3 perldelta - what's new for perl v5.6.0
7 This document describes differences between the 5.005 release and this one.
9 =head1 Core Enhancements
11 =head2 Interpreter cloning, threads, and concurrency
13 Perl 5.005_63 introduces the beginnings of support for running multiple
14 interpreters concurrently in different threads. In conjunction with
15 the perl_clone() API call, which can be used to selectively duplicate
16 the state of any given interpreter, it is possible to compile a
17 piece of code once in an interpreter, clone that interpreter
18 one or more times, and run all the resulting interpreters in distinct
21 On the Windows platform, this feature is used to emulate fork() at the
22 interpreter level. See L<perlfork> for details about that.
24 This feature is still in evolution. It is eventually meant to be used
25 to selectively clone a subroutine and data reachable from that
26 subroutine in a separate interpreter and run the cloned subroutine
27 in a separate thread. Since there is no shared data between the
28 interpreters, little or no locking will be needed (unless parts of
29 the symbol table are explicitly shared). This is obviously intended
30 to be an easy-to-use replacement for the existing threads support.
32 Support for cloning interpreters and interpreter concurrency can be
33 enabled using the -Dusethreads Configure option (see win32/Makefile for
34 how to enable it on Windows.) The resulting perl executable will be
35 functionally identical to one that was built with -Dmultiplicity, but
36 the perl_clone() API call will only be available in the former.
38 -Dusethreads enables the cpp macro USE_ITHREADS by default, which in turn
39 enables Perl source code changes that provide a clear separation between
40 the op tree and the data it operates with. The former is immutable, and
41 can therefore be shared between an interpreter and all of its clones,
42 while the latter is considered local to each interpreter, and is therefore
43 copied for each clone.
45 Note that building Perl with the -Dusemultiplicity Configure option
46 is adequate if you wish to run multiple B<independent> interpreters
47 concurrently in different threads. -Dusethreads only provides the
48 additional functionality of the perl_clone() API call and other
49 support for running B<cloned> interpreters concurrently.
51 NOTE: This is an experimental feature. Implementation details are
54 =head2 Lexically scoped warning categories
56 You can now control the granularity of warnings emitted by perl at a finer
57 level using the C<use warnings> pragma. L<warnings> and L<perllexwarn>
58 have copious documentation on this feature.
60 =head2 Unicode and UTF-8 support
62 Perl now uses UTF-8 as its internal representation for character
63 strings. The C<utf8> and C<bytes> pragmas are used to control this support
64 in the current lexical scope. See L<perlunicode>, L<utf8> and L<bytes> for
67 This feature is expected to evolve quickly to support some form of I/O
68 disciplines that can be used to specify the kind of input and output data
69 (bytes or characters). Until that happens, additional modules from CPAN
70 will be needed to complete the toolkit for dealing with Unicode.
72 NOTE: This should be considered an experimental feature. Implementation
73 details are subject to change.
75 =head2 Support for interpolating named characters
77 The new C<\N> escape interpolates named characters within strings.
78 For example, C<"Hi! \N{WHITE SMILING FACE}"> evaluates to a string
79 with a unicode smiley face at the end.
81 =head2 "our" declarations
83 An "our" declaration introduces a value that can be best understood
84 as a lexically scoped symbolic alias to a global variable in the
85 package that was current where the variable was declared. This is
86 mostly useful as an alternative to the C<vars> pragma, but also provides
87 the opportunity to introduce typing and other attributes for such
88 variables. See L<perlfunc/our>.
90 =head2 Support for strings represented as a vector of ordinals
92 Literals of the form C<v1.2.3.4> are now parsed as a string composed
93 of characters with the specified ordinals. This is an alternative, more
94 readable way to construct (possibly unicode) strings instead of
95 interpolating characters, as in C<"\x{1}\x{2}\x{3}\x{4}">. The leading
96 C<v> may be omitted if there are more than two ordinals, so C<1.2.3> is
97 parsed the same as C<v1.2.3>.
99 Strings written in this form are also useful to represent version "numbers".
100 It is easy to compare such version "numbers" (which are really just plain
101 strings) using any of the usual string comparison operators C<eq>, C<ne>,
102 C<lt>, C<gt>, etc., or perform bitwise string operations on them using C<|>,
105 In conjunction with the new C<$^V> magic variable (which contains
106 the perl version as a string), such literals can be used as a readable way
107 to check if you're running a particular version of Perl:
109 # this will parse in older versions of Perl also
110 if ($^V and $^V gt v5.6.0) {
111 # new features supported
114 C<require> and C<use> also have some special magic to support such literals.
115 They will be interpreted as a version rather than as a module name:
117 require v5.6.0; # croak if $^V lt v5.6.0
118 use v5.6.0; # same, but croaks at compile-time
120 Alternatively, the C<v> may be omitted if there is more than one dot:
125 Also, C<sprintf> and C<printf> support the Perl-specific format flag C<%v>
126 to print ordinals of characters in arbitrary strings:
128 printf "v%vd", $^V; # prints current version, such as "v5.5.650"
129 printf "%*vX", ":", $addr; # formats IPv6 address
130 printf "%*vb", " ", $bits; # displays bitstring
132 See L<perldata/"Scalar value constructors"> for additional information.
134 =head2 Improved Perl version numbering system
136 Beginning with Perl version 5.6.0, the version number convention has been
137 changed to a "dotted integer" scheme that is more commonly found in open
140 Maintenance versions of v5.6.0 will be released as v5.6.1, v5.6.2 etc.
141 The next development series following v5.6.0 will be numbered v5.7.x,
142 beginning with v5.7.0, and the next major production release following
143 v5.6.0 will be v5.8.0.
145 The English module now sets $PERL_VERSION to $^V (a string value) rather
146 than C<$]> (a numeric value). (This is a potential incompatibility.
147 Send us a report via perlbug if you are affected by this.)
149 The v1.2.3 syntax is also now legal in Perl.
150 See L<Support for strings represented as a vector of ordinals> for more on that.
152 To cope with the new versioning system's use of at least three significant
153 digits for each version component, the method used for incrementing the
154 subversion number has also changed slightly. We assume that versions older
155 than v5.6.0 have been incrementing the subversion component in multiples of
156 10. Versions after v5.6.0 will increment them by 1. Thus, using the new
157 notation, 5.005_03 is the "same" as v5.5.30, and the first maintenance
158 version following v5.6.0 will be v5.6.1 (which should be read as being
159 equivalent to a floating point value of 5.006_001 in the older format,
162 =head2 New syntax for declaring subroutine attributes
164 Formerly, if you wanted to mark a subroutine as being a method call or
165 as requiring an automatic lock() when it is entered, you had to declare
166 that with a C<use attrs> pragma in the body of the subroutine.
167 That can now be accomplished with declaration syntax, like this:
169 sub mymethod : locked method ;
171 sub mymethod : locked method {
175 sub othermethod :locked :method ;
177 sub othermethod :locked :method {
182 (Note how only the first C<:> is mandatory, and whitespace surrounding
183 the C<:> is optional.)
185 F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes
186 with the stubs they provide. See L<attributes>.
188 =head2 File and directory handles can be autovivified
190 Similar to how constructs such as C<< $x->[0] >> autovivify a reference,
191 handle constructors (open(), opendir(), pipe(), socketpair(), sysopen(),
192 socket(), and accept()) now autovivify a file or directory handle
193 if the handle passed to them is an uninitialized scalar variable. This
194 allows the constructs such as C<open(my $fh, ...)> and C<open(local $fh,...)>
195 to be used to create filehandles that will conveniently be closed
196 automatically when the scope ends, provided there are no other references
197 to them. This largely eliminates the need for typeglobs when opening
198 filehandles that must be passed around, as in the following example:
202 or die "Can't open '@_': $!";
207 my $f = myopen("</etc/motd");
209 # $f implicitly closed here
212 =head2 open() with more than two arguments
214 If open() is passed three arguments instead of two, the second argument
215 is used as the mode and the third argument is taken to be the file name.
216 This is primarily useful for protecting against unintended magic behavior
217 of the traditional two-argument form. See L<perlfunc/open>.
219 =head2 64-bit support
221 Any platform that has 64-bit integers either
223 (1) natively as longs or ints
224 (2) via special compiler flags
225 (3) using long long or int64_t
227 is able to use "quads" (64-bit integers) as follows:
233 constants (decimal, hexadecimal, octal, binary) in the code
237 arguments to oct() and hex()
241 arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)
249 pack() and unpack() "q" and "Q" formats
253 in basic arithmetics: + - * / % (NOTE: operating close to the limits
254 of the integer values may produce surprising results)
258 in bit arithmetics: & | ^ ~ << >> (NOTE: these used to be forced
259 to be 32 bits wide but now operate on the full native width.)
267 Note that unless you have the case (a) you will have to configure
268 and compile Perl using the -Duse64bitint Configure flag.
270 NOTE: The Configure flags -Duselonglong and -Duse64bits have been
271 deprecated. Use -Duse64bitint instead.
273 There are actually two modes of 64-bitness: the first one is achieved
274 using Configure -Duse64bitint and the second one using Configure
275 -Duse64bitall. The difference is that the first one is minimal and
276 the second one maximal. The first works in more places than the second.
278 The C<use64bitint> does only as much as is required to get 64-bit
279 integers into Perl (this may mean, for example, using "long longs")
280 while your memory may still be limited to 2 gigabytes (because your
281 pointers could still be 32-bit). Note that the name C<64bitint> does
282 not imply that your C compiler will be using 64-bit C<int>s (it might,
283 but it doesn't have to): the C<use64bitint> means that you will be
284 able to have 64 bits wide scalar values.
286 The C<use64bitall> goes all the way by attempting to switch also
287 integers (if it can), longs (and pointers) to being 64-bit. This may
288 create an even more binary incompatible Perl than -Duse64bitint: the
289 resulting executable may not run at all in a 32-bit box, or you may
290 have to reboot/reconfigure/rebuild your operating system to be 64-bit
293 Natively 64-bit systems like Alpha and Cray need neither -Duse64bitint
296 Last but not least: note that due to Perl's habit of always using
297 floating point numbers, the quads are still not true integers.
298 When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned,
299 -9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they
300 are silently promoted to floating point numbers, after which they will
301 start losing precision (in their lower digits).
303 NOTE: 64-bit support is still experimental on most platforms.
304 Existing support only covers the LP64 data model. In particular, the
305 LLP64 data model is not yet supported. 64-bit libraries and system
306 APIs on many platforms have not stabilized--your mileage may vary.
308 =head2 Large file support
310 If you have filesystems that support "large files" (files larger than
311 2 gigabytes), you may now also be able to create and access them from
314 NOTE: The default action is to enable large file support, if
315 available on the platform.
317 If the large file support is on, and you have a Fcntl constant
318 O_LARGEFILE, the O_LARGEFILE is automatically added to the flags
321 Beware that unless your filesystem also supports "sparse files" seeking
322 to umpteen petabytes may be inadvisable.
324 Note that in addition to requiring a proper file system to do large
325 files you may also need to adjust your per-process (or your
326 per-system, or per-process-group, or per-user-group) maximum filesize
327 limits before running Perl scripts that try to handle large files,
328 especially if you intend to write such files.
