than 0 or 1 in a binary number.
2802 Interpretation of the binary number stopped before the offending digit.
2804 =item Illegal number of bits in vec
2806 (F) The number of bits in vec() (the third argument) must be a power of
2807 two from 1 to 32 (or 64, if your platform supports that).
2809 =item Integer overflow in %s number
2811 (W overflow) The hexadecimal, octal or binary number you have specified either
2812 as a literal or as an argument to hex() or oct() is too big for your
2813 architecture, and has been converted to a floating point number. On a
2814 32-bit architecture the largest hexadecimal, octal or binary number
2815 representable without overflow is 0xFFFFFFFF, 037777777777, or
2816 0b11111111111111111111111111111111 respectively. Note that Perl
2817 transparently promotes all numbers to a floating point representation
2818 internally--subject to loss of precision errors in subsequent
2821 =item Invalid %s attribute: %s
2823 The indicated attribute for a subroutine or variable was not recognized
2824 by Perl or by a user-supplied handler. See L<attributes>.
2826 =item Invalid %s attributes: %s
2828 The indicated attributes for a subroutine or variable were not recognized
2829 by Perl or by a user-supplied handler. See L<attributes>.
2831 =item invalid [] range "%s" in regexp
2833 The offending range is now explicitly displayed.
2835 =item Invalid separator character %s in attribute list
2837 (F) Something other than a colon or whitespace was seen between the
2838 elements of an attribute list. If the previous attribute
2839 had a parenthesised parameter list, perhaps that list was terminated
2840 too soon. See L<attributes>.
2842 =item Invalid separator character %s in subroutine attribute list
2844 (F) Something other than a colon or whitespace was seen between the
2845 elements of a subroutine attribute list. If the previous attribute
2846 had a parenthesised parameter list, perhaps that list was terminated
2849 =item leaving effective %s failed
2851 (F) While under the C<use filetest> pragma, switching the real and
2852 effective uids or gids failed.
2854 =item Lvalue subs returning %s not implemented yet
2856 (F) Due to limitations in the current implementation, array and hash
2857 values cannot be returned in subroutines used in lvalue context.
2858 See L<perlsub/"Lvalue subroutines">.
2860 =item Method %s not permitted
2864 =item Missing %sbrace%s on \N{}
2866 (F) Wrong syntax of character name literal C<\N{charname}> within
2867 double-quotish context.
2869 =item Missing command in piped open
2871 (W pipe) You used the C<open(FH, "| command")> or C<open(FH, "command |")>
2872 construction, but the command was missing or blank.
2874 =item Missing name in "my sub"
2876 (F) The reserved syntax for lexically scoped subroutines requires that they
2877 have a name with which they can be found.
2879 =item No %s specified for -%c
2881 (F) The indicated command line switch needs a mandatory argument, but
2882 you haven't specified one.
2884 =item No package name allowed for variable %s in "our"
2886 (F) Fully qualified variable names are not allowed in "our" declarations,
2887 because that doesn't make much sense under existing semantics. Such
2888 syntax is reserved for future extensions.
2890 =item No space allowed after -%c
2892 (F) The argument to the indicated command line switch must follow immediately
2893 after the switch, without intervening spaces.
2895 =item no UTC offset information; assuming local time is UTC
2897 (S) A warning peculiar to VMS. Perl was unable to find the local
2898 timezone offset, so it's assuming that local system time is equivalent
2899 to UTC. If it's not, define the logical name F<SYS$TIMEZONE_DIFFERENTIAL>
2900 to translate to the number of seconds which need to be added to UTC to
2903 =item Octal number > 037777777777 non-portable
2905 (W portable) The octal number you specified is larger than 2**32-1 (4294967295)
2906 and therefore non-portable between systems. See L<perlport> for more
2907 on portability concerns.
2909 See also L<perlport> for writing portable code.
2911 =item panic: del_backref
2913 (P) Failed an internal consistency check while trying to reset a weak
2916 =item panic: kid popen errno read
2918 (F) forked child returned an incomprehensible message about its errno.
