3 perldelta - what's new for perl v5.6 (as of v5.005_62)
7 This is an unsupported alpha release, meant for intrepid Perl developers
8 only. The included sources may not even build correctly on some platforms.
9 Subscribing to perl5-porters is the best way to monitor and contribute
10 to the progress of development releases (see www.perl.org for info).
12 This document describes differences between the 5.005 release and this one.
14 =head1 Incompatible Changes
16 =head2 Perl Source Incompatibilities
18 Beware that any new warnings that have been added or enhanced old
19 warnings are B<not> considered incompatible changes.
21 Since all new warnings must be explicitly requested via the C<-w>
22 switch or the C<warnings> pragma, it is ultimately the programmer's
23 responsibility to ensure that warnings are enabled judiciously.
27 =item STOP is a new keyword
29 In addition to C<BEGIN>, C<INIT> and C<END>, subroutines named
30 C<STOP> are now special. They are queued up for execution at the
31 end of compilation, and cannot be called directly.
33 =item Treatment of list slices of undef has changed
35 When taking a slice of a literal list (as opposed to a slice of
36 an array or hash), Perl used to return an empty list if the
37 result happened to be composed of all undef values.
39 The new behavior is to produce an empty list if (and only if)
40 the original list was empty. Consider the following example:
42 @a = (1,undef,undef,2)[2,1,2];
44 The old behavior would have resulted in @a having no elements.
45 The new behavior ensures it has three undefined elements.
47 Note in particular that the behavior of slices of the following
48 cases remains unchanged:
52 @a = (anything_returning_empty_list())[2,1,2];
58 =item Possibly changed pseudo-random number generator
60 In 5.005_0x and earlier, perl's rand() function used the C library
61 rand(3) function. As of 5.005_52, Configure tests for drand48(),
62 random(), and rand() (in that order) and picks the first one it finds.
63 Perl programs that depend on reproducing a specific set of pseudo-random
64 numbers will now likely produce different output. You can use
65 C<sh Configure -Drandfunc=rand> to obtain the old behavior.
67 =item Hashing function for hash keys has changed
69 Perl hashes are not order preserving. The apparently random order
70 encountered when iterating on the contents of a hash is determined
71 by the hashing algorithm used. To improve the distribution of lower
72 bits in the hashed value, the algorithm has changed slightly as of
73 5.005_52. When iterating over hashes, this may yield a random order
74 that is B<different> from that of previous versions.
76 =item C<undef> fails on read only values
78 Using the C<undef> operator on a readonly value (such as $1) has
79 the same effect as assigning C<undef> to the readonly value--it
82 =item Close-on-exec bit may be set on pipe() handles
84 On systems that support a close-on-exec flag on filehandles, the
85 flag will be set for any handles created by pipe(), if that is
86 warranted by the value of $^F that may be in effect. Earlier
87 versions neglected to set the flag for handles created with
88 pipe(). See L<perlfunc/pipe> and L<perlvar/$^F>.
90 =item Writing C<"$$1"> to mean C<"${$}1"> is unsupported
92 Perl 5.004 deprecated the interpretation of C<$$1> and
93 similar within interpolated strings to mean C<$$ . "1">,
96 In Perl 5.6 and later, C<"$$1"> always means C<"${$1}">.
98 =item values(%h) and C<\(%h)> operate on aliases to values, not copies
100 each(), values() and hashes in a list context return the actual
101 values in the hash, instead of copies (as they used to in earlier
102 versions). Typical idioms for using these constructs copy the
103 returned values, but this can make a significant difference when
104 creating references to the returned values.
106 Keys in the hash are still returned as copies when iterating on
109 =item vec(EXPR,OFFSET,BITS) enforces powers-of-two BITS
111 vec() generates a run-time error if the BITS argument is not
112 a valid power-of-two integer.
114 =item Text of some diagnostic output has changed
116 Most references to internal Perl operations in diagnostics
117 have been changed to be more descriptive. This may be an
118 issue for programs that may incorrectly rely on the exact
119 text of diagnostics for proper functioning.
121 =item C<%@> has been removed
123 The undocumented special variable C<%@> that used to accumulate
124 "background" errors (such as those that happen in DESTROY())
125 has been removed, because it could potentially result in memory
130 =head2 C Source Incompatibilities
134 =item C<PERL_POLLUTE>
136 Release 5.005 grandfathered old global symbol names by providing preprocessor
137 macros for extension source compatibility. As of release 5.6, these
138 preprocessor definitions are not available by default. You need to explicitly
139 compile perl with C<-DPERL_POLLUTE> to get these definitions. For
140 extensions still using the old symbols, this option can be
141 specified via MakeMaker:
143 perl Makefile.PL POLLUTE=1
145 =item C<PERL_IMPLICIT_CONTEXT>
147 This new build option provides a set of macros for all API functions
148 such that an implicit interpreter/thread context argument is passed to
149 every API function. As a result of this, something like C<sv_setsv(foo,bar)>
150 amounts to a macro invocation that actually translates to something like
151 C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
152 to not have any significant source compatibility issues, the difference
153 between a macro and a real function call will need to be considered.
155 This means that there B<is> a source compatibility issue as a result of
156 this if your extensions attempt to use pointers to any of the Perl API
159 Note that the above issue is not relevant to the default build of
160 Perl, whose interfaces continue to match those of prior versions
161 (but subject to the other options described here).
163 PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built
164 with one of -Dusethreads, -Dusemultiplicity, or both.
166 See L<perlguts/"The Perl API"> for detailed information on the
167 ramifications of building Perl using this option.
169 =item C<PERL_POLLUTE_MALLOC>
171 Enabling Perl's malloc in release 5.005 and earlier caused
172 the namespace of system versions of the malloc family of functions to
173 be usurped by the Perl versions, since by default they used the
176 Besides causing problems on platforms that do not allow these functions to
177 be cleanly replaced, this also meant that the system versions could not
178 be called in programs that used Perl's malloc. Previous versions of Perl
179 have allowed this behaviour to be suppressed with the HIDEMYMALLOC and
180 EMBEDMYMALLOC preprocessor definitions.
182 As of release 5.6, Perl's malloc family of functions have default names
183 distinct from the system versions. You need to explicitly compile perl with
184 C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
185 and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
188 Note that these functions do B<not> constitute Perl's memory allocation API.
189 See L<perlguts/"Memory Allocation"> for further information about that.
193 =head2 Compatible C Source API Changes
197 =item C<PATCHLEVEL> is now C<PERL_VERSION>
199 The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
200 are now available by default from perl.h, and reflect the base revision,
201 patchlevel, and subversion respectively. C<PERL_REVISION> had no
202 prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
203 previously available as C<PATCHLEVEL> and C<SUBVERSION>.
205 The new names cause less pollution of the B<cpp> namespace and reflect what
206 the numbers have come to stand for in common practice. For compatibility,
207 the old names are still supported when F<patchlevel.h> is explicitly
208 included (as required before), so there is no source incompatibility
211 =item Support for C++ exceptions
213 change#3386, also needs perlguts documentation
214 [TODO - Chip Salzenberg <chip@perlsupport.com>]
218 =head2 Binary Incompatibilities
220 The default build of this release is binary compatible with the 5.005
221 release or its maintenance versions.
223 The usethreads or usemultiplicity builds are B<not> binary compatible
224 with the corresponding builds in 5.005.