330 Finally, in addition to your process/process group maximum filesize
331 limits, you may have quota limits on your filesystems that stop you
332 (your user id or your user group id) from using large files.
334 Adjusting your process/user/group/file system/operating system limits
335 is outside the scope of Perl core language. For process limits, you
336 may try increasing the limits using your shell's limits/limit/ulimit
337 command before running Perl. The BSD::Resource extension (not
338 included with the standard Perl distribution) may also be of use, it
339 offers the getrlimit/setrlimit interface that can be used to adjust
340 process resource usage limits, including the maximum filesize limit.
344 In some systems you may be able to use long doubles to enhance the
345 range and precision of your double precision floating point numbers
346 (that is, Perl's numbers). Use Configure -Duselongdouble to enable
347 this support (if it is available).
351 You can "Configure -Dusemorebits" to turn on both the 64-bit support
352 and the long double support.
354 =head2 Enhanced support for sort() subroutines
356 Perl subroutines with a prototype of C<($$)>, and XSUBs in general, can
357 now be used as sort subroutines. In either case, the two elements to
358 be compared are passed as normal parameters in @_. See L<perlfunc/sort>.
360 For unprototyped sort subroutines, the historical behavior of passing
361 the elements to be compared as the global variables $a and $b remains
364 =head2 C<sort $coderef @foo> allowed
366 sort() did not accept a subroutine reference as the comparison
367 function in earlier versions. This is now permitted.
369 =head2 File globbing implemented internally
371 Perl now uses the File::Glob implementation of the glob() operator
372 automatically. This avoids using an external csh process and the
373 problems associated with it.
375 NOTE: This is currently an experimental feature. Interfaces and
376 implementation are subject to change.
378 =item Support for CHECK blocks
380 In addition to C<BEGIN>, C<INIT>, C<END>, C<DESTROY> and C<AUTOLOAD>,
381 subroutines named C<CHECK> are now special. These are queued up during
382 compilation and behave similar to END blocks, except they are called at
383 the end of compilation rather than at the end of execution. They cannot
386 =head2 POSIX character class syntax [: :] supported
388 For example to match alphabetic characters use /[[:alpha:]]/.
389 See L<perlre> for details.
391 =item Better pseudo-random number generator
393 In 5.005_0x and earlier, perl's rand() function used the C library
394 rand(3) function. As of 5.005_52, Configure tests for drand48(),
395 random(), and rand() (in that order) and picks the first one it finds.
397 These changes should result in better random numbers from rand().
399 =head2 Improved C<qw//> operator
401 The C<qw//> operator is now evaluated at compile time into a true list
402 instead of being replaced with a run time call to C<split()>. This
403 removes the confusing misbehaviour of C<qw//> in scalar context, which
404 had inherited that behaviour from split().
408 $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
410 now correctly prints "3|a", instead of "2|a".
412 =item Better worst-case behavior of hashes
414 Small changes in the hashing algorithm have been implemented in
415 order to improve the distribution of lower order bits in the
416 hashed value. This is expected to yield better performance on
417 keys that are repeated sequences.
419 =head2 pack() format 'Z' supported
421 The new format type 'Z' is useful for packing and unpacking null-terminated
422 strings. See L<perlfunc/"pack">.
424 =head2 pack() format modifier '!' supported
426 The new format type modifier '!' is useful for packing and unpacking
427 native shorts, ints, and longs. See L<perlfunc/"pack">.
429 =head2 pack() and unpack() support counted strings
431 The template character '/' can be used to specify a counted string
432 type to be packed or unpacked. See L<perlfunc/"pack">.
434 =head2 Comments in pack() templates
436 The '#' character in a template introduces a comment up to
437 end of the line. This facilitates documentation of pack()
440 =head2 Weak references
442 In previous versions of Perl, you couldn't cache objects so as
443 to allow them to be deleted if the last reference from outside
444 the cache is deleted. The reference in the cache would hold a
445 reference count on the object and the objects would never be
448 Another familiar problem is with circular references. When an
449 object references itself, its reference count would never go
450 down to zero, and it would not get destroyed until the program
453 Weak references solve this by allowing you to "weaken" any
454 reference, that is, make it not count towards the reference count.
455 When the last non-weak reference to an object is deleted, the object
456 is destroyed and all the weak references to the object are
457 automatically undef-ed.
459 To use this feature, you need the WeakRef package from CPAN, which
460 contains additional documentation.
462 NOTE: This is an experimental feature. Details are subject to change.
464 =head2 Binary numbers supported
466 Binary numbers are now supported as literals, in s?printf formats, and
470 printf "The answer is: %b\n", oct("0b101010");
472 =head2 Lvalue subroutines
474 Subroutines can now return modifiable lvalues.
475 See L<perlsub/"Lvalue subroutines">.
477 NOTE: This is an experimental feature. Details are subject to change.
479 =head2 Some arrows may be omitted in calls through references
481 Perl now allows the arrow to be omitted in many constructs
482 involving subroutine calls through references. For example,
483 C<< $foo[10]->('foo') >> may now be written C<$foo[10]('foo')>.
484 This is rather similar to how the arrow may be omitted from
485 C<< $foo[10]->{'foo'} >>. Note however, that the arrow is still
486 required for C<< foo(10)->('bar') >>.
488 =head2 Boolean assignment operators are legal lvalues
490 Constructs such as C<($a ||= 2) += 1> are now allowed.
492 =head2 exists() is supported on subroutine names
494 The exists() builtin now works on subroutine names. A subroutine
495 is considered to exist if it has been declared (even if implicitly).
496 See L<perlfunc/exists> for examples.
498 =head2 exists() and delete() are supported on array elements
500 The exists() and delete() builtins now work on simple arrays as well.
501 The behavior is similar to that on hash elements.
503 exists() can be used to check whether an array element has been
504 initialized. This avoids autovivifying array elements that don't exist.
505 If the array is tied, the EXISTS() method in the corresponding tied
506 package will be invoked.
508 delete() may be used to remove an element from the array and return
509 it. The array element at that position returns to its uninitialized
510 state, so that testing for the same element with exists() will return
511 false. If the element happens to be the one at the end, the size of
512 the array also shrinks up to the highest element that tests true for
513 exists(), or 0 if none such is found. If the array is tied, the DELETE()
514 method in the corresponding tied package will be invoked.
516 See L<perlfunc/exists> and L<perlfunc/delete> for examples.
518 =head2 Pseudo-hashes work better
520 Dereferencing some types of reference values in a pseudo-hash,
521 such as C<< $ph->{foo}[1] >>, was accidentally disallowed. This has
524 When applied to a pseudo-hash element, exists() now reports whether
525 the specified value exists, not merely if the key is valid.
527 delete() now works on pseudo-hashes. When given a pseudo-hash element
528 or slice it deletes the values corresponding to the keys (but not the keys
529 themselves). See L<perlref/"Pseudo-hashes: Using an array as a hash">.
531 Pseudo-hash slices with constant keys are now optimized to array lookups
534 List assignments to pseudo-hash slices are now supported.
536 The C<fields> pragma now provides ways to create pseudo-hashes, via
537 fields::new() and fields::phash(). See L<fields>.
539 NOTE: The pseudo-hash data type continues to be experimental.
540 Limiting oneself to the interface elements provided by the
541 fields pragma will provide protection from any future changes.
543 =head2 Automatic flushing of output buffers
545 fork(), exec(), system(), qx//, and pipe open()s now flush buffers
546 of all files opened for output when the operation was attempted. This
547 mostly eliminates confusing buffering mishaps suffered by users unaware
548 of how Perl internally handles I/O.
550 This is not supported on some platforms like Solaris where a suitably
551 correct implementation of fflush(NULL) isn't available.
553 =head2 Better diagnostics on meaningless filehandle operations
555 Constructs such as C<< open(<FH>) >> and C<< close(<FH>) >>
556 are compile time errors. Attempting to read from filehandles that
557 were opened only for writing will now produce warnings (just as
558 writing to read-only filehandles does).
560 =head2 Where possible, buffered data discarded from duped input filehandle
562 C<< open(NEW, "<&OLD") >> now attempts to discard any data that
563 was previously read and buffered in C<OLD> before duping the handle.
564 On platforms where doing this is allowed, the next read operation
565 on C<NEW> will return the same data as the corresponding operation
566 on C<OLD>. Formerly, it would have returned the data from the start
567 of the following disk block instead.
569 =head2 eof() has the same old magic as <>
571 C<eof()> would return true if no attempt to read from C<< <> >> had
572 yet been made. C<eof()> has been changed to have a little magic of its
573 own, it now opens the C<< <> >> files.
575 =head2 binmode() can be used to set :crlf and :raw modes
577 binmode() now accepts a second argument that specifies a discipline
578 for the handle in question. The two pseudo-disciplines ":raw" and
579 ":crlf" are currently supported on DOS-derivative platforms.
580 See L<perlfunc/"binmode"> and L<open>.
582 =head2 C<-T> filetest recognizes UTF-8 encoded files as "text"
584 The algorithm used for the C<-T> filetest has been enhanced to
585 correctly identify UTF-8 content as "text".
587 =head2 system(), backticks and pipe open now reflect exec() failure
589 On Unix and similar platforms, system(), qx() and open(FOO, "cmd |")
590 etc., are implemented via fork() and exec(). When the underlying
591 exec() fails, earlier versions did not report the error properly,
592 since the exec() happened to be in a different process.
594 The child process now communicates with the parent about the
595 error in launching the external command, which allows these
596 constructs to return with their usual error value and set $!.
598 =head2 Improved diagnostics
600 Line numbers are no longer suppressed (under most likely circumstances)
601 during the global destruction phase.
603 Diagnostics emitted from code running in threads other than the main
604 thread are now accompanied by the thread ID.
606 Embedded null characters in diagnostics now actually show up. They
607 used to truncate the message in prior versions.
609 $foo::a and $foo::b are now exempt from "possible typo" warnings only
610 if sort() is encountered in package C<foo>.
612 Unrecognized alphabetic escapes encountered when parsing quote
613 constructs now generate a warning, since they may take on new
614 semantics in later versions of Perl.
616 Many diagnostics now report the internal operation in which the warning
617 was provoked, like so:
619 Use of uninitialized value in concatenation (.) at (eval 1) line 1.
620 Use of uninitialized value in print at (eval 1) line 1.
622 Diagnostics that occur within eval may also report the file and line
623 number where the eval is located, in addition to the eval sequence
624 number and the line number within the evaluated text itself. For
627 Not enough arguments for scalar at (eval 4)[newlib/perl5db.pl:1411] line 2, at EOF
629 =head2 Diagnostics follow STDERR
631 Diagnostic output now goes to whichever file the C<STDERR> handle
632 is pointing at, instead of always going to the underlying C runtime
635 =item More consistent close-on-exec behavior
637 On systems that support a close-on-exec flag on filehandles, the
638 flag is now set for any handles created by pipe(), socketpair(),
639 socket(), and accept(), if that is warranted by the value of $^F
640 that may be in effect. Earlier versions neglected to set the flag
641 for handles created with these operators. See L<perlfunc/pipe>,
642 L<perlfunc/socketpair>, L<perlfunc/socket>, L<perlfunc/accept>,
645 =head2 syswrite() ease-of-use
647 The length argument of C<syswrite()> has become optional.
649 =head2 Better syntax checks on parenthesized unary operators
653 print defined(&foo,&bar,&baz);
654 print uc("foo","bar","baz");
657 used to be accidentally allowed in earlier versions, and produced
658 unpredictable behaviour. Some produced ancillary warnings
659 when used in this way; others silently did the wrong thing.