2920 =item panic: magic_killbackrefs
2922 (P) Failed an internal consistency check while trying to reset all weak
2923 references to an object.
2925 =item Parentheses missing around "%s" list
2927 (W parenthesis) You said something like
2933 my ($foo, $bar) = @_;
2935 Remember that "my", "our", and "local" bind tighter than comma.
2937 =item Possible unintended interpolation of %s in string
2939 (W ambiguous) It used to be that Perl would try to guess whether you
2940 wanted an array interpolated or a literal @. It no longer does this;
2941 arrays are now I<always> interpolated into strings. This means that
2942 if you try something like:
2944 print "fred@example.com";
2946 and the array C<@example> doesn't exist, Perl is going to print
2947 C<fred.com>, which is probably not what you wanted. To get a literal
2948 C<@> sign in a string, put a backslash before it, just as you would
2949 to get a literal C<$> sign.
2951 =item Possible Y2K bug: %s
2953 (W y2k) You are concatenating the number 19 with another number, which
2954 could be a potential Year 2000 problem.
2956 =item pragma "attrs" is deprecated, use "sub NAME : ATTRS" instead
2958 (W deprecated) You have written something like this:
2962 use attrs qw(locked);
2965 You should use the new declaration syntax instead.
2971 The C<use attrs> pragma is now obsolete, and is only provided for
2972 backward-compatibility. See L<perlsub/"Subroutine Attributes">.
2975 =item Premature end of script headers
2979 =item Repeat count in pack overflows
2981 (F) You can't specify a repeat count so large that it overflows
2982 your signed integers. See L<perlfunc/pack>.
2984 =item Repeat count in unpack overflows
2986 (F) You can't specify a repeat count so large that it overflows
2987 your signed integers. See L<perlfunc/unpack>.
2989 =item realloc() of freed memory ignored
2991 (S) An internal routine called realloc() on something that had already
2994 =item Reference is already weak
2996 (W misc) You have attempted to weaken a reference that is already weak.
2997 Doing so has no effect.
2999 =item setpgrp can't take arguments
3001 (F) Your system has the setpgrp() from BSD 4.2, which takes no arguments,
3002 unlike POSIX setpgid(), which takes a process ID and process group ID.
3004 =item Strange *+?{} on zero-length expression
3006 (W regexp) You applied a regular expression quantifier in a place where it
3007 makes no sense, such as on a zero-width assertion.
3008 Try putting the quantifier inside the assertion instead. For example,
3009 the way to match "abc" provided that it is followed by three
3010 repetitions of "xyz" is C</abc(?=(?:xyz){3})/>, not C</abc(?=xyz){3}/>.
3012 =item switching effective %s is not implemented
3014 (F) While under the C<use filetest> pragma, we cannot switch the
3015 real and effective uids or gids.
3017 =item This Perl can't reset CRTL environ elements (%s)
3019 =item This Perl can't set CRTL environ elements (%s=%s)
3021 (W internal) Warnings peculiar to VMS. You tried to change or delete an element
3022 of the CRTL's internal environ array, but your copy of Perl wasn't
3023 built with a CRTL that contained the setenv() function. You'll need to
3024 rebuild Perl with a CRTL that does, or redefine F<PERL_ENV_TABLES> (see
3025 L<perlvms>) so that the environ array isn't the target of the change to
3026 %ENV which produced the warning.
3028 =item Too late to run %s block
3030 (W void) A CHECK or INIT block is being defined during run time proper,
3031 when the opportunity to run them has already passed. Perhaps you are
3032 loading a file with C<require> or C<do> when you should be using
3033 C<use> instead. Or perhaps you should put the C<require> or C<do>
3034 inside a BEGIN block.