226 =head1 Installation and Configuration Improvements
228 =head2 New Configure flags
230 The following new flags may be enabled on the Configure command line
231 by running Configure with C<-Dflag>.
239 =head2 -Dusethreads and -Duse64bits now more daring
241 The Configure options enabling the use of threads and the use of
242 64-bitness are now more daring in the sense that they no more have
243 an explicit list of operating systems of known threads/64-bit
244 capabilities. In other words: if your operating system has the
245 necessary APIs, you should be able just to go ahead and use them.
246 See also L<"64-bit support">.
250 Some platforms have "long doubles", floating point numbers of even
251 larger range than ordinary "doubles". To enable using long doubles for
252 Perl's scalars, use -Duselongdouble.
256 You can enable both -Duse64bits and -Dlongdouble by -Dusemorebits.
257 See also L<"64-bit support">.
259 =head2 -Duselargefiles
261 Some platforms support large files, files larger than two gigabytes.
262 See L<"Large file support"> for more information.
264 =head2 installusrbinperl
266 You can use "Configure -Uinstallusrbinperl" which causes installperl
267 to skip installing perl also as /usr/bin/perl. This is useful if you
268 prefer not to modify /usr/bin for some reason or another but harmful
269 because many scripts assume to find Perl in /usr/bin/perl.
273 You can use "Configure -Dusesocks" which causes Perl to probe
274 for the SOCKS (v5, not v4) proxy protocol library,
275 http://www.socks.nec.com/
279 You can "post-edit" the Configure variables using the Configure C<-A>
280 flag. The editing happens immediately after the platform specific
281 hints files have been processed but before the actual configuration
282 process starts. Run C<Configure -h> to find out the full C<-A> syntax.
284 =head2 Enhanced Installation Directories
286 The installation structure has been enriched to improve the support for
287 maintaining multiple versions of perl, to provide locations for
288 vendor-supplied modules and scripts, and to ease maintenance of
289 locally-added modules and scripts. See the section on Installation
290 Directories in the INSTALL file for complete details. For most users
291 building and installing from source, the defaults should be fine.
295 =head2 Unicode and UTF-8 support
297 Perl can optionally use UTF-8 as its internal representation for character
298 strings. The C<utf8> pragma enables this support in the current lexical
299 scope. See L<utf8> for more information.
301 =head2 Lexically scoped warning categories
303 You can now control the granularity of warnings emitted by perl at a finer
304 level using the C<use warnings> pragma. See L<warnings> and L<perllexwarn>
307 =head2 Lvalue subroutines
309 WARNING: This is an experimental feature.
312 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>,
313 Tuomas Lukka <lukka@fas.harvard.edu>)]
315 =head2 "our" declarations
317 An "our" declaration introduces a value that can be best understood
318 as a lexically scoped symbolic alias to a global variable in the
319 current package. This is mostly useful as an alternative to the
320 C<vars> pragma, but also provides the opportunity to introduce
321 typing and other attributes for such variables. See L<perlfunc/our>.
323 =head2 Weak references
325 WARNING: This is an experimental feature.
327 change#3385, also need perlguts documentation
329 [TODO - Tuomas Lukka <lukka@fas.harvard.edu>]
331 =head2 File globbing implemented internally
333 WARNING: This is currently an experimental feature. Interfaces and
334 implementation are likely to change.
336 Perl now uses the File::Glob implementation of the glob() operator
337 automatically. This avoids using an external csh process and the
338 problems associated with it.
340 =head2 Binary numbers supported
342 Binary numbers are now supported as literals, in s?printf formats, and
346 printf "The answer is: %b\n", oct("0b101010");
348 =head2 Some arrows may be omitted in calls through references
350 Perl now allows the arrow to be omitted in many constructs
351 involving subroutine calls through references. For example,
352 C<$foo[10]->('foo')> may now be written C<$foo[10]('foo')>.
353 This is rather similar to how the arrow may be omitted from
354 C<$foo[10]->{'foo'}>. Note however, that the arrow is still
355 required for C<foo(10)->('bar')>.
357 =head2 syswrite() ease-of-use
359 The length argument of C<syswrite()> has become optional.
361 =head2 Filehandles can be autovivified
363 Similar to how constructs such as C<$x->[0]> autovivify a reference,
364 open() now autovivifies a filehandle if the first argument is an
365 uninitialized variable. This allows the constructs C<open(my $fh, ...)> and
366 C<open(local $fh,...)> to be used to create filehandles that will
367 conveniently be closed automatically when the scope ends, provided there
368 are no other references to them. This largely eliminates the need for
369 typeglobs when opening filehandles that must be passed around, as in the
374 or die "Can't open '@_': $!";
379 my $f = myopen("</etc/motd");
381 # $f implicitly closed here
384 [TODO - this idiom needs more pod penetration]
386 =head2 64-bit support
388 All platforms that have 64-bit integers either (a) natively as longs
389 or ints (b) via special compiler flags (c) using long long are able to
390 use "quads" (64-integers) as follows:
396 constants (decimal, hexadecimal, octal, binary) in the code
400 arguments to oct() and hex()
404 arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)
412 pack() and unpack() "q" and "Q" formats
416 in basic arithmetics: + - * / %
420 vec() (but see the below note about bit arithmetics)
424 Note that unless you have the case (a) you will have to configure
425 and compile Perl using the -Duse64bits Configure flag.
427 Unfortunately bit arithmetics (&, |, ^, ~, <<, >>) for numbers are not
428 64-bit clean, they are explictly forced to be 32-bit. Bit arithmetics
429 for bit vectors (created by vec()) are not limited in their width.
431 Last but not least: note that due to Perl's habit of always using
432 floating point numbers the quads are still not true integers.
433 When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned,
434 -9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they
435 are silently promoted to floating point numbers, after which they will
436 start losing precision (their lower digits).
438 =head2 Large file support
440 If you have filesystems that support "large files" (files larger than
441 2 gigabytes), you may now also be able to create and access them from
442 Perl. You have to use Configure -Duselargefiles. Turning on the
443 large file support turns on also the 64-bit support, for obvious reasons.
445 Note that in addition to requiring a proper file system to do large
446 files you may also need to adjust your per-process (or your
447 per-system, or per-process-group, or per-user-group) maximum filesize
448 limits before running Perl scripts that try to handle large files,
449 especially if you intend to write such files.
451 Finally, in addition to your process/process group maximum filesize
452 limits, you may have quota limits on your filesystems that stop you
453 (your user id or your user group id) from using large files.
455 Adjusting your process/user/group/file system/operating system limits
456 is outside the scope of Perl core language. For process limits, you
457 may try increasing the limits using your shell's limits/limit/ulimit
458 command before running Perl. The BSD::Resource extension (not
459 included with the standard Perl distribution) may also be of use, it
460 offers the getrlimit/setrlimit interface that can be used to adjust
461 process resource usage limits, including the maximum filesize limit.
465 In some systems you may be able to use long doubles to enhance the
466 range of precision of your double precision floating point numbers
467 (that is, Perl's numbers). Use Configure -Duselongdouble to enable
468 this support (if it is available).
472 You can Configure -Dusemorebits to turn on both the 64-bit support
473 and the long double support.
475 =head2 Enhanced support for sort() subroutines
477 Perl subroutines with a prototype of C<($$)> and XSUBs in general can
478 now be used as sort subroutines. In either case, the two elements to
479 be compared as passed as normal parameters in @_. See L<perlfunc/sort>.