661 The parenthesized forms of most unary operators that expect a single
662 argument now ensure that they are not called with more than one
663 argument, making the cases shown above syntax errors. The usual
666 print defined &foo, &bar, &baz;
667 print uc "foo", "bar", "baz";
670 remains unchanged. See L<perlop>.
672 =head2 Bit operators support full native integer width
674 The bit operators (& | ^ ~ << >>) now operate on the full native
675 integral width (the exact size of which is available in $Config{ivsize}).
676 For example, if your platform is either natively 64-bit or if Perl
677 has been configured to use 64-bit integers, these operations apply
678 to 8 bytes (as opposed to 4 bytes on 32-bit platforms).
679 For portability, be sure to mask off the excess bits in the result of
680 unary C<~>, e.g., C<~$x & 0xffffffff>.
682 =head2 Improved security features
684 More potentially unsafe operations taint their results for improved
687 The C<passwd> and C<shell> fields returned by the getpwent(), getpwnam(),
688 and getpwuid() are now tainted, because the user can affect their own
689 encrypted password and login shell.
691 The variable modified by shmread(), and messages returned by msgrcv()
692 (and its object-oriented interface IPC::SysV::Msg::rcv) are also tainted,
693 because other untrusted processes can modify messages and shared memory
694 segments for their own nefarious purposes.
696 =item More functional bareword prototype (*)
698 Bareword prototypes have been rationalized to enable them to be used
699 to override builtins that accept barewords and interpret them in
700 a special way, such as C<require> or C<do>.
702 Arguments prototyped as C<*> will now be visible within the subroutine
703 as either a simple scalar or as a reference to a typeglob.
704 See L<perlsub/Prototypes>.
706 =head2 C<require> and C<do> may be overridden
708 C<require> and C<do 'file'> operations may be overridden locally
709 by importing subroutines of the same name into the current package
710 (or globally by importing them into the CORE::GLOBAL:: namespace).
711 Overriding C<require> will also affect C<use>, provided the override
712 is visible at compile-time.
713 See L<perlsub/"Overriding Built-in Functions">.
715 =head2 $^X variables may now have names longer than one character
717 Formerly, $^X was synonymous with ${"\cX"}, but $^XY was a syntax
718 error. Now variable names that begin with a control character may be
719 arbitrarily long. However, for compatibility reasons, these variables
720 I<must> be written with explicit braces, as C<${^XY}> for example.
721 C<${^XYZ}> is synonymous with ${"\cXYZ"}. Variable names with more
722 than one control character, such as C<${^XY^Z}>, are illegal.
724 The old syntax has not changed. As before, `^X' may be either a
725 literal control-X character or the two-character sequence `caret' plus
726 `X'. When braces are omitted, the variable name stops after the
727 control character. Thus C<"$^XYZ"> continues to be synonymous with
728 C<$^X . "YZ"> as before.
730 As before, lexical variables may not have names beginning with control
731 characters. As before, variables whose names begin with a control
732 character are always forced to be in package `main'. All such variables
733 are reserved for future extensions, except those that begin with
734 C<^_>, which may be used by user programs and are guaranteed not to
735 acquire special meaning in any future version of Perl.
737 =head2 New variable $^C reflects C<-c> switch
739 C<$^C> has a boolean value that reflects whether perl is being run
740 in compile-only mode (i.e. via the C<-c> switch). Since
741 BEGIN blocks are executed under such conditions, this variable
742 enables perl code to determine whether actions that make sense
743 only during normal running are warranted. See L<perlvar>.
745 =head2 New variable $^V contains Perl version as a string
747 C<$^V> contains the Perl version number as a string composed of
748 characters whose ordinals match the version numbers, i.e. v5.6.0.
749 This may be used in string comparisons.
751 See C<Support for strings represented as a vector of ordinals> for an
754 =head2 Optional Y2K warnings
756 If Perl is built with the cpp macro C<PERL_Y2KWARN> defined,
757 it emits optional warnings when concatenating the number 19
760 This behavior must be specifically enabled when running Configure.
761 See F<INSTALL> and F<README.Y2K>.
763 =head1 Modules and Pragmata
771 While used internally by Perl as a pragma, this module also
772 provides a way to fetch subroutine and variable attributes.
777 The Perl Compiler suite has been extensively reworked for this
778 release. More of the standard Perl testsuite passes when run
779 under the Compiler, but there is still a significant way to
780 go to achieve production quality compiled executables.
782 NOTE: The Compiler suite remains highly experimental. The
783 generated code may not be correct, even when it manages to execute
788 Overall, Benchmark results exhibit lower average error and better timing
791 You can now run tests for I<n> seconds instead of guessing the right
792 number of tests to run: e.g., timethese(-5, ...) will run each
793 code for at least 5 CPU seconds. Zero as the "number of repetitions"
794 means "for at least 3 CPU seconds". The output format has also
795 changed. For example:
797 use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
799 will now output something like this:
801 Benchmark: running a, b, each for at least 5 CPU seconds...
802 a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
803 b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
805 New features: "each for at least N CPU seconds...", "wallclock secs",
806 and the "@ operations/CPU second (n=operations)".
808 timethese() now returns a reference to a hash of Benchmark objects containing
809 the test results, keyed on the names of the tests.
811 timethis() now returns the iterations field in the Benchmark result object
814 timethese(), timethis(), and the new cmpthese() (see below) can also take
815 a format specifier of 'none' to suppress output.
817 A new function countit() is just like timeit() except that it takes a
818 TIME instead of a COUNT.
820 A new function cmpthese() prints a chart comparing the results of each test
821 returned from a timethese() call. For each possible pair of tests, the
822 percentage speed difference (iters/sec or seconds/iter) is shown.
824 For other details, see L<Benchmark>.
828 The ByteLoader is a dedicated extension to generate and run
829 Perl bytecode. See L<ByteLoader>.
833 References can now be used.
835 The new version also allows a leading underscore in constant names, but
836 disallows a double leading underscore (as in "__LINE__"). Some other names
837 are disallowed or warned against, including BEGIN, END, etc. Some names
838 which were forced into main:: used to fail silently in some cases; now they're
839 fatal (outside of main::) and an optional warning (inside of main::).
840 The ability to detect whether a constant had been set with a given name has
847 This pragma implements the C<\N> string escape. See L<charnames>.
851 A C<Maxdepth> setting can be specified to avoid venturing
852 too deeply into deep data structures. See L<Data::Dumper>.
854 The XSUB implementation of Dump() is now automatically called if the
855 C<Useqq> setting is not in use.
857 Dumping C<qr//> objects works correctly.
861 C<DB> is an experimental module that exposes a clean abstraction
862 to Perl's debugging API.
866 DB_File can now be built with Berkeley DB versions 1, 2 or 3.
867 See C<ext/DB_File/Changes>.
871 Devel::DProf, a Perl source code profiler has been added. See
872 L<Devel::DProf> and L<dprofpp>.
876 The Devel::Peek module provides access to the internal representation
877 of Perl variables and data. It is a data debugging tool for the XS programmer.
881 The Dumpvalue module provides screen dumps of Perl data.
885 DynaLoader now supports a dl_unload_file() function on platforms that
886 support unloading shared objects using dlclose().
888 Perl can also optionally arrange to unload all extension shared objects
889 loaded by Perl. To enable this, build Perl with the Configure option
890 C<-Accflags=-DDL_UNLOAD_ALL_AT_EXIT>. (This maybe useful if you are
891 using Apache with mod_perl.)
895 $PERL_VERSION now stands for C<$^V> (a string value) rather than for C<$]>
900 Env now supports accessing environment variables like PATH as array
905 More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
906 large file (more than 4GB) access (NOTE: the O_LARGEFILE is
907 automatically added to sysopen() flags if large file support has been
908 configured, as is the default), Free/Net/OpenBSD locking behaviour
909 flags F_FLOCK, F_POSIX, Linux F_SHLCK, and O_ACCMODE: the combined
910 mask of O_RDONLY, O_WRONLY, and O_RDWR. The seek()/sysseek()
911 constants SEEK_SET, SEEK_CUR, and SEEK_END are available via the
912 C<:seek> tag. The chmod()/stat() S_IF* constants and S_IS* functions
913 are available via the C<:mode> tag.
917 A compare_text() function has been added, which allows custom
918 comparison functions. See L<File::Compare>.
922 File::Find now works correctly when the wanted() function is either
923 autoloaded or is a symbolic reference.
925 A bug that caused File::Find to lose track of the working directory
926 when pruning top-level directories has been fixed.
928 File::Find now also supports several other options to control its
929 behavior. It can follow symbolic links if the C<follow> option is
930 specified. Enabling the C<no_chdir> option will make File::Find skip
931 changing the current directory when walking directories. The C<untaint>
932 flag can be useful when running with taint checks enabled.
938 This extension implements BSD-style file globbing. By default,
939 it will also be used for the internal implementation of the glob()
940 operator. See L<File::Glob>.
944 New methods have been added to the File::Spec module: devnull() returns
945 the name of the null device (/dev/null on Unix) and tmpdir() the name of
946 the temp directory (normally /tmp on Unix). There are now also methods
947 to convert between absolute and relative filenames: abs2rel() and
948 rel2abs(). For compatibility with operating systems that specify volume
949 names in file paths, the splitpath(), splitdir(), and catdir() methods
952 =item File::Spec::Functions
954 The new File::Spec::Functions modules provides a function interface
955 to the File::Spec module. Allows shorthand
957 $fullname = catfile($dir1, $dir2, $file);
961 $fullname = File::Spec->catfile($dir1, $dir2, $file);
965 Getopt::Long licensing has changed to allow the Perl Artistic License
966 as well as the GPL. It used to be GPL only, which got in the way of
967 non-GPL applications that wanted to use Getopt::Long.
969 Getopt::Long encourages the use of Pod::Usage to produce help
970 messages. For example:
976 GetOptions('help|?' => \$help, man => \$man) or pod2usage(2);
977 pod2usage(1) if $help;
978 pod2usage(-exitstatus => 0, -verbose => 2) if $man;
984 sample - Using GetOpt::Long and Pod::Usage
988 sample [options] [file ...]
991 -help brief help message
992 -man full documentation
1000 Print a brief help message and exits.
1004 Prints the manual page and exits.
1010 B<This program> will read the given input file(s) and do something
1011 useful with the contents thereof.
1015 See L<Pod::Usage> for details.
1017 A bug that prevented the non-option call-back <> from being
1018 specified as the first argument has been fixed.
1020 To specify the characters < and > as option starters, use ><. Note,
1021 however, that changing option starters is strongly deprecated.
1025 write() and syswrite() will now accept a single-argument
1026 form of the call, for consistency with Perl's syswrite().
1028 You can now create a TCP-based IO::Socket::INET without forcing
1029 a connect attempt. This allows you to configure its options
1030 (like making it non-blocking) and then call connect() manually.
1032 A bug that prevented the IO::Socket::protocol() accessor
1033 from ever returning the correct value has been corrected.
1035 IO::Socket::connect now uses non-blocking IO instead of alarm()
1036 to do connect timeouts.