3036 =item Unknown open() mode '%s'
3038 (F) The second argument of 3-argument open() is not among the list
3039 of valid modes: C<< < >>, C<< > >>, C<<< >> >>>, C<< +< >>,
3040 C<< +> >>, C<<< +>> >>>, C<-|>, C<|->.
3042 =item Unknown process %x sent message to prime_env_iter: %s
3044 (P) An error peculiar to VMS. Perl was reading values for %ENV before
3045 iterating over it, and someone else stuck a message in the stream of
3046 data Perl expected. Someone's very confused, or perhaps trying to
3047 subvert Perl's population of %ENV for nefarious purposes.
3049 =item Unrecognized escape \\%c passed through
3051 (W misc) You used a backslash-character combination which is not recognized
3052 by Perl. The character was understood literally.
3054 =item Unterminated attribute parameter in attribute list
3056 (F) The lexer saw an opening (left) parenthesis character while parsing an
3057 attribute list, but the matching closing (right) parenthesis
3058 character was not found. You may need to add (or remove) a backslash
3059 character to get your parentheses to balance. See L<attributes>.
3061 =item Unterminated attribute list
3063 (F) The lexer found something other than a simple identifier at the start
3064 of an attribute, and it wasn't a semicolon or the start of a
3065 block. Perhaps you terminated the parameter list of the previous attribute
3066 too soon. See L<attributes>.
3068 =item Unterminated attribute parameter in subroutine attribute list
3070 (F) The lexer saw an opening (left) parenthesis character while parsing a
3071 subroutine attribute list, but the matching closing (right) parenthesis
3072 character was not found. You may need to add (or remove) a backslash
3073 character to get your parentheses to balance.
3075 =item Unterminated subroutine attribute list
3077 (F) The lexer found something other than a simple identifier at the start
3078 of a subroutine attribute, and it wasn't a semicolon or the start of a
3079 block. Perhaps you terminated the parameter list of the previous attribute
3082 =item Value of CLI symbol "%s" too long
3084 (W misc) A warning peculiar to VMS. Perl tried to read the value of an %ENV
3085 element from a CLI symbol table, and found a resultant string longer
3086 than 1024 characters. The return value has been truncated to 1024
3089 =item Version number must be a constant number
3091 (P) The attempt to translate a C<use Module n.n LIST> statement into
3092 its equivalent C<BEGIN> block found an internal inconsistency with
3103 Compatibility tests for C<sub : attrs> vs the older C<use attrs>.
3107 Tests for new environment scalar capability (e.g., C<use Env qw($BAR);>).
3111 Tests for new environment array capability (e.g., C<use Env qw(@PATH);>).
3115 IO constants (SEEK_*, _IO*).
3119 Directory-related IO methods (new, read, close, rewind, tied delete).
3121 =item lib/io_multihomed
3123 INET sockets with multi-homed hosts.
3135 Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>.
3139 File test operators.
3143 Verify operations that access pad objects (lexicals and temporaries).
3147 Verify C<exists &sub> operations.
3151 =head1 Incompatible Changes
3153 =head2 Perl Source Incompatibilities
3155 Beware that any new warnings that have been added or old ones
3156 that have been enhanced are B<not> considered incompatible changes.
3158 Since all new warnings must be explicitly requested via the C<-w>
3159 switch or the C<warnings> pragma, it is ultimately the programmer's
3160 responsibility to ensure that warnings are enabled judiciously.
3164 =item CHECK is a new keyword
3166 All subroutine definitions named CHECK are now special. See
3167 C</"Support for CHECK blocks"> for more information.
3169 =item Treatment of list slices of undef has changed
3171 There is a potential incompatibility in the behavior of list slices
3172 that are comprised entirely of undefined values.
3173 See L</"Behavior of list slices is more consistent">.
3175 =item Format of $English::PERL_VERSION is different
3177 The English module now sets $PERL_VERSION to $^V (a string value) rather
3178 than C<$]> (a numeric value). This is a potential incompatibility.