481 For unprototyped sort subroutines, the historical behavior of passing
482 the elements to be compared as the global variables $a and $b remains
485 =head2 Better syntax checks on parenthesized unary operators
489 print defined(&foo,&bar,&baz);
490 print uc("foo","bar","baz");
493 used to be accidentally allowed in earlier versions, and produced
494 unpredictable behaviour. Some produced ancillary warnings
495 when used in this way; others silently did the wrong thing.
497 The parenthesized forms of most unary operators that expect a single
498 argument now ensure that they are not called with more than one
499 argument, making the cases shown above syntax errors. The usual
502 print defined &foo, &bar, &baz;
503 print uc "foo", "bar", "baz";
506 remains unchanged. See L<perlop>.
508 =head2 POSIX character class syntax [: :] supported
510 For example to match alphabetic characters use /[[:alpha:]]/.
511 See L<perlre> for details.
513 =head2 Improved C<qw//> operator
515 The C<qw//> operator is now evaluated at compile time into a true list
516 instead of being replaced with a run time call to C<split()>. This
517 removes the confusing misbehaviour of C<qw//> in scalar context, which
518 had inherited that behaviour from split().
522 $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
524 now correctly prints "3|a", instead of "2|a".
526 =head2 pack() format 'Z' supported
528 The new format type 'Z' is useful for packing and unpacking null-terminated
529 strings. See L<perlfunc/"pack">.
531 =head2 pack() format modifier '!' supported
533 The new format type modifier '!' is useful for packing and unpacking
534 native shorts, ints, and longs. See L<perlfunc/"pack">.
536 =head2 pack() and unpack() support counted strings
538 The template character '/' can be used to specify a counted string
539 type to be packed or unpacked. See L<perlfunc/"pack">.
541 =head2 Comments in pack() templates
543 The '#' character in a template introduces a comment up to
544 end of the line. This facilitates documentation of pack()
547 =head2 $^X variables may now have names longer than one character
549 Formerly, $^X was synonymous with ${"\cX"}, but $^XY was a syntax
550 error. Now variable names that begin with a control character may be
551 arbitrarily long. However, for compatibility reasons, these variables
552 I<must> be written with explicit braces, as C<${^XY}> for example.
553 C<${^XYZ}> is synonymous with ${"\cXYZ"}. Variable names with more
554 than one control character, such as C<${^XY^Z}>, are illegal.
556 The old syntax has not changed. As before, `^X' may be either a
557 literal control-X character or the two-character sequence `caret' plus
558 `X'. When braces are omitted, the variable name stops after the
559 control character. Thus C<"$^XYZ"> continues to be synonymous with
560 C<$^X . "YZ"> as before.
562 As before, lexical variables may not have names beginning with control
563 characters. As before, variables whose names begin with a control
564 character are always forced to be in package `main'. All such variables
565 are reserved for future extensions, except those that begin with
566 C<^_>, which may be used by user programs and are guaranteed not to
567 acquire special meaning in any future version of Perl.
569 =head2 C<use attrs> implicit in subroutine attributes
571 Formerly, if you wanted to mark a subroutine as being a method call or
572 as requiring an automatic lock() when it is entered, you had to declare
573 that with a C<use attrs> pragma in the body of the subroutine.
574 That can now be accomplished with a declaration syntax, like this:
576 sub mymethod : locked, method ;
578 sub mymethod : locked, method {
582 F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes
583 with the stubs they provide. See L<attributes>.
585 =head2 Regular expression improvements
587 change#2827,2373,2372,2365,1813,1800,4112,4158,4215,4301
588 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
590 =head2 Overloading improvements
593 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
595 =head2 open() with more than two arguments
597 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
599 =head2 Support for interpolating named characters
602 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
604 =head2 Experimental support for user-hooks in @INC
606 [TODO - Ken Fox <kfox@ford.com>]
608 =head2 C<require> and C<do> may be overridden
610 C<require> and C<do 'file'> operations may be overridden locally
611 by importing subroutines of the same name into the current package
612 (or globally by importing them into the CORE::GLOBAL:: namespace).
613 Overriding C<require> will also affect C<use>, provided the override
614 is visible at compile-time.
615 See L<perlsub/"Overriding Built-in Functions">.
617 =head2 New variable $^C reflects C<-c> switch
619 C<$^C> has a boolean value that reflects whether perl is being run
620 in compile-only mode (i.e. via the C<-c> switch). Since
621 BEGIN blocks are executed under such conditions, this variable
622 enables perl code to determine whether actions that make sense
623 only during normal running are warranted. See L<perlvar>.
627 Arbitrary code can be queued for execution when Perl has finished
628 parsing the program (i.e. when the compile phase ends) using STOP
629 blocks. These behave similar to END blocks, except for being
630 called at the end of compilation rather than at the end of execution.
632 =head2 Optional Y2K warnings
634 If Perl is built with the cpp macro C<PERL_Y2KWARN> defined,
635 it emits optional warnings when concatenating the number 19
638 This behavior must be specifically enabled when running Configure.
639 See L<INSTALL> and L<README.Y2K>.
641 =head1 Significant bug fixes
643 =head2 E<lt>HANDLEE<gt> on empty files
645 With C<$/> set to C<undef>, slurping an empty file returns a string of
646 zero length (instead of C<undef>, as it used to) the first time the
647 HANDLE is read. Further reads yield C<undef>.
649 This means that the following will append "foo" to an empty file (it used
652 perl -0777 -pi -e 's/^/foo/' empty_file
656 perl -pi -e 's/^/foo/' empty_file
658 is unchanged (it continues to leave the file empty).
660 =head2 C<eval '...'> improvements
662 Line numbers (as reflected by caller() and most diagnostics) within
663 C<eval '...'> were often incorrect when here documents were involved.
664 This has been corrected.
666 Lexical lookups for variables appearing in C<eval '...'> within
667 functions that were themselves called within an C<eval '...'> were
668 searching the wrong place for lexicals. The lexical search now
669 correctly ends at the subroutine's block boundary.
671 Parsing of here documents used to be flawed when they appeared as
672 the replacement expression in C<eval 's/.../.../e'>. This has
675 =head2 All compilation errors are true errors
677 Some "errors" encountered at compile time were by neccessity
678 generated as warnings followed by eventual termination of the
679 program. This enabled more such errors to be reported in a
680 single run, rather than causing a hard stop at the first error
681 that was encountered.
683 The mechanism for reporting such errors has been reimplemented
684 to queue compile-time errors and report them at the end of the
685 compilation as true errors rather than as warnings. This fixes
686 cases where error messages leaked through in the form of warnings
687 when code was compiled at run time using C<eval STRING>, and
688 also allows such errors to be reliably trapped using __DIE__ hooks.
690 =head2 Automatic flushing of output buffers
692 fork(), exec(), system(), qx//, and pipe open()s now flush buffers
693 of all files opened for output when the operation
694 was attempted. This mostly eliminates confusing
695 buffering mishaps suffered by users unaware of how Perl internally
698 =head2 Better diagnostics on meaningless filehandle operations
700 Constructs such as C<open(E<lt>FHE<gt>)> and C<close(E<lt>FHE<gt>)>
701 are compile time errors. Attempting to read from filehandles that
702 were opened only for writing will now produce warnings (just as
703 writing to read-only filehandles does).