1038 IO::Socket::accept now uses select() instead of alarm() for doing
1041 IO::Socket::INET->new now sets $! correctly on failure. $@ is
1042 still set for backwards compatibility.
1046 Java Perl Lingo is now distributed with Perl. See jpl/README
1047 for more information.
1051 C<use lib> now weeds out any trailing duplicate entries.
1052 C<no lib> removes all named entries.
1056 The bitwise operations C<<< << >>>, C<<< >> >>>, C<&>, C<|>,
1057 and C<~> are now supported on bigints.
1061 The accessor methods Re, Im, arg, abs, rho, and theta can now also
1062 act as mutators (accessor $z->Re(), mutator $z->Re(3)).
1064 The class method C<display_format> and the corresponding object method
1065 C<display_format>, in addition to accepting just one argument, now can
1066 also accept a parameter hash. Recognized keys of a parameter hash are
1067 C<"style">, which corresponds to the old one parameter case, and two
1068 new parameters: C<"format">, which is a printf()-style format string
1069 (defaults usually to C<"%.15g">, you can revert to the default by
1070 setting the format string to C<undef>) used for both parts of a
1071 complex number, and C<"polar_pretty_print"> (defaults to true),
1072 which controls whether an attempt is made to try to recognize small
1073 multiples and rationals of pi (2pi, pi/2) at the argument (angle) of a
1074 polar complex number.
1076 The potentially disruptive change is that in list context both methods
1077 now I<return the parameter hash>, instead of only the value of the
1078 C<"style"> parameter.
1082 A little bit of radial trigonometry (cylindrical and spherical),
1083 radial coordinate conversions, and the great circle distance were added.
1085 =item Pod::Parser, Pod::InputObjects
1087 Pod::Parser is a base class for parsing and selecting sections of
1088 pod documentation from an input stream. This module takes care of
1089 identifying pod paragraphs and commands in the input and hands off the
1090 parsed paragraphs and commands to user-defined methods which are free
1091 to interpret or translate them as they see fit.
1093 Pod::InputObjects defines some input objects needed by Pod::Parser, and
1094 for advanced users of Pod::Parser that need more about a command besides
1097 As of release 5.6.0 of Perl, Pod::Parser is now the officially sanctioned
1098 "base parser code" recommended for use by all pod2xxx translators.
1099 Pod::Text (pod2text) and Pod::Man (pod2man) have already been converted
1100 to use Pod::Parser and efforts to convert Pod::HTML (pod2html) are already
1101 underway. For any questions or comments about pod parsing and translating
1102 issues and utilities, please use the pod-people@perl.org mailing list.
1104 For further information, please see L<Pod::Parser> and L<Pod::InputObjects>.
1106 =item Pod::Checker, podchecker
1108 This utility checks pod files for correct syntax, according to
1109 L<perlpod>. Obvious errors are flagged as such, while warnings are
1110 printed for mistakes that can be handled gracefully. The checklist is
1111 not complete yet. See L<Pod::Checker>.
1113 =item Pod::ParseUtils, Pod::Find
1115 These modules provide a set of gizmos that are useful mainly for pod
1116 translators. L<Pod::Find|Pod::Find> traverses directory structures and
1117 returns found pod files, along with their canonical names (like
1118 C<File::Spec::Unix>). L<Pod::ParseUtils|Pod::ParseUtils> contains
1119 B<Pod::List> (useful for storing pod list information), B<Pod::Hyperlink>
1120 (for parsing the contents of C<LE<lt>E<gt>> sequences) and B<Pod::Cache>
1121 (for caching information about pod files, e.g., link nodes).
1123 =item Pod::Select, podselect
1125 Pod::Select is a subclass of Pod::Parser which provides a function
1126 named "podselect()" to filter out user-specified sections of raw pod
1127 documentation from an input stream. podselect is a script that provides
1128 access to Pod::Select from other scripts to be used as a filter.
1131 =item Pod::Usage, pod2usage
1133 Pod::Usage provides the function "pod2usage()" to print usage messages for
1134 a Perl script based on its embedded pod documentation. The pod2usage()
1135 function is generally useful to all script authors since it lets them
1136 write and maintain a single source (the pods) for documentation, thus
1137 removing the need to create and maintain redundant usage message text
1138 consisting of information already in the pods.
1140 There is also a pod2usage script which can be used from other kinds of
1141 scripts to print usage messages from pods (even for non-Perl scripts
1142 with pods embedded in comments).
1144 For details and examples, please see L<Pod::Usage>.
1146 =item Pod::Text and Pod::Man
1148 Pod::Text has been rewritten to use Pod::Parser. While pod2text() is
1149 still available for backwards compatibility, the module now has a new
1150 preferred interface. See L<Pod::Text> for the details. The new Pod::Text
1151 module is easily subclassed for tweaks to the output, and two such
1152 subclasses (Pod::Text::Termcap for man-page-style bold and underlining
1153 using termcap information, and Pod::Text::Color for markup with ANSI color
1154 sequences) are now standard.
1156 pod2man has been turned into a module, Pod::Man, which also uses
1157 Pod::Parser. In the process, several outstanding bugs related to quotes
1158 in section headers, quoting of code escapes, and nested lists have been
1159 fixed. pod2man is now a wrapper script around this module.
1163 An EXISTS method has been added to this module (and sdbm_exists() has
1164 been added to the underlying sdbm library), so one can now call exists
1165 on an SDBM_File tied hash and get the correct result, rather than a
1168 A bug that may have caused data loss when more than one disk block
1169 happens to be read from the database in a single FETCH() has been
1174 Sys::Syslog now uses XSUBs to access facilities from syslog.h so it
1175 no longer requires syslog.ph to exist.
1179 Sys::Hostname now uses XSUBs to call the C library's gethostname() or
1180 uname() if they exist.
1182 =item Term::ANSIColor
1184 Term::ANSIColor is a very simple module to provide easy and readable
1185 access to the ANSI color and highlighting escape sequences, supported by
1186 most ANSI terminal emulators. It is now included standard.
1190 The timelocal() and timegm() functions used to silently return bogus
1191 results when the date fell outside the machine's integer range. They
1192 now consistently croak() if the date falls in an unsupported range.
1196 The error return value in list context has been changed for all functions
1197 that return a list of values. Previously these functions returned a list
1198 with a single element C<undef> if an error occurred. Now these functions
1199 return the empty list in these situations. This applies to the following
1205 The remaining functions are unchanged and continue to return C<undef> on
1206 error even in list context.
1208 The Win32::SetLastError(ERROR) function has been added as a complement
1209 to the Win32::GetLastError() function.
1211 The new Win32::GetFullPathName(FILENAME) returns the full absolute
1212 pathname for FILENAME in scalar context. In list context it returns
1213 a two-element list containing the fully qualified directory name and
1214 the filename. See L<Win32>.
1218 The XSLoader extension is a simpler alternative to DynaLoader.
1223 A new feature called "DBM Filters" has been added to all the
1224 DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
1225 DBM Filters add four new methods to each DBM module:
1232 These can be used to filter key-value pairs before the pairs are
1233 written to the database or just after they are read from the database.
1234 See L<perldbmfilter> for further information.
1240 C<use attrs> is now obsolete, and is only provided for
1241 backward-compatibility. It's been replaced by the C<sub : attributes>
1242 syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>.
1244 Lexical warnings pragma, C<use warnings;>, to control optional warnings.
1247 C<use filetest> to control the behaviour of filetests (C<-r> C<-w>
1248 ...). Currently only one subpragma implemented, "use filetest
1249 'access';", that uses access(2) or equivalent to check permissions
1250 instead of using stat(2) as usual. This matters in filesystems
1251 where there are ACLs (access control lists): the stat(2) might lie,
1252 but access(2) knows better.
1254 The C<open> pragma can be used to specify default disciplines for
1255 handle constructors (e.g. open()) and for qx//. The two
1256 pseudo-disciplines C<:raw> and C<:crlf> are currently supported on
1257 DOS-derivative platforms (i.e. where binmode is not a no-op).
1258 See also L</"binmode() can be used to set :crlf and :raw modes">.
1260 =head1 Utility Changes
1264 C<dprofpp> is used to display profile data generated using C<Devel::DProf>.
1269 The C<find2perl> utility now uses the enhanced features of the File::Find
1270 module. The -depth and -follow options are supported. Pod documentation
1271 is also included in the script.
1275 The C<h2xs> tool can now work in conjunction with C<C::Scan> (available
1276 from CPAN) to automatically parse real-life header files. The C<-M>,
1277 C<-a>, C<-k>, and C<-o> options are new.
1281 C<perlcc> now supports the C and Bytecode backends. By default,
1282 it generates output from the simple C backend rather than the
1283 optimized C backend.
1285 Support for non-Unix platforms has been improved.
1289 C<perldoc> has been reworked to avoid possible security holes.
1290 It will not by default let itself be run as the superuser, but you
1291 may still use the B<-U> switch to try to make it drop privileges
1294 =head2 The Perl Debugger
1296 Many bug fixes and enhancements were added to F<perl5db.pl>, the
1297 Perl debugger. The help documentation was rearranged. New commands
1298 include C<< < ? >>, C<< > ? >>, and C<< { ? >> to list out current
1299 actions, C<man I<docpage>> to run your doc viewer on some perl
1300 docset, and support for quoted options. The help information was
1301 rearranged, and should be viewable once again if you're using B<less>
1302 as your pager. A serious security hole was plugged--you should
1303 immediately remove all older versions of the Perl debugger as
1304 installed in previous releases, all the way back to perl3, from
1305 your system to avoid being bitten by this.
1307 =head1 Improved Documentation
1309 Many of the platform-specific README files are now part of the perl
1310 installation. See L<perl> for the complete list.
1316 The official list of public Perl API functions.
1320 A tutorial for beginners on object-oriented Perl.
1322 =item perlcompile.pod
1324 An introduction to using the Perl Compiler suite.
1326 =item perldbmfilter.pod
1328 A howto document on using the DBM filter facility.
1332 All material unrelated to running the Perl debugger, plus all
1333 low-level guts-like details that risked crushing the casual user
1334 of the debugger, have been relocated from the old manpage to the
1337 =item perldebguts.pod
1339 This new manpage contains excessively low-level material not related
1340 to the Perl debugger, but slightly related to debugging Perl itself.
1341 It also contains some arcane internal details of how the debugging
1342 process works that may only be of interest to developers of Perl
1347 Notes on the fork() emulation currently available for the Windows platform.
1349 =item perlfilter.pod
1351 An introduction to writing Perl source filters.
1355 Some guidelines for hacking the Perl source code.
1357 =item perlintern.pod
1359 A list of internal functions in the Perl source code.
1360 (List is currently empty.)
1362 =item perllexwarn.pod
1364 Introduction and reference information about lexically scoped
1367 =item perlnumber.pod
1369 Detailed information about numbers as they are represented in Perl.
1371 =item perlopentut.pod
1373 A tutorial on using open() effectively.
1375 =item perlreftut.pod
1377 A tutorial that introduces the essentials of references.
1381 A tutorial on managing class data for object modules.
1385 Discussion of the most often wanted features that may someday be
1388 =item perlunicode.pod
1390 An introduction to Unicode support features in Perl.
1394 =head1 Performance enhancements
1396 =head2 Simple sort() using { $a <=> $b } and the like are optimized
1398 Many common sort() operations using a simple inlined block are now
1399 optimized for faster performance.