3179 Send us a report via perlbug if you are affected by this.
3181 See L</"Improved Perl version numbering system"> for the reasons for
3184 =item Literals of the form C<1.2.3> parse differently
3186 Previously, numeric literals with more than one dot in them were
3187 interpreted as a floating point number concatenated with one or more
3188 numbers. Such "numbers" are now parsed as strings composed of the
3191 For example, C<print 97.98.99> used to output C<97.9899> in earlier
3192 versions, but now prints C<abc>.
3194 See L</"Support for strings represented as a vector of ordinals">.
3196 =item Possibly changed pseudo-random number generator
3198 Perl programs that depend on reproducing a specific set of pseudo-random
3199 numbers may now produce different output due to improvements made to the
3200 rand() builtin. You can use C<sh Configure -Drandfunc=rand> to obtain
3203 See L</"Better pseudo-random number generator">.
3205 =item Hashing function for hash keys has changed
3207 Even though Perl hashes are not order preserving, the apparently
3208 random order encountered when iterating on the contents of a hash
3209 is actually determined by the hashing algorithm used. Improvements
3210 in the algorithm may yield a random order that is B<different> from
3211 that of previous versions, especially when iterating on hashes.
3213 See L</"Better worst-case behavior of hashes"> for additional
3216 =item C<undef> fails on read only values
3218 Using the C<undef> operator on a readonly value (such as $1) has
3219 the same effect as assigning C<undef> to the readonly value--it
3220 throws an exception.
3222 =item Close-on-exec bit may be set on pipe and socket handles
3224 Pipe and socket handles are also now subject to the close-on-exec
3225 behavior determined by the special variable $^F.
3227 See L</"More consistent close-on-exec behavior">.
3229 =item Writing C<"$$1"> to mean C<"${$}1"> is unsupported
3231 Perl 5.004 deprecated the interpretation of C<$$1> and
3232 similar within interpolated strings to mean C<$$ . "1">,
3233 but still allowed it.
3235 In Perl 5.6.0 and later, C<"$$1"> always means C<"${$1}">.
3237 =item delete(), each(), values() and C<\(%h)>
3239 operate on aliases to values, not copies
3241 delete(), each(), values() and hashes (e.g. C<\(%h)>)
3242 in a list context return the actual
3243 values in the hash, instead of copies (as they used to in earlier
3244 versions). Typical idioms for using these constructs copy the
3245 returned values, but this can make a significant difference when
3246 creating references to the returned values. Keys in the hash are still
3247 returned as copies when iterating on a hash.
3249 See also L</"delete(), each(), values() and hash iteration are faster">.
3251 =item vec(EXPR,OFFSET,BITS) enforces powers-of-two BITS
3253 vec() generates a run-time error if the BITS argument is not
3254 a valid power-of-two integer.
3256 =item Text of some diagnostic output has changed
3258 Most references to internal Perl operations in diagnostics
3259 have been changed to be more descriptive. This may be an
3260 issue for programs that may incorrectly rely on the exact
3261 text of diagnostics for proper functioning.
3263 =item C<%@> has been removed
3265 The undocumented special variable C<%@> that used to accumulate
3266 "background" errors (such as those that happen in DESTROY())
3267 has been removed, because it could potentially result in memory
3270 =item Parenthesized not() behaves like a list operator
3272 The C<not> operator now falls under the "if it looks like a function,
3273 it behaves like a function" rule.
3275 As a result, the parenthesized form can be used with C<grep> and C<map>.
3276 The following construct used to be a syntax error before, but it works
3279 grep not($_), @things;
3281 On the other hand, using C<not> with a literal list slice may not
3282 work. The following previously allowed construct:
3284 print not (1,2,3)[0];
3286 needs to be written with additional parentheses now:
3288 print not((1,2,3)[0]);
3290 The behavior remains unaffected when C<not> is not followed by parentheses.
3292 =item Semantics of bareword prototype C<(*)> have changed
3294 The semantics of the bareword prototype C<*> have changed. Perl 5.005
3295 always coerced simple scalar arguments to a typeglob, which wasn't useful
3296 in situations where the subroutine must distinguish between a simple
3297 scalar and a typeglob. The new behavior is to not coerce bareword
3298 arguments to a typeglob. The value will always be visible as either
3299 a simple scalar or as a reference to a typeglob.