705 =head2 Where possible, buffered data discarded from duped input filehandle
707 C<open(NEW, "E<lt>&OLD")> now attempts to discard any data that
708 was previously read and buffered in C<OLD> before duping the handle.
709 On platforms where doing this is allowed, the next read operation
710 on C<NEW> will return the same data as the corresponding operation
711 on C<OLD>. Formerly, it would have returned the data from the start
712 of the following disk block instead.
714 =head2 system(), backticks and pipe open now reflect exec() failure
716 On Unix and similar platforms, system(), qx() and open(FOO, "cmd |")
717 etc., are implemented via fork() and exec(). When the underlying
718 exec() fails, earlier versions did not report the error properly,
719 since the exec() happened to be in a different process.
721 The child process now communicates with the parent about the
722 error in launching the external command, which allows these
723 constructs to return with their usual error value and set $!.
725 =head2 Implicitly closed filehandles are safer
727 Sometimes implicitly closed filehandles (as when they are localized,
728 and Perl automatically closes them on exiting the scope) could
729 inadvertently set $? or $!. This has been corrected.
731 =head2 C<(\$)> prototype and C<$foo{a}>
733 An scalar reference prototype now correctly allows a hash or
734 array element in that slot.
736 =head2 Pseudo-hashes work better
738 Dereferencing some types of reference values in a pseudo-hash,
739 such as C<$ph->{foo}[1]>, was accidentally disallowed. This has
742 When applied to a pseudo-hash element, exists() now reports whether
743 the specified value exists, not merely if the key is valid.
745 =head2 C<goto &sub> and AUTOLOAD
747 The C<goto &sub> construct works correctly when C<&sub> happens
750 =head2 C<-bareword> allowed under C<use integer>
752 The autoquoting of barewords preceded by C<-> did not work
753 in prior versions when the C<integer> pragma was enabled.
756 =head2 Boolean assignment operators are legal lvalues
758 Constructs such as C<($a ||= 2) += 1> are now allowed.
760 =head2 C<sort $coderef @foo> allowed
762 sort() did not accept a subroutine reference as the comparison
763 function in earlier versions. This is now permitted.
765 =head2 Failures in DESTROY()
767 When code in a destructor threw an exception, it went unnoticed
768 in earlier versions of Perl, unless someone happened to be
769 looking in $@ just after the point the destructor happened to
770 run. Such failures are now visible as warnings when warnings are
773 =head2 Locale bugs fixed
775 printf() and sprintf() previously reset the numeric locale
776 back to the default "C" locale. This has been fixed.
778 Numbers formatted according to the local numeric locale
779 (such as using a decimal comma instead of a decimal dot) caused
780 "isn't numeric" warnings, even while the operations accessing
781 those numbers produced correct results. The warnings are gone.
785 The C<eval 'return sub {...}'> construct could sometimes leak
786 memory. This has been fixed.
788 Operations that aren't filehandle constructors used to leak memory
789 when used on invalid filehandles. This has been fixed.
791 Constructs that modified C<@_> could fail to deallocate values
792 in C<@_> and thus leak memory. This has been corrected.
794 =head2 Spurious subroutine stubs after failed subroutine calls
796 Perl could sometimes create empty subroutine stubs when a
797 subroutine was not found in the package. Such cases stopped
798 later method lookups from progressing into base packages.
799 This has been corrected.
801 =head2 Consistent numeric conversions
804 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
806 =head2 Taint failures under C<-U>
808 When running in unsafe mode, taint violations could sometimes
809 cause silent failures. This has been fixed.
811 =head2 END blocks and the C<-c> switch
813 Prior versions used to run BEGIN B<and> END blocks when Perl was
814 run in compile-only mode. Since this is typically not the expected
815 behavior, END blocks are not executed anymore when the C<-c> switch
818 See L<STOP blocks> for how to run things when the compile phase ends.
820 =head2 Potential to leak DATA filehandles
822 Using the C<__DATA__> token creates an implicit filehandle to
823 the file that contains the token. It is the program's
824 responsibility to close it when it is done reading from it.
826 This caveat is now better explained in the documentation.
829 =head2 Diagnostics follow STDERR
831 Diagnostic output now goes to whichever file the C<STDERR> handle
832 is pointing at, instead of always going to the underlying C runtime
835 =head2 Other fixes for better diagnostics
837 Line numbers are no longer suppressed (under most likely circumstances)
838 during the global destruction phase.
840 Diagnostics emitted from code running in threads other than the main
841 thread are now accompanied by the thread ID.
843 Embedded null characters in diagnostics now actually show up. They
844 used to truncate the message in prior versions.
846 $foo::a and $foo::b are now exempt from "possible typo" warnings only
847 if sort() is encountered in package foo.
849 Unrecognized alphabetic escapes encountered when parsing quote
850 constructs now generate a warning, since they may take on new
851 semantics in later versions of Perl.
853 =head1 Performance enhancements
855 =head2 Simple sort() using { $a <=> $b } and the like are optimized
857 Many common sort() operations using a simple inlined block are now
858 optimized for faster performance.
860 =head2 Optimized assignments to lexical variables
862 Certain operations in the RHS of assignment statements have been
863 optimized to directly set the lexical variable on the LHS,
864 eliminating redundant copying overheads.
866 =head2 Method lookups optimized
868 [TODO - Chip Salzenberg <chip@perlsupport.com>]
870 =head2 Faster mechanism to invoke XSUBs
873 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
875 =head2 Perl_malloc() improvements
878 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
880 =head2 Faster subroutine calls
882 Minor changes in how subroutine calls are handled internally
883 provide marginal improvements in performance.
885 =head1 Platform specific changes
887 =head2 Additional supported platforms
893 VM/ESA is now supported.
897 Siemens BS2000 is now supported under the POSIX Shell.
901 The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
906 GNU/Hurd is now supported.
910 Rhapsody is now supported.
914 EPOC is is now supported (on Psion 5).
924 Perl now works with djgpp 2.02 (and 2.03 alpha).
928 Environment variable names are not converted to uppercase any more.
932 Wrong exit code from backticks now fixed.
936 This port is still using its own builtin globbing.
942 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
946 [TODO - Charles Bailey <bailey@newman.upenn.edu>]
950 Site library searches failed to look for ".../site/5.XXX/lib"
951 if ".../site/5.XXXYY/lib" wasn't found. This has been corrected.
953 When given a pathname that consists only of a drivename, such
954 as C<A:>, opendir() and stat() now use the current working
955 directory for the drive rather than the drive root.
957 The builtin XSUB functions in the Win32:: namespace are
958 documented. See L<Win32>.
960 $^X now contains the full path name of the running executable.
962 A Win32::GetLongPathName() function is provided to complement
963 Win32::GetFullPathName() and Win32::GetShortPathName(). See L<Win32>.
965 POSIX::uname() is supported.
967 system(1,...) now returns true process IDs rather than process
968 handles. kill() accepts any real process id, rather than strictly
969 return values from system(1,...).
971 The C<Shell> module is supported.
973 Rudimentary support for building under command.com in Windows 95
984 Compatibility tests for C<sub : attrs> vs the older C<use attrs>.
988 IO constants (SEEK_*, _IO*).
992 Directory-related IO methods (new, read, close, rewind, tied delete).
994 =item lib/io_multihomed
996 INET sockets with multi-homed hosts.
1008 Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>.
1012 File test operators.
1016 Verify operations that access pad objects (lexicals and temporaries).
1020 =head1 Modules and Pragmata
1028 While used internally by Perl as a pragma, this module also
1029 provides a way to fetch subroutine and variable attributes.