1401 =head2 Optimized assignments to lexical variables
1403 Certain operations in the RHS of assignment statements have been
1404 optimized to directly set the lexical variable on the LHS,
1405 eliminating redundant copying overheads.
1407 =head2 Faster subroutine calls
1409 Minor changes in how subroutine calls are handled internally
1410 provide marginal improvements in performance.
1412 =item delete(), each(), values() and hash iteration are faster
1414 The hash values returned by delete(), each(), values() and hashes in a
1415 list context are the actual values in the hash, instead of copies.
1416 This results in significantly better performance, because it eliminates
1417 needless copying in most situations.
1419 =head1 Installation and Configuration Improvements
1421 =head2 -Dusethreads means something different
1423 The -Dusethreads flag now enables the experimental interpreter-based thread
1424 support by default. To get the flavor of experimental threads that was in
1425 5.005 instead, you need to run Configure with "-Dusethreads -Duse5005threads".
1427 As of v5.6.0, interpreter-threads support is still lacking a way to
1428 create new threads from Perl (i.e., C<use Thread;> will not work with
1429 interpreter threads). C<use Thread;> continues to be available when you
1430 specify the -Duse5005threads option to Configure, bugs and all.
1432 NOTE: Support for threads continues to be an experimental feature.
1433 Interfaces and implementation are subject to sudden and drastic changes.
1435 =head2 New Configure flags
1437 The following new flags may be enabled on the Configure command line
1438 by running Configure with C<-Dflag>.
1441 usethreads useithreads (new interpreter threads: no Perl API yet)
1442 usethreads use5005threads (threads as they were in 5.005)
1444 use64bitint (equal to now deprecated 'use64bits')
1450 usesocks (only SOCKS v5 supported)
1452 =head2 Threadedness and 64-bitness now more daring
1454 The Configure options enabling the use of threads and the use of
1455 64-bitness are now more daring in the sense that they no more have an
1456 explicit list of operating systems of known threads/64-bit
1457 capabilities. In other words: if your operating system has the
1458 necessary APIs and datatypes, you should be able just to go ahead and
1459 use them, for threads by Configure -Dusethreads, and for 64 bits
1460 either explicitly by Configure -Duse64bitint or implicitly if your
1461 system has 64-bit wide datatypes. See also L<"64-bit support">.
1465 Some platforms have "long doubles", floating point numbers of even
1466 larger range than ordinary "doubles". To enable using long doubles for
1467 Perl's scalars, use -Duselongdouble.
1469 =head2 -Dusemorebits
1471 You can enable both -Duse64bitint and -Duselongdouble with -Dusemorebits.
1472 See also L<"64-bit support">.
1474 =head2 -Duselargefiles
1476 Some platforms support system APIs that are capable of handling large files
1477 (typically, files larger than two gigabytes). Perl will try to use these
1478 APIs if you ask for -Duselargefiles.
1480 See L<"Large file support"> for more information.
1482 =head2 installusrbinperl
1484 You can use "Configure -Uinstallusrbinperl" which causes installperl
1485 to skip installing perl also as /usr/bin/perl. This is useful if you
1486 prefer not to modify /usr/bin for some reason or another but harmful
1487 because many scripts assume to find Perl in /usr/bin/perl.
1489 =head2 SOCKS support
1491 You can use "Configure -Dusesocks" which causes Perl to probe
1492 for the SOCKS proxy protocol library (v5, not v4). For more information
1495 http://www.socks.nec.com/
1499 You can "post-edit" the Configure variables using the Configure C<-A>
1500 switch. The editing happens immediately after the platform specific
1501 hints files have been processed but before the actual configuration
1502 process starts. Run C<Configure -h> to find out the full C<-A> syntax.
1504 =head2 Enhanced Installation Directories
1506 The installation structure has been enriched to improve the support
1507 for maintaining multiple versions of perl, to provide locations for
1508 vendor-supplied modules, scripts, and manpages, and to ease maintenance
1509 of locally-added modules, scripts, and manpages. See the section on
1510 Installation Directories in the INSTALL file for complete details.
1511 For most users building and installing from source, the defaults should
1514 If you previously used C<Configure -Dsitelib> or C<-Dsitearch> to set
1515 special values for library directories, you might wish to consider using
1516 the new C<-Dsiteprefix> setting instead. Also, if you wish to re-use a
1517 config.sh file from an earlier version of perl, you should be sure to
1518 check that Configure makes sensible choices for the new directories.
1519 See INSTALL for complete details.
1521 =head1 Platform specific changes
1523 =head2 Supported platforms
1529 The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
1534 GNU/Hurd is now supported.
1538 Rhapsody/Darwin is now supported.
1542 EPOC is now supported (on Psion 5).
1546 The cygwin port (formerly cygwin32) has been greatly improved.
1556 Perl now works with djgpp 2.02 (and 2.03 alpha).
1560 Environment variable names are not converted to uppercase any more.
1564 Incorrect exit codes from backticks have been fixed.
1568 This port continues to use its own builtin globbing (not File::Glob).
1572 =head2 OS390 (OpenEdition MVS)
1574 Support for this EBCDIC platform has not been renewed in this release.
1575 There are difficulties in reconciling Perl's standardization on UTF-8
1576 as its internal representation for characters with the EBCDIC character
1577 set, because the two are incompatible.
1579 It is unclear whether future versions will renew support for this
1580 platform, but the possibility exists.
1584 Numerous revisions and extensions to configuration, build, testing, and
1585 installation process to accommodate core changes and VMS-specific options.
1587 Expand %ENV-handling code to allow runtime mapping to logical names,
1588 CLI symbols, and CRTL environ array.
1590 Extension of subprocess invocation code to accept filespecs as command
1593 Add to Perl command line processing the ability to use default file types and
1594 to recognize Unix-style C<2E<gt>&1>.
1596 Expansion of File::Spec::VMS routines, and integration into ExtUtils::MM_VMS.
1598 Extension of ExtUtils::MM_VMS to handle complex extensions more flexibly.
1600 Barewords at start of Unix-syntax paths may be treated as text rather than
1601 only as logical names.
1603 Optional secure translation of several logical names used internally by Perl.
1605 Miscellaneous bugfixing and porting of new core code to VMS.
1607 Thanks are gladly extended to the many people who have contributed VMS
1608 patches, testing, and ideas.
1612 Perl can now emulate fork() internally, using multiple interpreters running
1613 in different concurrent threads. This support must be enabled at build
1614 time. See L<perlfork> for detailed information.
1616 When given a pathname that consists only of a drivename, such as C<A:>,
1617 opendir() and stat() now use the current working directory for the drive
1618 rather than the drive root.
1620 The builtin XSUB functions in the Win32:: namespace are documented. See
1623 $^X now contains the full path name of the running executable.
1625 A Win32::GetLongPathName() function is provided to complement
1626 Win32::GetFullPathName() and Win32::GetShortPathName(). See L<Win32>.
1628 POSIX::uname() is supported.
1630 system(1,...) now returns true process IDs rather than process
1631 handles. kill() accepts any real process id, rather than strictly
1632 return values from system(1,...).
1634 For better compatibility with Unix, C<kill(0, $pid)> can now be used to
1635 test whether a process exists.
1637 The C<Shell> module is supported.
1639 Better support for building Perl under command.com in Windows 95
1642 Scripts are read in binary mode by default to allow ByteLoader (and
1643 the filter mechanism in general) to work properly. For compatibility,
1644 the DATA filehandle will be set to text mode if a carriage return is
1645 detected at the end of the line containing the __END__ or __DATA__
1646 token; if not, the DATA filehandle will be left open in binary mode.
1647 Earlier versions always opened the DATA filehandle in text mode.
1649 The glob() operator is implemented via the C<File::Glob> extension,
1650 which supports glob syntax of the C shell. This increases the flexibility
1651 of the glob() operator, but there may be compatibility issues for
1652 programs that relied on the older globbing syntax. If you want to
1653 preserve compatibility with the older syntax, you might want to run
1654 perl with C<-MFile::DosGlob>. For details and compatibility information,
1657 =head1 Significant bug fixes
1659 =head2 <HANDLE> on empty files
1661 With C<$/> set to C<undef>, "slurping" an empty file returns a string of
1662 zero length (instead of C<undef>, as it used to) the first time the
1663 HANDLE is read after C<$/> is set to C<undef>. Further reads yield
1666 This means that the following will append "foo" to an empty file (it used
1669 perl -0777 -pi -e 's/^/foo/' empty_file
1673 perl -pi -e 's/^/foo/' empty_file
1675 is unchanged (it continues to leave the file empty).
1677 =head2 C<eval '...'> improvements
1679 Line numbers (as reflected by caller() and most diagnostics) within
1680 C<eval '...'> were often incorrect where here documents were involved.
1681 This has been corrected.
1683 Lexical lookups for variables appearing in C<eval '...'> within
1684 functions that were themselves called within an C<eval '...'> were
1685 searching the wrong place for lexicals. The lexical search now
1686 correctly ends at the subroutine's block boundary.
1688 The use of C<return> within C<eval {...}> caused $@ not to be reset
1689 correctly when no exception occurred within the eval. This has
1692 Parsing of here documents used to be flawed when they appeared as
1693 the replacement expression in C<eval 's/.../.../e'>. This has
1696 =head2 All compilation errors are true errors
1698 Some "errors" encountered at compile time were by necessity
1699 generated as warnings followed by eventual termination of the
1700 program. This enabled more such errors to be reported in a
1701 single run, rather than causing a hard stop at the first error
1702 that was encountered.
1704 The mechanism for reporting such errors has been reimplemented
1705 to queue compile-time errors and report them at the end of the
1706 compilation as true errors rather than as warnings. This fixes
1707 cases where error messages leaked through in the form of warnings
1708 when code was compiled at run time using C<eval STRING>, and
1709 also allows such errors to be reliably trapped using C<eval "...">.
1711 =head2 Implicitly closed filehandles are safer
1713 Sometimes implicitly closed filehandles (as when they are localized,
1714 and Perl automatically closes them on exiting the scope) could
1715 inadvertently set $? or $!. This has been corrected.
1718 =head2 Behavior of list slices is more consistent
1720 When taking a slice of a literal list (as opposed to a slice of
1721 an array or hash), Perl used to return an empty list if the
1722 result happened to be composed of all undef values.
1724 The new behavior is to produce an empty list if (and only if)
1725 the original list was empty. Consider the following example:
1727 @a = (1,undef,undef,2)[2,1,2];
1729 The old behavior would have resulted in @a having no elements.
1730 The new behavior ensures it has three undefined elements.
1732 Note in particular that the behavior of slices of the following
1733 cases remains unchanged:
1736 @a = (getpwent)[7,0];
1737 @a = (anything_returning_empty_list())[2,1,2];
1739 @a = @c{'a','b','c'};
1743 =head2 C<(\$)> prototype and C<$foo{a}>
1745 A scalar reference prototype now correctly allows a hash or
1746 array element in that slot.
1748 =head2 C<goto &sub> and AUTOLOAD
1750 The C<goto &sub> construct works correctly when C<&sub> happens
1753 =head2 C<-bareword> allowed under C<use integer>
1755 The autoquoting of barewords preceded by C<-> did not work
1756 in prior versions when the C<integer> pragma was enabled.