3301 See L</"More functional bareword prototype (*)">.
3303 =item Semantics of bit operators may have changed on 64-bit platforms
3305 If your platform is either natively 64-bit or if Perl has been
3306 configured to used 64-bit integers, i.e., $Config{ivsize} is 8,
3307 there may be a potential incompatibility in the behavior of bitwise
3308 numeric operators (& | ^ ~ << >>). These operators used to strictly
3309 operate on the lower 32 bits of integers in previous versions, but now
3310 operate over the entire native integral width. In particular, note
3311 that unary C<~> will produce different results on platforms that have
3312 different $Config{ivsize}. For portability, be sure to mask off
3313 the excess bits in the result of unary C<~>, e.g., C<~$x & 0xffffffff>.
3315 See L</"Bit operators support full native integer width">.
3317 =item More builtins taint their results
3319 As described in L</"Improved security features">, there may be more
3320 sources of taint in a Perl program.
3322 To avoid these new tainting behaviors, you can build Perl with the
3323 Configure option C<-Accflags=-DINCOMPLETE_TAINTS>. Beware that the
3324 ensuing perl binary may be insecure.
3328 =head2 C Source Incompatibilities
3332 =item C<PERL_POLLUTE>
3334 Release 5.005 grandfathered old global symbol names by providing preprocessor
3335 macros for extension source compatibility. As of release 5.6.0, these
3336 preprocessor definitions are not available by default. You need to explicitly
3337 compile perl with C<-DPERL_POLLUTE> to get these definitions. For
3338 extensions still using the old symbols, this option can be
3339 specified via MakeMaker:
3341 perl Makefile.PL POLLUTE=1
3343 =item C<PERL_IMPLICIT_CONTEXT>
3345 This new build option provides a set of macros for all API functions
3346 such that an implicit interpreter/thread context argument is passed to
3347 every API function. As a result of this, something like C<sv_setsv(foo,bar)>
3348 amounts to a macro invocation that actually translates to something like
3349 C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
3350 to not have any significant source compatibility issues, the difference
3351 between a macro and a real function call will need to be considered.
3353 This means that there B<is> a source compatibility issue as a result of
3354 this if your extensions attempt to use pointers to any of the Perl API
3357 Note that the above issue is not relevant to the default build of
3358 Perl, whose interfaces continue to match those of prior versions
3359 (but subject to the other options described here).
3361 See L<perlguts/"The Perl API"> for detailed information on the
3362 ramifications of building Perl with this option.
3364 NOTE: PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built
3365 with one of -Dusethreads, -Dusemultiplicity, or both. It is not
3366 intended to be enabled by users at this time.
3368 =item C<PERL_POLLUTE_MALLOC>
3370 Enabling Perl's malloc in release 5.005 and earlier caused the namespace of
3371 the system's malloc family of functions to be usurped by the Perl versions,
3372 since by default they used the same names. Besides causing problems on
3373 platforms that do not allow these functions to be cleanly replaced, this
3374 also meant that the system versions could not be called in programs that
3375 used Perl's malloc. Previous versions of Perl have allowed this behaviour
3376 to be suppressed with the HIDEMYMALLOC and EMBEDMYMALLOC preprocessor
3379 As of release 5.6.0, Perl's malloc family of functions have default names
3380 distinct from the system versions. You need to explicitly compile perl with
3381 C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
3382 and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
3385 Note that these functions do B<not> constitute Perl's memory allocation API.
3386 See L<perlguts/"Memory Allocation"> for further information about that.
3390 =head2 Compatible C Source API Changes
3394 =item C<PATCHLEVEL> is now C<PERL_VERSION>
3396 The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
3397 are now available by default from perl.h, and reflect the base revision,
3398 patchlevel, and subversion respectively. C<PERL_REVISION> had no
3399 prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
3400 previously available as C<PATCHLEVEL> and C<SUBVERSION>.