1034 The Perl Compiler suite has been extensively reworked for this
1037 [TODO - Vishal Bhatia <vishal@gol.com>,
1038 Nick Ing-Simmons <nick@ni-s.u-net.com>]
1042 The ByteLoader is a dedicated extension to generate and run
1043 Perl bytecode. See L<ByteLoader>.
1047 References can now be used. See L<constant>.
1052 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1056 A C<Maxdepth> setting can be specified to avoid venturing
1057 too deeply into deep data structures. See L<Data::Dumper>.
1059 Dumping C<qr//> objects works correctly.
1063 C<DB> is an experimental module that exposes a clean abstraction
1064 to Perl's debugging API.
1068 [TODO - Paul Marquess <paul.marquess@bt.com>]
1072 Devel::DProf, a Perl source code profiler has been added. See
1073 L<Devel::DProf> and L<dprofpp>.
1077 The Dumpvalue module provides screen dumps of Perl data.
1081 Overall, Benchmark results exhibit lower average error and better timing
1084 You can now run tests for I<n> seconds instead of guessing the right
1085 number of tests to run: e.g. timethese(-5, ...) will run each
1086 code for at least 5 CPU seconds. Zero as the "number of repetitions"
1087 means "for at least 3 CPU seconds". The output format has also
1088 changed. For example:
1090 use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
1092 will now output something like this:
1094 Benchmark: running a, b, each for at least 5 CPU seconds...
1095 a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
1096 b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
1098 New features: "each for at least N CPU seconds...", "wallclock secs",
1099 and the "@ operations/CPU second (n=operations)".
1101 timethese() now returns a reference to a hash of Benchmark objects containing
1102 the test results, keyed on the names of the tests.
1104 timethis() now returns the iterations field in the Benchmark result object
1107 timethese(), timethis(), and the new cmpthese() (see below) can also take
1108 a format specifier of 'none' to suppress output.
1110 A new function countit() is just like timeit() except that it takes a
1111 TIME instead of a COUNT.
1113 A new function cmpthese() prints a chart comparing the results of each test
1114 returned from a timethese() call. For each possible pair of tests, the
1115 percentage speed difference (iters/sec or seconds/iter) is shown.
1117 For other details, see L<Benchmark>.
1121 The Devel::Peek module provides access to the internal representation
1122 of Perl variables and data. It is a data debugging tool for the XS programmer.
1124 =item ExtUtils::MakeMaker
1126 change#4135, also needs docs in module pod
1127 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1131 More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
1132 large (more than 4G) file access (64-bit support is not yet
1133 working, though, so no need to get overly excited), Free/Net/OpenBSD
1134 locking behaviour flags F_FLOCK, F_POSIX, Linux F_SHLCK, and
1135 O_ACCMODE: the mask of O_RDONLY, O_WRONLY, and O_RDWR.
1139 A compare_text() function has been added, which allows custom
1140 comparison functions. See L<File::Compare>.
1144 File::Find now works correctly when the wanted() function is either
1145 autoloaded or is a symbolic reference.
1147 A bug that caused File::Find to lose track of the working directory
1148 when pruning top-level directories has been fixed.
1150 File::Find now also supports several other options to control its
1151 behavior. It can follow symbolic links if the C<follow> option is
1152 specified. Enabling the C<no_chdir> option will make File::Find skip
1153 changing the current directory when walking directories. The C<untaint>
1154 flag can be useful when running with taint checks enabled.
1160 This extension implements BSD-style file globbing. By default,
1161 it will also be used for the internal implementation of the glob()
1162 operator. See L<File::Glob>.
1166 New methods have been added to the File::Spec module: devnull() returns
1167 the name of the null device (/dev/null on Unix) and tmpdir() the name of
1168 the temp directory (normally /tmp on Unix). There are now also methods
1169 to convert between absolute and relative filenames: abs2rel() and
1170 rel2abs(). For compatibility with operating systems that specify volume
1171 names in file paths, the splitpath(), splitdir(), and catdir() methods
1174 =item File::Spec::Functions
1176 The new File::Spec::Functions modules provides a function interface
1177 to the File::Spec module. Allows shorthand
1179 $fullname = catfile($dir1, $dir2, $file);
1183 $fullname = File::Spec->catfile($dir1, $dir2, $file);
1187 Getopt::Long licensing has changed to allow the Perl Artistic License
1188 as well as the GPL. It used to be GPL only, which got in the way of
1189 non-GPL applications that wanted to use Getopt::Long.
1191 Getopt::Long encourages the use of Pod::Usage to produce help
1192 messages. For example:
1198 GetOptions('help|?' => \$help, man => \$man) or pod2usage(2);
1199 pod2usage(1) if $help;
1200 pod2usage(-exitstatus => 0, -verbose => 2) if $man;
1206 sample - Using GetOpt::Long and Pod::Usage
1210 sample [options] [file ...]
1213 -help brief help message
1214 -man full documentation
1222 Print a brief help message and exits.
1226 Prints the manual page and exits.
1232 B<This program> will read the given input file(s) and do someting
1233 useful with the contents thereof.
1237 See L<Pod::Usage> for details.
1239 A bug that prevented the non-option call-back E<lt>E<gt> from being
1240 specified as the first argument has been fixed.
1242 To specify the characters E<lt> and E<gt> as option starters, use
1243 E<gt>E<lt>. Note, however, that changing option starters is strongly
1248 write() and syswrite() will now accept a single-argument
1249 form of the call, for consistency with Perl's syswrite().
1251 You can now create a TCP-based IO::Socket::INET without forcing
1252 a connect attempt. This allows you to configure its options
1253 (like making it non-blocking) and then call connect() manually.
1255 A bug that prevented the IO::Socket::protocol() accessor
1256 from ever returning the correct value has been corrected.
1260 Java Perl Lingo is now distributed with Perl. See jpl/README
1261 for more information.
1265 C<use lib> now weeds out any trailing duplicate entries.
1266 C<no lib> removes all named entries.
1270 The bitwise operations C<E<lt>E<lt>>, C<E<gt>E<gt>>, C<&>, C<|>,
1271 and C<~> are now supported on bigints.
1275 The accessor methods Re, Im, arg, abs, rho, and theta can now also
1276 act as mutators (accessor $z->Re(), mutator $z->Re(3)).
1280 A little bit of radial trigonometry (cylindrical and spherical),
1281 radial coordinate conversions, and the great circle distance were added.
1285 [TODO - Brad Appleton <bradapp@enteract.com>]
1287 =item Pod::Text and Pod::Man
1289 [TODO - Russ Allbery <rra@stanford.edu>]
1293 An EXISTS method has been added to this module (and sdbm_exists() has
1294 been added to the underlying sdbm library), so one can now call exists
1295 on an SDBM_File tied hash and get the correct result, rather than a
1298 A bug that may have caused data loss when more than one disk block
1299 happens to be read from the database in a single FETCH() has been
1304 The timelocal() and timegm() functions used to silently return bogus
1305 results when the date fell outside the machine's integer range. They
1306 now consistently croak() if the date falls in an unsupported range.
1310 The error return value in list context has been changed for all functions
1311 that return a list of values. Previously these functions returned a list
1312 with a single element C<undef> if an error occurred. Now these functions
1313 return the empty list in these situations. This applies to the following
1319 The remaining functions are unchanged and continue to return C<undef> on
1320 error even in list context.