1757 This has been fixed.
1759 =head2 Failures in DESTROY()
1761 When code in a destructor threw an exception, it went unnoticed
1762 in earlier versions of Perl, unless someone happened to be
1763 looking in $@ just after the point the destructor happened to
1764 run. Such failures are now visible as warnings when warnings are
1767 =head2 Locale bugs fixed
1769 printf() and sprintf() previously reset the numeric locale
1770 back to the default "C" locale. This has been fixed.
1772 Numbers formatted according to the local numeric locale
1773 (such as using a decimal comma instead of a decimal dot) caused
1774 "isn't numeric" warnings, even while the operations accessing
1775 those numbers produced correct results. These warnings have been
1780 The C<eval 'return sub {...}'> construct could sometimes leak
1781 memory. This has been fixed.
1783 Operations that aren't filehandle constructors used to leak memory
1784 when used on invalid filehandles. This has been fixed.
1786 Constructs that modified C<@_> could fail to deallocate values
1787 in C<@_> and thus leak memory. This has been corrected.
1789 =head2 Spurious subroutine stubs after failed subroutine calls
1791 Perl could sometimes create empty subroutine stubs when a
1792 subroutine was not found in the package. Such cases stopped
1793 later method lookups from progressing into base packages.
1794 This has been corrected.
1796 =head2 Taint failures under C<-U>
1798 When running in unsafe mode, taint violations could sometimes
1799 cause silent failures. This has been fixed.
1801 =head2 END blocks and the C<-c> switch
1803 Prior versions used to run BEGIN B<and> END blocks when Perl was
1804 run in compile-only mode. Since this is typically not the expected
1805 behavior, END blocks are not executed anymore when the C<-c> switch
1806 is used, or if compilation fails.
1808 See L<CHECK blocks> for how to run things when the compile phase ends.
1810 =head2 Potential to leak DATA filehandles
1812 Using the C<__DATA__> token creates an implicit filehandle to
1813 the file that contains the token. It is the program's
1814 responsibility to close it when it is done reading from it.
1816 This caveat is now better explained in the documentation.
1819 =head1 New or Changed Diagnostics
1823 =item "%s" variable %s masks earlier declaration in same %s
1825 (W misc) A "my" or "our" variable has been redeclared in the current scope or statement,
1826 effectively eliminating all access to the previous instance. This is almost
1827 always a typographical error. Note that the earlier variable will still exist
1828 until the end of the scope or until all closure referents to it are
1831 =item "my sub" not yet implemented
1833 (F) Lexically scoped subroutines are not yet implemented. Don't try that
1836 =item "our" variable %s redeclared
1838 (W misc) You seem to have already declared the same global once before in the
1839 current lexical scope.
1841 =item '!' allowed only after types %s
1843 (F) The '!' is allowed in pack() and unpack() only after certain types.
1844 See L<perlfunc/pack>.
1846 =item / cannot take a count
1848 (F) You had an unpack template indicating a counted-length string,
1849 but you have also specified an explicit size for the string.
1850 See L<perlfunc/pack>.
1852 =item / must be followed by a, A or Z
1854 (F) You had an unpack template indicating a counted-length string,
1855 which must be followed by one of the letters a, A or Z
1856 to indicate what sort of string is to be unpacked.
1857 See L<perlfunc/pack>.
1859 =item / must be followed by a*, A* or Z*
1861 (F) You had a pack template indicating a counted-length string,
1862 Currently the only things that can have their length counted are a*, A* or Z*.
1863 See L<perlfunc/pack>.
1865 =item / must follow a numeric type
1867 (F) You had an unpack template that contained a '#',
1868 but this did not follow some numeric unpack specification.
1869 See L<perlfunc/pack>.
1871 =item /%s/: Unrecognized escape \\%c passed through
1873 (W regexp) You used a backslash-character combination which is not recognized
1874 by Perl. This combination appears in an interpolated variable or a
1875 C<'>-delimited regular expression. The character was understood literally.
1877 =item /%s/: Unrecognized escape \\%c in character class passed through
1879 (W regexp) You used a backslash-character combination which is not recognized
1880 by Perl inside character classes. The character was understood literally.
1882 =item /%s/ should probably be written as "%s"
1884 (W syntax) You have used a pattern where Perl expected to find a string,
1885 as in the first argument to C<join>. Perl will treat the true
1886 or false result of matching the pattern against $_ as the string,
1887 which is probably not what you had in mind.
1889 =item %s() called too early to check prototype
1891 (W prototype) You've called a function that has a prototype before the parser saw a
1892 definition or declaration for it, and Perl could not check that the call
1893 conforms to the prototype. You need to either add an early prototype
1894 declaration for the subroutine in question, or move the subroutine
1895 definition ahead of the call to get proper prototype checking. Alternatively,
1896 if you are certain that you're calling the function correctly, you may put
1897 an ampersand before the name to avoid the warning. See L<perlsub>.
1899 =item %s argument is not a HASH or ARRAY element
1901 (F) The argument to exists() must be a hash or array element, such as:
1906 =item %s argument is not a HASH or ARRAY element or slice
1908 (F) The argument to delete() must be either a hash or array element, such as:
1913 or a hash or array slice, such as:
1915 @foo[$bar, $baz, $xyzzy]
1916 @{$ref->[12]}{"susie", "queue"}
1918 =item %s argument is not a subroutine name
1920 (F) The argument to exists() for C<exists &sub> must be a subroutine
1921 name, and not a subroutine call. C<exists &sub()> will generate this error.
1923 =item %s package attribute may clash with future reserved word: %s
1925 (W reserved) A lowercase attribute name was used that had a package-specific handler.
1926 That name might have a meaning to Perl itself some day, even though it
1927 doesn't yet. Perhaps you should use a mixed-case attribute name, instead.
1930 =item (in cleanup) %s
1932 (W misc) This prefix usually indicates that a DESTROY() method raised
1933 the indicated exception. Since destructors are usually called by
1934 the system at arbitrary points during execution, and often a vast
1935 number of times, the warning is issued only once for any number
1936 of failures that would otherwise result in the same message being
1939 Failure of user callbacks dispatched using the C<G_KEEPERR> flag
1940 could also result in this warning. See L<perlcall/G_KEEPERR>.
1942 =item <> should be quotes
1944 (F) You wrote C<< require <file> >> when you should have written
1947 =item Attempt to join self
1949 (F) You tried to join a thread from within itself, which is an
1950 impossible task. You may be joining the wrong thread, or you may
1951 need to move the join() to some other thread.
1953 =item Bad evalled substitution pattern
1955 (F) You've used the /e switch to evaluate the replacement for a
1956 substitution, but perl found a syntax error in the code to evaluate,
1957 most likely an unexpected right brace '}'.
1959 =item Bad realloc() ignored
1961 (S) An internal routine called realloc() on something that had never been
1962 malloc()ed in the first place. Mandatory, but can be disabled by
1963 setting environment variable C<PERL_BADFREE> to 1.
1965 =item Bareword found in conditional
1967 (W bareword) The compiler found a bareword where it expected a conditional,
1968 which often indicates that an || or && was parsed as part of the
1969 last argument of the previous construct, for example:
1973 It may also indicate a misspelled constant that has been interpreted
1976 use constant TYPO => 1;
1977 if (TYOP) { print "foo" }
1979 The C<strict> pragma is useful in avoiding such errors.
1981 =item Binary number > 0b11111111111111111111111111111111 non-portable
1983 (W portable) The binary number you specified is larger than 2**32-1
1984 (4294967295) and therefore non-portable between systems. See
1985 L<perlport> for more on portability concerns.
1987 =item Bit vector size > 32 non-portable
1989 (W portable) Using bit vector sizes larger than 32 is non-portable.
1991 =item Buffer overflow in prime_env_iter: %s
1993 (W internal) A warning peculiar to VMS. While Perl was preparing to iterate over
1994 %ENV, it encountered a logical name or symbol definition which was too long,
1995 so it was truncated to the string shown.
1997 =item Can't check filesystem of script "%s"
1999 (P) For some reason you can't check the filesystem of the script for nosuid.
2001 =item Can't declare class for non-scalar %s in "%s"
2003 (S) Currently, only scalar variables can declared with a specific class
2004 qualifier in a "my" or "our" declaration. The semantics may be extended
2005 for other types of variables in future.
2007 =item Can't declare %s in "%s"
2009 (F) Only scalar, array, and hash variables may be declared as "my" or
2010 "our" variables. They must have ordinary identifiers as names.
2012 =item Can't ignore signal CHLD, forcing to default
2014 (W signal) Perl has detected that it is being run with the SIGCHLD signal
2015 (sometimes known as SIGCLD) disabled. Since disabling this signal
2016 will interfere with proper determination of exit status of child
2017 processes, Perl has reset the signal to its default value.
2018 This situation typically indicates that the parent program under
2019 which Perl may be running (e.g., cron) is being very careless.
2021 =item Can't modify non-lvalue subroutine call
2023 (F) Subroutines meant to be used in lvalue context should be declared as
2024 such, see L<perlsub/"Lvalue subroutines">.
2026 =item Can't read CRTL environ
2028 (S) A warning peculiar to VMS. Perl tried to read an element of %ENV
2029 from the CRTL's internal environment array and discovered the array was
2030 missing. You need to figure out where your CRTL misplaced its environ
2031 or define F<PERL_ENV_TABLES> (see L<perlvms>) so that environ is not searched.
2033 =item Can't remove %s: %s, skipping file
2035 (S) You requested an inplace edit without creating a backup file. Perl
2036 was unable to remove the original file to replace it with the modified
2037 file. The file was left unmodified.
2039 =item Can't return %s from lvalue subroutine
2041 (F) Perl detected an attempt to return illegal lvalues (such
2042 as temporary or readonly values) from a subroutine used as an lvalue.
2043 This is not allowed.
2045 =item Can't weaken a nonreference
2047 (F) You attempted to weaken something that was not a reference. Only
2048 references can be weakened.
2050 =item Character class [:%s:] unknown
2052 (F) The class in the character class [: :] syntax is unknown.
2055 =item Character class syntax [%s] belongs inside character classes
2057 (W unsafe) The character class constructs [: :], [= =], and [. .] go
2058 I<inside> character classes, the [] are part of the construct,
2059 for example: /[012[:alpha:]345]/. Note that [= =] and [. .]
2060 are not currently implemented; they are simply placeholders for
2063 =item Constant is not %s reference
2065 (F) A constant value (perhaps declared using the C<use constant> pragma)
2066 is being dereferenced, but it amounts to the wrong type of reference. The
2067 message indicates the type of reference that was expected. This usually
2068 indicates a syntax error in dereferencing the constant value.
2069 See L<perlsub/"Constant Functions"> and L<constant>.
2071 =item constant(%s): %s
2073 (F) The parser found inconsistencies either while attempting to define an
2074 overloaded constant, or when trying to find the character name specified
2075 in the C<\N{...}> escape. Perhaps you forgot to load the corresponding
2076 C<overload> or C<charnames> pragma? See L<charnames> and L<overload>.
2078 =item CORE::%s is not a keyword
2080 (F) The CORE:: namespace is reserved for Perl keywords.