3402 The new names cause less pollution of the B<cpp> namespace and reflect what
3403 the numbers have come to stand for in common practice. For compatibility,
3404 the old names are still supported when F<patchlevel.h> is explicitly
3405 included (as required before), so there is no source incompatibility
3410 =head2 Binary Incompatibilities
3412 In general, the default build of this release is expected to be binary
3413 compatible for extensions built with the 5.005 release or its maintenance
3414 versions. However, specific platforms may have broken binary compatibility
3415 due to changes in the defaults used in hints files. Therefore, please be
3416 sure to always check the platform-specific README files for any notes to
3419 The usethreads or usemultiplicity builds are B<not> binary compatible
3420 with the corresponding builds in 5.005.
3422 On platforms that require an explicit list of exports (AIX, OS/2 and Windows,
3423 among others), purely internal symbols such as parser functions and the
3424 run time opcodes are not exported by default. Perl 5.005 used to export
3425 all functions irrespective of whether they were considered part of the
3428 For the full list of public API functions, see L<perlapi>.
3430 =head1 Known Problems
3432 =head2 Localizing a tied hash element may leak memory
3434 As of the 5.6.1 release, there is a known leak when code such as this
3438 tie my %tie_hash => 'Tie::StdHash';
3442 local($tie_hash{Foo}) = 1; # leaks
3444 =head2 Known test failures
3452 Subtest #15 of lib/b.t may fail under 64-bit builds on platforms such
3453 as HP-UX PA64 and Linux IA64. The issue is still being investigated.
3455 The lib/io_multihomed test may hang in HP-UX if Perl has been
3456 configured to be 64-bit. Because other 64-bit platforms do not
3457 hang in this test, HP-UX is suspect. All other tests pass
3458 in 64-bit HP-UX. The test attempts to create and connect to
3459 "multihomed" sockets (sockets which have multiple IP addresses).
3461 Note that 64-bit support is still experimental.
3465 Failure of Thread tests
3467 The subtests 19 and 20 of lib/thr5005.t test are known to fail due to
3468 fundamental problems in the 5.005 threading implementation. These are
3469 not new failures--Perl 5.005_0x has the same bugs, but didn't have these
3470 tests. (Note that support for 5.005-style threading remains experimental.)
3474 NEXTSTEP 3.3 POSIX test failure
3476 In NEXTSTEP 3.3p2 the implementation of the strftime(3) in the
3477 operating system libraries is buggy: the %j format numbers the days of
3478 a month starting from zero, which, while being logical to programmers,
3479 will cause the subtests 19 to 27 of the lib/posix test may fail.
3483 Tru64 (aka Digital UNIX, aka DEC OSF/1) lib/sdbm test failure with gcc
3485 If compiled with gcc 2.95 the lib/sdbm test will fail (dump core).
3486 The cure is to use the vendor cc, it comes with the operating system
3487 and produces good code.
3491 =head2 EBCDIC platforms not fully supported
3493 In earlier releases of Perl, EBCDIC environments like OS390 (also
3494 known as Open Edition MVS) and VM-ESA were supported. Due to changes
3495 required by the UTF-8 (Unicode) support, the EBCDIC platforms are not
3496 supported in Perl 5.6.0.
3498 The 5.6.1 release improves support for EBCDIC platforms, but they
3499 are not fully supported yet.
3501 =head2 UNICOS/mk CC failures during Configure run
3503 In UNICOS/mk the following errors may appear during the Configure run:
3505 Guessing which symbols your C compiler and preprocessor define...
3506 CC-20 cc: ERROR File = try.c, Line = 3
3508 bad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K
3510 4 errors detected in the compilation of "try.c".