1322 The Win32::SetLastError(ERROR) function has been added as a complement
1323 to the Win32::GetLastError() function.
1325 The new Win32::GetFullPathName(FILENAME) returns the full absolute
1326 pathname for FILENAME in scalar context. In list context it returns
1327 a two-element list containing the fully qualified directory name and
1328 the filename. See L<Win32>.
1332 A new feature called "DBM Filters" has been added to all the
1333 DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
1334 DBM Filters add four new methods to each DBM module:
1341 These can be used to filter key-value pairs before the pairs are
1342 written to the database or just after they are read from the database.
1343 See L<perldbmfilter> for further information.
1349 C<use attrs> is now obsolete, and is only provided for
1350 backward-compatibility. It's been replaced by the C<sub : attributes>
1351 syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>.
1353 C<use utf8> to enable UTF-8 and Unicode support.
1355 C<use caller 'encoding'> allows modules to inherit pragmatic attributes
1356 from the caller's context. C<encoding> is currently the only supported
1359 Lexical warnings pragma, C<use warnings;>, to control optional warnings.
1362 C<use filetest> to control the behaviour of filetests (C<-r> C<-w>
1363 ...). Currently only one subpragma implemented, "use filetest
1364 'access';", that uses access(2) or equivalent to check permissions
1365 instead of using stat(2) as usual. This matters in filesystems
1366 where there are ACLs (access control lists): the stat(2) might lie,
1367 but access(2) knows better.
1369 =head1 Utility Changes
1373 [TODO - Kurt Starsinic <kstar@chapin.edu>]
1377 C<perlcc> now supports the C and Bytecode backends. By default,
1378 it generates output from the simple C backend rather than the
1379 optimized C backend.
1381 Support for non-Unix platforms has been improved.
1386 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1388 =head1 Documentation Changes
1392 =item perlcompile.pod
1394 An introduction to using the Perl Compiler suite.
1396 =item perlfilter.pod
1398 An introduction to writing Perl source filters.
1402 Some guidelines for hacking the Perl source code.
1404 =item perlopentut.pod
1406 A tutorial on using open() effectively.
1408 =item perlreftut.pod
1410 A tutorial that introduces the essentials of references.
1414 A tutorial on managing class data for object modules.
1418 =head1 New or Changed Diagnostics
1422 =item "my sub" not yet implemented
1424 (F) Lexically scoped subroutines are not yet implemented. Don't try that
1427 =item '!' allowed only after types %s
1429 (F) The '!' is allowed in pack() and unpack() only after certain types.
1430 See L<perlfunc/pack>.
1432 =item / cannot take a count
1434 (F) You had an unpack template indicating a counted-length string,
1435 but you have also specified an explicit size for the string.
1436 See L<perlfunc/pack>.
1438 =item / must be followed by a, A or Z
1440 (F) You had an unpack template indicating a counted-length string,
1441 which must be followed by one of the letters a, A or Z
1442 to indicate what sort of string is to be unpacked.
1443 See L<perlfunc/pack>.
1445 =item / must be followed by a*, A* or Z*
1447 (F) You had a pack template indicating a counted-length string,
1448 Currently the only things that can have their length counted are a*, A* or Z*.
1449 See L<perlfunc/pack>.
1451 =item / must follow a numeric type
1453 (F) You had an unpack template that contained a '#',
1454 but this did not follow some numeric unpack specification.
1455 See L<perlfunc/pack>.
1457 =item /%s/: Unrecognized escape \\%c passed through
1459 (W) You used a backslash-character combination which is not recognized
1460 by Perl. This combination appears in an interpolated variable or a
1461 C<'>-delimited regular expression. The character was understood literally.
1463 =item /%s/: Unrecognized escape \\%c in character class passed through
1465 (W) You used a backslash-character combination which is not recognized
1466 by Perl inside character classes. The character was understood literally.
1468 =item /%s/ should probably be written as "%s"
1470 (W) You have used a pattern where Perl expected to find a string,
1471 as in the first argument to C<join>. Perl will treat the true
1472 or false result of matching the pattern against $_ as the string,
1473 which is probably not what you had in mind.
1475 =item %s() called too early to check prototype
1477 (W) You've called a function that has a prototype before the parser saw a
1478 definition or declaration for it, and Perl could not check that the call
1479 conforms to the prototype. You need to either add an early prototype
1480 declaration for the subroutine in question, or move the subroutine
1481 definition ahead of the call to get proper prototype checking. Alternatively,
1482 if you are certain that you're calling the function correctly, you may put
1483 an ampersand before the name to avoid the warning. See L<perlsub>.
1485 =item %s package attribute may clash with future reserved word: %s
1487 (W) A lowercase attribute name was used that had a package-specific handler.
1488 That name might have a meaning to Perl itself some day, even though it
1489 doesn't yet. Perhaps you should use a mixed-case attribute name, instead.
1492 =item (in cleanup) %s
1494 (W) This prefix usually indicates that a DESTROY() method raised
1495 the indicated exception. Since destructors are usually called by
1496 the system at arbitrary points during execution, and often a vast
1497 number of times, the warning is issued only once for any number
1498 of failures that would otherwise result in the same message being
1501 Failure of user callbacks dispatched using the C<G_KEEPERR> flag
1502 could also result in this warning. See L<perlcall/G_KEEPERR>.
1504 =item <> should be quotes
1506 (F) You wrote C<require E<lt>fileE<gt>> when you should have written
1509 =item Attempt to join self
1511 (F) You tried to join a thread from within itself, which is an
1512 impossible task. You may be joining the wrong thread, or you may
1513 need to move the join() to some other thread.
1515 =item Bad evalled substitution pattern
1517 (F) You've used the /e switch to evaluate the replacement for a
1518 substitution, but perl found a syntax error in the code to evaluate,
1519 most likely an unexpected right brace '}'.
1521 =item Bad realloc() ignored
1523 (S) An internal routine called realloc() on something that had never been
1524 malloc()ed in the first place. Mandatory, but can be disabled by
1525 setting environment variable C<PERL_BADFREE> to 1.
1527 =item Binary number > 0b11111111111111111111111111111111 non-portable
1529 (W) The binary number you specified is larger than 2**32-1
1530 (4294967295) and therefore non-portable between systems. See
1531 L<perlport> for more on portability concerns.
1533 =item Bit vector size > 32 non-portable
1535 (W) Using bit vector sizes larger than 32 is non-portable.
1537 =item Buffer overflow in prime_env_iter: %s
1539 (W) A warning peculiar to VMS. While Perl was preparing to iterate over
1540 %ENV, it encountered a logical name or symbol definition which was too long,
1541 so it was truncated to the string shown.
1543 =item Can't check filesystem of script "%s"
1545 (P) For some reason you can't check the filesystem of the script for nosuid.
1547 =item Can't ignore signal CHLD, forcing to default
1549 (W) Perl has detected that it is being run with the SIGCHLD signal
1550 (sometimes known as SIGCLD) disabled. Since disabling this signal
1551 will interfere with proper determination of exit status of child
1552 processes, Perl has reset the signal to its default value.
1553 This situation typically indicates that the parent program under
1554 which Perl may be running (e.g. cron) is being very careless.
1556 =item Can't modify non-lvalue subroutine call
1558 (F) Subroutines meant to be used in lvalue context should be declared as
1559 such, see L<perlsub/"Lvalue subroutines">.