2082 =item defined(@array) is deprecated
2084 (D) defined() is not usually useful on arrays because it checks for an
2085 undefined I<scalar> value. If you want to see if the array is empty,
2086 just use C<if (@array) { # not empty }> for example.
2088 =item defined(%hash) is deprecated
2090 (D) defined() is not usually useful on hashes because it checks for an
2091 undefined I<scalar> value. If you want to see if the hash is empty,
2092 just use C<if (%hash) { # not empty }> for example.
2094 =item Did not produce a valid header
2098 =item (Did you mean "local" instead of "our"?)
2100 (W misc) Remember that "our" does not localize the declared global variable.
2101 You have declared it again in the same lexical scope, which seems superfluous.
2103 =item Document contains no data
2107 =item entering effective %s failed
2109 (F) While under the C<use filetest> pragma, switching the real and
2110 effective uids or gids failed.
2112 =item false [] range "%s" in regexp
2114 (W regexp) A character class range must start and end at a literal character, not
2115 another character class like C<\d> or C<[:alpha:]>. The "-" in your false
2116 range is interpreted as a literal "-". Consider quoting the "-", "\-".
2119 =item Filehandle %s opened only for output
2121 (W io) You tried to read from a filehandle opened only for writing. If you
2122 intended it to be a read/write filehandle, you needed to open it with
2123 "+<" or "+>" or "+>>" instead of with "<" or nothing. If
2124 you intended only to read from the file, use "<". See
2127 =item flock() on closed filehandle %s
2129 (W closed) The filehandle you're attempting to flock() got itself closed some
2130 time before now. Check your logic flow. flock() operates on filehandles.
2131 Are you attempting to call flock() on a dirhandle by the same name?
2133 =item Global symbol "%s" requires explicit package name
2135 (F) You've said "use strict vars", which indicates that all variables
2136 must either be lexically scoped (using "my"), declared beforehand using
2137 "our", or explicitly qualified to say which package the global variable
2140 =item Hexadecimal number > 0xffffffff non-portable
2142 (W portable) The hexadecimal number you specified is larger than 2**32-1
2143 (4294967295) and therefore non-portable between systems. See
2144 L<perlport> for more on portability concerns.
2146 =item Ill-formed CRTL environ value "%s"
2148 (W internal) A warning peculiar to VMS. Perl tried to read the CRTL's internal
2149 environ array, and encountered an element without the C<=> delimiter
2150 used to separate keys from values. The element is ignored.
2152 =item Ill-formed message in prime_env_iter: |%s|
2154 (W internal) A warning peculiar to VMS. Perl tried to read a logical name
2155 or CLI symbol definition when preparing to iterate over %ENV, and
2156 didn't see the expected delimiter between key and value, so the
2159 =item Illegal binary digit %s
2161 (F) You used a digit other than 0 or 1 in a binary number.
2163 =item Illegal binary digit %s ignored
2165 (W digit) You may have tried to use a digit other than 0 or 1 in a binary number.
2166 Interpretation of the binary number stopped before the offending digit.
2168 =item Illegal number of bits in vec
2170 (F) The number of bits in vec() (the third argument) must be a power of
2171 two from 1 to 32 (or 64, if your platform supports that).
2173 =item Integer overflow in %s number
2175 (W overflow) The hexadecimal, octal or binary number you have specified either
2176 as a literal or as an argument to hex() or oct() is too big for your
2177 architecture, and has been converted to a floating point number. On a
2178 32-bit architecture the largest hexadecimal, octal or binary number
2179 representable without overflow is 0xFFFFFFFF, 037777777777, or
2180 0b11111111111111111111111111111111 respectively. Note that Perl
2181 transparently promotes all numbers to a floating point representation
2182 internally--subject to loss of precision errors in subsequent
2185 =item Invalid %s attribute: %s
2187 The indicated attribute for a subroutine or variable was not recognized
2188 by Perl or by a user-supplied handler. See L<attributes>.
2190 =item Invalid %s attributes: %s
2192 The indicated attributes for a subroutine or variable were not recognized
2193 by Perl or by a user-supplied handler. See L<attributes>.
2195 =item invalid [] range "%s" in regexp
2197 The offending range is now explicitly displayed.
2199 =item Invalid separator character %s in attribute list
2201 (F) Something other than a colon or whitespace was seen between the
2202 elements of an attribute list. If the previous attribute
2203 had a parenthesised parameter list, perhaps that list was terminated
2204 too soon. See L<attributes>.
2206 =item Invalid separator character %s in subroutine attribute list
2208 (F) Something other than a colon or whitespace was seen between the
2209 elements of a subroutine attribute list. If the previous attribute
2210 had a parenthesised parameter list, perhaps that list was terminated
2213 =item leaving effective %s failed
2215 (F) While under the C<use filetest> pragma, switching the real and
2216 effective uids or gids failed.
2218 =item Lvalue subs returning %s not implemented yet
2220 (F) Due to limitations in the current implementation, array and hash
2221 values cannot be returned in subroutines used in lvalue context.
2222 See L<perlsub/"Lvalue subroutines">.
2224 =item Method %s not permitted
2228 =item Missing %sbrace%s on \N{}
2230 (F) Wrong syntax of character name literal C<\N{charname}> within
2231 double-quotish context.
2233 =item Missing command in piped open
2235 (W pipe) You used the C<open(FH, "| command")> or C<open(FH, "command |")>
2236 construction, but the command was missing or blank.
2238 =item Missing name in "my sub"
2240 (F) The reserved syntax for lexically scoped subroutines requires that they
2241 have a name with which they can be found.
2243 =item No %s specified for -%c
2245 (F) The indicated command line switch needs a mandatory argument, but
2246 you haven't specified one.
2248 =item No package name allowed for variable %s in "our"
2250 (F) Fully qualified variable names are not allowed in "our" declarations,
2251 because that doesn't make much sense under existing semantics. Such
2252 syntax is reserved for future extensions.
2254 =item No space allowed after -%c
2256 (F) The argument to the indicated command line switch must follow immediately
2257 after the switch, without intervening spaces.
2259 =item no UTC offset information; assuming local time is UTC
2261 (S) A warning peculiar to VMS. Perl was unable to find the local
2262 timezone offset, so it's assuming that local system time is equivalent
2263 to UTC. If it's not, define the logical name F<SYS$TIMEZONE_DIFFERENTIAL>
2264 to translate to the number of seconds which need to be added to UTC to
2267 =item Octal number > 037777777777 non-portable
2269 (W portable) The octal number you specified is larger than 2**32-1 (4294967295)
2270 and therefore non-portable between systems. See L<perlport> for more
2271 on portability concerns.
2273 See also L<perlport> for writing portable code.
2275 =item panic: del_backref
2277 (P) Failed an internal consistency check while trying to reset a weak
2280 =item panic: kid popen errno read
2282 (F) forked child returned an incomprehensible message about its errno.
2284 =item panic: magic_killbackrefs
2286 (P) Failed an internal consistency check while trying to reset all weak
2287 references to an object.
2289 =item Parentheses missing around "%s" list
2291 (W parenthesis) You said something like
2297 my ($foo, $bar) = @_;
2299 Remember that "my", "our", and "local" bind tighter than comma.
2301 =item Possible Y2K bug: %s
2303 (W y2k) You are concatenating the number 19 with another number, which
2304 could be a potential Year 2000 problem.
2306 =item pragma "attrs" is deprecated, use "sub NAME : ATTRS" instead
2308 (W deprecated) You have written something like this:
2312 use attrs qw(locked);
2315 You should use the new declaration syntax instead.
2321 The C<use attrs> pragma is now obsolete, and is only provided for
2322 backward-compatibility. See L<perlsub/"Subroutine Attributes">.
2325 =item Premature end of script headers
2329 =item Repeat count in pack overflows
2331 (F) You can't specify a repeat count so large that it overflows
2332 your signed integers. See L<perlfunc/pack>.
2334 =item Repeat count in unpack overflows
2336 (F) You can't specify a repeat count so large that it overflows
2337 your signed integers. See L<perlfunc/unpack>.
2339 =item realloc() of freed memory ignored
2341 (S) An internal routine called realloc() on something that had already
2344 =item Reference is already weak
2346 (W misc) You have attempted to weaken a reference that is already weak.
2347 Doing so has no effect.
2349 =item setpgrp can't take arguments
2351 (F) Your system has the setpgrp() from BSD 4.2, which takes no arguments,
2352 unlike POSIX setpgid(), which takes a process ID and process group ID.
2354 =item Strange *+?{} on zero-length expression
2356 (W regexp) You applied a regular expression quantifier in a place where it
2357 makes no sense, such as on a zero-width assertion.
2358 Try putting the quantifier inside the assertion instead. For example,
2359 the way to match "abc" provided that it is followed by three
2360 repetitions of "xyz" is C</abc(?=(?:xyz){3})/>, not C</abc(?=xyz){3}/>.
2362 =item switching effective %s is not implemented
2364 (F) While under the C<use filetest> pragma, we cannot switch the
2365 real and effective uids or gids.
2367 =item This Perl can't reset CRTL environ elements (%s)
2369 =item This Perl can't set CRTL environ elements (%s=%s)
2371 (W internal) Warnings peculiar to VMS. You tried to change or delete an element
2372 of the CRTL's internal environ array, but your copy of Perl wasn't
2373 built with a CRTL that contained the setenv() function. You'll need to
2374 rebuild Perl with a CRTL that does, or redefine F<PERL_ENV_TABLES> (see
2375 L<perlvms>) so that the environ array isn't the target of the change to
2376 %ENV which produced the warning.
2378 =item Too late to run %s block
2380 (W void) A CHECK or INIT block is being defined during run time proper,
2381 when the opportunity to run them has already passed. Perhaps you are
2382 loading a file with C<require> or C<do> when you should be using
2383 C<use> instead. Or perhaps you should put the C<require> or C<do>
2384 inside a BEGIN block.
2386 =item Unknown open() mode '%s'
2388 (F) The second argument of 3-argument open() is not among the list
2389 of valid modes: C<< < >>, C<< > >>, C<<< >> >>>, C<< +< >>,
2390 C<< +> >>, C<<< +>> >>>, C<-|>, C<|->.
2392 =item Unknown process %x sent message to prime_env_iter: %s
2394 (P) An error peculiar to VMS. Perl was reading values for %ENV before
2395 iterating over it, and someone else stuck a message in the stream of
2396 data Perl expected. Someone's very confused, or perhaps trying to
2397 subvert Perl's population of %ENV for nefarious purposes.
2399 =item Unrecognized escape \\%c passed through
2401 (W misc) You used a backslash-character combination which is not recognized
2402 by Perl. The character was understood literally.
2404 =item Unterminated attribute parameter in attribute list
2406 (F) The lexer saw an opening (left) parenthesis character while parsing an
2407 attribute list, but the matching closing (right) parenthesis
2408 character was not found. You may need to add (or remove) a backslash
2409 character to get your parentheses to balance. See L<attributes>.
2411 =item Unterminated attribute list
2413 (F) The lexer found something other than a simple identifier at the start
2414 of an attribute, and it wasn't a semicolon or the start of a
2415 block. Perhaps you terminated the parameter list of the previous attribute
2416 too soon. See L<attributes>.