3512 The culprit is the broken awk of UNICOS/mk. The effect is fortunately
3513 rather mild: Perl itself is not adversely affected by the error, only
3514 the h2ph utility coming with Perl, and that is rather rarely needed
3517 =head2 Arrow operator and arrays
3519 When the left argument to the arrow operator C<< -> >> is an array, or
3520 the C<scalar> operator operating on an array, the result of the
3521 operation must be considered erroneous. For example:
3526 These expressions will get run-time errors in some future release of
3529 =head2 Experimental features
3531 As discussed above, many features are still experimental. Interfaces and
3532 implementation of these features are subject to change, and in extreme cases,
3533 even subject to removal in some future release of Perl. These features
3534 include the following:
3542 =item 64-bit support
3544 =item Lvalue subroutines
3546 =item Weak references
3548 =item The pseudo-hash data type
3550 =item The Compiler suite
3552 =item Internal implementation of file globbing
3556 =item The regular expression code constructs:
3558 C<(?{ code })> and C<(??{ code })>
3562 =head1 Obsolete Diagnostics
3566 =item Character class syntax [: :] is reserved for future extensions
3568 (W) Within regular expression character classes ([]) the syntax beginning
3569 with "[:" and ending with ":]" is reserved for future extensions.
3570 If you need to represent those character sequences inside a regular
3571 expression character class, just quote the square brackets with the
3572 backslash: "\[:" and ":\]".
3574 =item Ill-formed logical name |%s| in prime_env_iter
3576 (W) A warning peculiar to VMS. A logical name was encountered when preparing
3577 to iterate over %ENV which violates the syntactic rules governing logical
3578 names. Because it cannot be translated normally, it is skipped, and will not
3579 appear in %ENV. This may be a benign occurrence, as some software packages
3580 might directly modify logical name tables and introduce nonstandard names,
3581 or it may indicate that a logical name table has been corrupted.
3583 =item In string, @%s now must be written as \@%s
3585 The description of this error used to say:
3587 (Someday it will simply assume that an unbackslashed @
3588 interpolates an array.)
3590 That day has come, and this fatal error has been removed. It has been
3591 replaced by a non-fatal warning instead.
3592 See L</Arrays now always interpolate into double-quoted strings> for
3595 =item Probable precedence problem on %s
3597 (W) The compiler found a bareword where it expected a conditional,
3598 which often indicates that an || or && was parsed as part of the
3599 last argument of the previous construct, for example:
3603 =item regexp too big
3605 (F) The current implementation of regular expressions uses shorts as
3606 address offsets within a string. Unfortunately this means that if
3607 the regular expression compiles to longer than 32767, it'll blow up.
3608 Usually when you want a regular expression this big, there is a better
3609 way to do it with multiple statements. See L<perlre>.
3611 =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
3613 (D) Perl versions before 5.004 misinterpreted any type marker followed
3614 by "$" and a digit. For example, "$$0" was incorrectly taken to mean
3615 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
3617 However, the developers of Perl 5.004 could not fix this bug completely,
3618 because at least two widely-used modules depend on the old meaning of
3619 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
3620 old (broken) way inside strings; but it generates this message as a
3621 warning. And in Perl 5.005, this special treatment will cease.
3625 =head1 Reporting Bugs
3627 If you find what you think is a bug, you might check the
3628 articles recently posted to the comp.lang.perl.misc newsgroup.
3629 There may also be information at http://www.perl.com/ , the Perl
3632 If you believe you have an unreported bug, please run the B<perlbug>
3633 program included with your release. Be sure to trim your bug down
3634 to a tiny but sufficient test case. Your bug report, along with the
3635 output of C<perl -V>, will be sent off to perlbug@perl.org to be
3636 analysed by the Perl porting team.
3640 The F<Changes> file for exhaustive details on what changed.
3642 The F<INSTALL> file for how to build Perl.
3644 The F<README> file for general stuff.
3646 The F<Artistic> and F<Copying> files for copyright information.
3650 Written by Gurusamy Sarathy <F<gsar@ActiveState.com>>, with many
3651 contributions from The Perl Porters.
3653 Send omissions or corrections to <F<perlbug@perl.org>>.