1561 =item Can't read CRTL environ
1563 (S) A warning peculiar to VMS. Perl tried to read an element of %ENV
1564 from the CRTL's internal environment array and discovered the array was
1565 missing. You need to figure out where your CRTL misplaced its environ
1566 or define F<PERL_ENV_TABLES> (see L<perlvms>) so that environ is not searched.
1568 =item Can't remove %s: %s, skipping file
1570 (S) You requested an inplace edit without creating a backup file. Perl
1571 was unable to remove the original file to replace it with the modified
1572 file. The file was left unmodified.
1574 =item Can't return %s from lvalue subroutine
1576 (F) Perl detected an attempt to return illegal lvalues (such
1577 as temporary or readonly values) from a subroutine used as an lvalue.
1578 This is not allowed.
1580 =item Can't weaken a nonreference
1582 (F) You attempted to weaken something that was not a reference. Only
1583 references can be weakened.
1585 =item Character class [:%s:] unknown
1587 (F) The class in the character class [: :] syntax is unknown.
1590 =item Character class syntax [%s] belongs inside character classes
1592 (W) The character class constructs [: :], [= =], and [. .] go
1593 I<inside> character classes, the [] are part of the construct,
1594 for example: /[012[:alpha:]345]/. Note that [= =] and [. .]
1595 are not currently implemented; they are simply placeholders for
1598 =item Constant is not %s reference
1600 (F) A constant value (perhaps declared using the C<use constant> pragma)
1601 is being dereferenced, but it amounts to the wrong type of reference. The
1602 message indicates the type of reference that was expected. This usually
1603 indicates a syntax error in dereferencing the constant value.
1604 See L<perlsub/"Constant Functions"> and L<constant>.
1606 =item constant(%s): %%^H is not localized
1608 (F) When setting compile-time-lexicalized hash %^H one should set the
1609 corresponding bit of $^H as well.
1611 =item constant(%s): %s
1613 (F) Compile-time-substitutions (such as overloaded constants and
1614 character names) were not correctly set up.
1616 =item defined(@array) is deprecated
1618 (D) defined() is not usually useful on arrays because it checks for an
1619 undefined I<scalar> value. If you want to see if the array is empty,
1620 just use C<if (@array) { # not empty }> for example.
1622 =item defined(%hash) is deprecated
1624 (D) defined() is not usually useful on hashes because it checks for an
1625 undefined I<scalar> value. If you want to see if the hash is empty,
1626 just use C<if (%hash) { # not empty }> for example.
1628 =item Did not produce a valid header
1632 =item Document contains no data
1636 =item entering effective %s failed
1638 (F) While under the C<use filetest> pragma, switching the real and
1639 effective uids or gids failed.
1641 =item false [] range "%s" in regexp
1643 (W) A character class range must start and end at a literal character, not
1644 another character class like C<\d> or C<[:alpha:]>. The "-" in your false
1645 range is interpreted as a literal "-". Consider quoting the "-", "\-".
1648 =item Filehandle %s opened only for output
1650 (W) You tried to read from a filehandle opened only for writing. If you
1651 intended it to be a read/write filehandle, you needed to open it with
1652 "+E<lt>" or "+E<gt>" or "+E<gt>E<gt>" instead of with "E<lt>" or nothing. If
1653 you intended only to read from the file, use "E<lt>". See
1656 =item Hexadecimal number > 0xffffffff non-portable
1658 (W) The hexadecimal number you specified is larger than 2**32-1
1659 (4294967295) and therefore non-portable between systems. See
1660 L<perlport> for more on portability concerns.
1662 =item Ill-formed CRTL environ value "%s"
1664 (W) A warning peculiar to VMS. Perl tried to read the CRTL's internal
1665 environ array, and encountered an element without the C<=> delimiter
1666 used to spearate keys from values. The element is ignored.
1668 =item Ill-formed message in prime_env_iter: |%s|
1670 (W) A warning peculiar to VMS. Perl tried to read a logical name
1671 or CLI symbol definition when preparing to iterate over %ENV, and
1672 didn't see the expected delimiter between key and value, so the
1675 =item Illegal binary digit %s
1677 (F) You used a digit other than 0 or 1 in a binary number.
1679 =item Illegal binary digit %s ignored
1681 (W) You may have tried to use a digit other than 0 or 1 in a binary number.
1682 Interpretation of the binary number stopped before the offending digit.
1684 =item Illegal number of bits in vec
1686 (F) The number of bits in vec() (the third argument) must be a power of
1687 two from 1 to 32 (or 64, if your platform supports that).
1689 =item Integer overflow in %s number
1691 (W) The hexadecimal, octal or binary number you have specified either
1692 as a literal or as an argument to hex() or oct() is too big for your
1693 architecture, and has been converted to a floating point number. On a
1694 32-bit architecture the largest hexadecimal, octal or binary number
1695 representable without overflow is 0xFFFFFFFF, 037777777777, or
1696 0b11111111111111111111111111111111 respectively. Note that Perl
1697 transparently promotes all numbers to a floating point representation
1698 internally--subject to loss of precision errors in subsequent
1701 =item Invalid %s attribute: %s
1703 The indicated attribute for a subroutine or variable was not recognized
1704 by Perl or by a user-supplied handler. See L<attributes>.
1706 =item Invalid %s attributes: %s
1708 The indicated attributes for a subroutine or variable were not recognized
1709 by Perl or by a user-supplied handler. See L<attributes>.
1711 =item invalid [] range "%s" in regexp
1713 The offending range is now explicitly displayed.
1715 =item Invalid separator character %s in attribute list
1717 (F) Something other than a comma or whitespace was seen between the
1718 elements of an attribute list. If the previous attribute
1719 had a parenthesised parameter list, perhaps that list was terminated
1720 too soon. See L<attributes>.
1722 =item Invalid separator character %s in subroutine attribute list
1724 (F) Something other than a comma or whitespace was seen between the
1725 elements of a subroutine attribute list. If the previous attribute
1726 had a parenthesised parameter list, perhaps that list was terminated
1729 =item leaving effective %s failed
1731 (F) While under the C<use filetest> pragma, switching the real and
1732 effective uids or gids failed.
1734 =item Lvalue subs returning %s not implemented yet
1736 (F) Due to limitations in the current implementation, array and hash
1737 values cannot be returned in subroutines used in lvalue context.
1738 See L<perlsub/"Lvalue subroutines">.
1740 =item Method %s not permitted
1744 =item Missing %sbrace%s on \N{}
1746 (F) Wrong syntax of character name literal C<\N{charname}> within
1747 double-quotish context.
1749 =item Missing command in piped open
1751 (W) You used the C<open(FH, "| command")> or C<open(FH, "command |")>
1752 construction, but the command was missing or blank.
1754 =item Missing name in "my sub"
1756 (F) The reserved syntax for lexically scoped subroutines requires that they
1757 have a name with which they can be found.
1759 =item no UTC offset information; assuming local time is UTC
1761 (S) A warning peculiar to VMS. Perl was unable to find the local
1762 timezone offset, so it's assuming that local system time is equivalent
1763 to UTC. If it's not, define the logical name F<SYS$TIMEZONE_DIFFERENTIAL>
1764 to translate to the number of seconds which need to be added to UTC to
1767 =item Octal number > 037777777777 non-portable
1769 (W) The octal number you specified is larger than 2**32-1 (4294967295)
1770 and therefore non-portable between systems. See L<perlport> for more
1771 on portability concerns.
1773 See also L<perlport> for writing portable code.
1775 =item panic: del_backref
1777 (P) Failed an internal consistency check while trying to reset a weak
1780 =item panic: kid popen errno read
1782 (F) forked child returned an incomprehensible message about its errno.
1784 =item panic: magic_killbackrefs
1786 (P) Failed an internal consistency check while trying to reset all weak
1787 references to an object.
1789 =item Possible Y2K bug: %s
1791 (W) You are concatenating the number 19 with another number, which
1792 could be a potential Year 2000 problem.
1794 =item Premature end of script headers
1798 =item Repeat count in pack overflows
1800 (F) You can't specify a repeat count so large that it overflows
1801 your signed integers. See L<perlfunc/pack>.
1803 =item Repeat count in unpack overflows
1805 (F) You can't specify a repeat count so large that it overflows
1806 your signed integers. See L<perlfunc/unpack>.
1808 =item realloc() of freed memory ignored
1810 (S) An internal routine called realloc() on something that had already
1813 =item Reference is already weak
1815 (W) You have attempted to weaken a reference that is already weak.
1816 Doing so has no effect.
1818 =item setpgrp can't take arguments
1820 (F) Your system has the setpgrp() from BSD 4.2, which takes no arguments,
1821 unlike POSIX setpgid(), which takes a process ID and process group ID.
1823 =item Strange *+?{} on zero-length expression
1825 (W) You applied a regular expression quantifier in a place where it
1826 makes no sense, such as on a zero-width assertion.
1827 Try putting the quantifier inside the assertion instead. For example,
1828 the way to match "abc" provided that it is followed by three
1829 repetitions of "xyz" is C</abc(?=(?:xyz){3})/>, not C</abc(?=xyz){3}/>.
1831 =item switching effective %s is not implemented
1833 (F) While under the C<use filetest> pragma, we cannot switch the
1834 real and effective uids or gids.
1836 =item This Perl can't reset CRTL environ elements (%s)
1838 =item This Perl can't set CRTL environ elements (%s=%s)
1840 (W) Warnings peculiar to VMS. You tried to change or delete an element
1841 of the CRTL's internal environ array, but your copy of Perl wasn't
1842 built with a CRTL that contained the setenv() function. You'll need to
1843 rebuild Perl with a CRTL that does, or redefine F<PERL_ENV_TABLES> (see
1844 L<perlvms>) so that the environ array isn't the target of the change to
1845 %ENV which produced the warning.
1847 =item Unknown open() mode '%s'
1849 (F) The second argument of 3-argument open() is not among the list
1850 of valid modes: C<L<lt>>, C<L<gt>>, C<E<gt>E<gt>>, C<+L<lt>>,
1851 C<+L<gt>>, C<+E<gt>E<gt>>, C<-|>, C<|->.
1853 =item Unknown process %x sent message to prime_env_iter: %s
1855 (P) An error peculiar to VMS. Perl was reading values for %ENV before
1856 iterating over it, and someone else stuck a message in the stream of
1857 data Perl expected. Someone's very confused, or perhaps trying to
1858 subvert Perl's population of %ENV for nefarious purposes.
1860 =item Unrecognized escape \\%c passed through
1862 (W) You used a backslash-character combination which is not recognized
1863 by Perl. The character was understood literally.
1865 =item Unterminated attribute parameter in attribute list
1867 (F) The lexer saw an opening (left) parenthesis character while parsing an
1868 attribute list, but the matching closing (right) parenthesis
1869 character was not found. You may need to add (or remove) a backslash
1870 character to get your parentheses to balance. See L<attributes>.
1872 =item Unterminated attribute list
1874 (F) The lexer found something other than a simple identifier at the start
1875 of an attribute, and it wasn't a semicolon or the start of a
1876 block. Perhaps you terminated the parameter list of the previous attribute
1877 too soon. See L<attributes>.
1879 =item Unterminated attribute parameter in subroutine attribute list
1881 (F) The lexer saw an opening (left) parenthesis character while parsing a
1882 subroutine attribute list, but the matching closing (right) parenthesis
1883 character was not found. You may need to add (or remove) a backslash
1884 character to get your parentheses to balance.
1886 =item Unterminated subroutine attribute list
1888 (F) The lexer found something other than a simple identifier at the start
1889 of a subroutine attribute, and it wasn't a semicolon or the start of a
1890 block. Perhaps you terminated the parameter list of the previous attribute
1893 =item Value of CLI symbol "%s" too long
1895 (W) A warning peculiar to VMS. Perl tried to read the value of an %ENV
1896 element from a CLI symbol table, and found a resultant string longer
1897 than 1024 characters. The return value has been truncated to 1024
1900 =item Version number must be a constant number
1902 (P) The attempt to translate a C<use Module n.n LIST> statement into
1903 its equivalent C<BEGIN> block found an internal inconsistency with
1908 =head1 Obsolete Diagnostics
1912 =item Character class syntax [: :] is reserved for future extensions
1914 (W) Within regular expression character classes ([]) the syntax beginning
1915 with "[:" and ending with ":]" is reserved for future extensions.
1916 If you need to represent those character sequences inside a regular
1917 expression character class, just quote the square brackets with the
1918 backslash: "\[:" and ":\]".
1920 =item Ill-formed logical name |%s| in prime_env_iter
1922 (W) A warning peculiar to VMS. A logical name was encountered when preparing
1923 to iterate over %ENV which violates the syntactic rules governing logical
1924 names. Because it cannot be translated normally, it is skipped, and will not
1925 appear in %ENV. This may be a benign occurrence, as some software packages
1926 might directly modify logical name tables and introduce nonstandard names,
1927 or it may indicate that a logical name table has been corrupted.
1929 =item regexp too big
1931 (F) The current implementation of regular expressions uses shorts as
1932 address offsets within a string. Unfortunately this means that if
1933 the regular expression compiles to longer than 32767, it'll blow up.
1934 Usually when you want a regular expression this big, there is a better
1935 way to do it with multiple statements. See L<perlre>.
1937 =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
1939 (D) Perl versions before 5.004 misinterpreted any type marker followed
1940 by "$" and a digit. For example, "$$0" was incorrectly taken to mean
1941 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
1943 However, the developers of Perl 5.004 could not fix this bug completely,
1944 because at least two widely-used modules depend on the old meaning of
1945 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
1946 old (broken) way inside strings; but it generates this message as a
1947 warning. And in Perl 5.005, this special treatment will cease.
1953 If you find what you think is a bug, you might check the
1954 articles recently posted to the comp.lang.perl.misc newsgroup.
1955 There may also be information at http://www.perl.com/perl/, the Perl
1958 If you believe you have an unreported bug, please run the B<perlbug>
1959 program included with your release. Make sure to trim your bug down
1960 to a tiny but sufficient test case. Your bug report, along with the
1961 output of C<perl -V>, will be sent off to perlbug@perl.com to be
1962 analysed by the Perl porting team.
1966 The F<Changes> file for exhaustive details on what changed.
1968 The F<INSTALL> file for how to build Perl.
1970 The F<README> file for general stuff.
1972 The F<Artistic> and F<Copying> files for copyright information.
1976 Written by Gurusamy Sarathy <F<gsar@activestate.com>>, with many
1977 contributions from The Perl Porters.
1979 Send omissions or corrections to <F<perlbug@perl.com>>.