2418 =item Unterminated attribute parameter in subroutine attribute list
2420 (F) The lexer saw an opening (left) parenthesis character while parsing a
2421 subroutine attribute list, but the matching closing (right) parenthesis
2422 character was not found. You may need to add (or remove) a backslash
2423 character to get your parentheses to balance.
2425 =item Unterminated subroutine attribute list
2427 (F) The lexer found something other than a simple identifier at the start
2428 of a subroutine attribute, and it wasn't a semicolon or the start of a
2429 block. Perhaps you terminated the parameter list of the previous attribute
2432 =item Value of CLI symbol "%s" too long
2434 (W misc) A warning peculiar to VMS. Perl tried to read the value of an %ENV
2435 element from a CLI symbol table, and found a resultant string longer
2436 than 1024 characters. The return value has been truncated to 1024
2439 =item Version number must be a constant number
2441 (P) The attempt to translate a C<use Module n.n LIST> statement into
2442 its equivalent C<BEGIN> block found an internal inconsistency with
2453 Compatibility tests for C<sub : attrs> vs the older C<use attrs>.
2457 Tests for new environment scalar capability (e.g., C<use Env qw($BAR);>).
2461 Tests for new environment array capability (e.g., C<use Env qw(@PATH);>).
2465 IO constants (SEEK_*, _IO*).
2469 Directory-related IO methods (new, read, close, rewind, tied delete).
2471 =item lib/io_multihomed
2473 INET sockets with multi-homed hosts.
2485 Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>.
2489 File test operators.
2493 Verify operations that access pad objects (lexicals and temporaries).
2497 Verify C<exists &sub> operations.
2501 =head1 Incompatible Changes
2503 =head2 Perl Source Incompatibilities
2505 Beware that any new warnings that have been added or old ones
2506 that have been enhanced are B<not> considered incompatible changes.
2508 Since all new warnings must be explicitly requested via the C<-w>
2509 switch or the C<warnings> pragma, it is ultimately the programmer's
2510 responsibility to ensure that warnings are enabled judiciously.
2514 =item CHECK is a new keyword
2516 All subroutine definitions named CHECK are now special. See
2517 C</"Support for CHECK blocks"> for more information.
2519 =item Treatment of list slices of undef has changed
2521 There is a potential incompatibility in the behavior of list slices
2522 that are comprised entirely of undefined values.
2523 See L</"Behavior of list slices is more consistent">.
2525 =head2 Format of $English::PERL_VERSION is different
2527 The English module now sets $PERL_VERSION to $^V (a string value) rather
2528 than C<$]> (a numeric value). This is a potential incompatibility.
2529 Send us a report via perlbug if you are affected by this.
2531 See L</"Improved Perl version numbering system"> for the reasons for
2534 =item Literals of the form C<1.2.3> parse differently
2536 Previously, numeric literals with more than one dot in them were
2537 interpreted as a floating point number concatenated with one or more
2538 numbers. Such "numbers" are now parsed as strings composed of the
2541 For example, C<print 97.98.99> used to output C<97.9899> in earlier
2542 versions, but now prints C<abc>.
2544 See L</"Support for strings represented as a vector of ordinals">.
2546 =item Possibly changed pseudo-random number generator
2548 Perl programs that depend on reproducing a specific set of pseudo-random
2549 numbers may now produce different output due to improvements made to the
2550 rand() builtin. You can use C<sh Configure -Drandfunc=rand> to obtain
2553 See L</"Better pseudo-random number generator">.
2555 =item Hashing function for hash keys has changed
2557 Even though Perl hashes are not order preserving, the apparently
2558 random order encountered when iterating on the contents of a hash
2559 is actually determined by the hashing algorithm used. Improvements
2560 in the algorithm may yield a random order that is B<different> from
2561 that of previous versions, especially when iterating on hashes.
2563 See L</"Better worst-case behavior of hashes"> for additional
2566 =item C<undef> fails on read only values
2568 Using the C<undef> operator on a readonly value (such as $1) has
2569 the same effect as assigning C<undef> to the readonly value--it
2570 throws an exception.
2572 =item Close-on-exec bit may be set on pipe and socket handles
2574 Pipe and socket handles are also now subject to the close-on-exec
2575 behavior determined by the special variable $^F.
2577 See L</"More consistent close-on-exec behavior">.
2579 =item Writing C<"$$1"> to mean C<"${$}1"> is unsupported
2581 Perl 5.004 deprecated the interpretation of C<$$1> and
2582 similar within interpolated strings to mean C<$$ . "1">,
2583 but still allowed it.
2585 In Perl 5.6.0 and later, C<"$$1"> always means C<"${$1}">.
2587 =item delete(), values() and C<\(%h)> operate on aliases to values, not copies
2589 delete(), each(), values() and hashes in a list context return the actual
2590 values in the hash, instead of copies (as they used to in earlier
2591 versions). Typical idioms for using these constructs copy the
2592 returned values, but this can make a significant difference when
2593 creating references to the returned values. Keys in the hash are still
2594 returned as copies when iterating on a hash.
2596 See also L</"delete(), each(), values() and hash iteration are faster">.
2598 =item vec(EXPR,OFFSET,BITS) enforces powers-of-two BITS
2600 vec() generates a run-time error if the BITS argument is not
2601 a valid power-of-two integer.
2603 =item Text of some diagnostic output has changed
2605 Most references to internal Perl operations in diagnostics
2606 have been changed to be more descriptive. This may be an
2607 issue for programs that may incorrectly rely on the exact
2608 text of diagnostics for proper functioning.
2610 =item C<%@> has been removed
2612 The undocumented special variable C<%@> that used to accumulate
2613 "background" errors (such as those that happen in DESTROY())
2614 has been removed, because it could potentially result in memory
2617 =item Parenthesized not() behaves like a list operator
2619 The C<not> operator now falls under the "if it looks like a function,
2620 it behaves like a function" rule.
2622 As a result, the parenthesized form can be used with C<grep> and C<map>.
2623 The following construct used to be a syntax error before, but it works
2626 grep not($_), @things;
2628 On the other hand, using C<not> with a literal list slice may not
2629 work. The following previously allowed construct:
2631 print not (1,2,3)[0];
2633 needs to be written with additional parentheses now:
2635 print not((1,2,3)[0]);
2637 The behavior remains unaffected when C<not> is not followed by parentheses.
2639 =item Semantics of bareword prototype C<(*)> have changed
2641 The semantics of the bareword prototype C<*> have changed. Perl 5.005
2642 always coerced simple scalar arguments to a typeglob, which wasn't useful
2643 in situations where the subroutine must distinguish between a simple
2644 scalar and a typeglob. The new behavior is to not coerce bareword
2645 arguments to a typeglob. The value will always be visible as either
2646 a simple scalar or as a reference to a typeglob.
2648 See L</"More functional bareword prototype (*)">.
2650 =head2 Semantics of bit operators may have changed on 64-bit platforms
2652 If your platform is either natively 64-bit or if Perl has been
2653 configured to used 64-bit integers, i.e., $Config{ivsize} is 8,
2654 there may be a potential incompatibility in the behavior of bitwise
2655 numeric operators (& | ^ ~ << >>). These operators used to strictly
2656 operate on the lower 32 bits of integers in previous versions, but now
2657 operate over the entire native integral width. In particular, note
2658 that unary C<~> will produce different results on platforms that have
2659 different $Config{ivsize}. For portability, be sure to mask off
2660 the excess bits in the result of unary C<~>, e.g., C<~$x & 0xffffffff>.
2662 See L</"Bit operators support full native integer width">.
2664 =head2 More builtins taint their results
2666 As described in L</"Improved security features">, there may be more
2667 sources of taint in a Perl program.
2669 To avoid these new tainting behaviors, you can build Perl with the
2670 Configure option C<-Accflags=-DINCOMPLETE_TAINTS>. Beware that the
2671 ensuing perl binary may be insecure.
2675 =head2 C Source Incompatibilities
2679 =item C<PERL_POLLUTE>
2681 Release 5.005 grandfathered old global symbol names by providing preprocessor
2682 macros for extension source compatibility. As of release 5.6.0, these
2683 preprocessor definitions are not available by default. You need to explicitly
2684 compile perl with C<-DPERL_POLLUTE> to get these definitions. For
2685 extensions still using the old symbols, this option can be
2686 specified via MakeMaker:
2688 perl Makefile.PL POLLUTE=1
2690 =item C<PERL_IMPLICIT_CONTEXT>
2692 This new build option provides a set of macros for all API functions
2693 such that an implicit interpreter/thread context argument is passed to
2694 every API function. As a result of this, something like C<sv_setsv(foo,bar)>
2695 amounts to a macro invocation that actually translates to something like
2696 C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
2697 to not have any significant source compatibility issues, the difference
2698 between a macro and a real function call will need to be considered.
2700 This means that there B<is> a source compatibility issue as a result of
2701 this if your extensions attempt to use pointers to any of the Perl API
2704 Note that the above issue is not relevant to the default build of
2705 Perl, whose interfaces continue to match those of prior versions
2706 (but subject to the other options described here).
2708 See L<perlguts/"The Perl API"> for detailed information on the
2709 ramifications of building Perl with this option.
2711 NOTE: PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built
2712 with one of -Dusethreads, -Dusemultiplicity, or both. It is not
2713 intended to be enabled by users at this time.
2715 =item C<PERL_POLLUTE_MALLOC>
2717 Enabling Perl's malloc in release 5.005 and earlier caused the namespace of
2718 the system's malloc family of functions to be usurped by the Perl versions,
2719 since by default they used the same names. Besides causing problems on
2720 platforms that do not allow these functions to be cleanly replaced, this
2721 also meant that the system versions could not be called in programs that
2722 used Perl's malloc. Previous versions of Perl have allowed this behaviour
2723 to be suppressed with the HIDEMYMALLOC and EMBEDMYMALLOC preprocessor
2726 As of release 5.6.0, Perl's malloc family of functions have default names
2727 distinct from the system versions. You need to explicitly compile perl with
2728 C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
2729 and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
2732 Note that these functions do B<not> constitute Perl's memory allocation API.
2733 See L<perlguts/"Memory Allocation"> for further information about that.
2737 =head2 Compatible C Source API Changes
2741 =item C<PATCHLEVEL> is now C<PERL_VERSION>
2743 The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
2744 are now available by default from perl.h, and reflect the base revision,
2745 patchlevel, and subversion respectively. C<PERL_REVISION> had no
2746 prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
2747 previously available as C<PATCHLEVEL> and C<SUBVERSION>.
2749 The new names cause less pollution of the B<cpp> namespace and reflect what
2750 the numbers have come to stand for in common practice. For compatibility,
2751 the old names are still supported when F<patchlevel.h> is explicitly
2752 included (as required before), so there is no source incompatibility
2757 =head2 Binary Incompatibilities
2759 In general, the default build of this release is expected to be binary
2760 compatible for extensions built with the 5.005 release or its maintenance
2761 versions. However, specific platforms may have broken binary compatibility
2762 due to changes in the defaults used in hints files. Therefore, please be
2763 sure to always check the platform-specific README files for any notes to
2766 The usethreads or usemultiplicity builds are B<not> binary compatible
2767 with the corresponding builds in 5.005.
2769 On platforms that require an explicit list of exports (AIX, OS/2 and Windows,
2770 among others), purely internal symbols such as parser functions and the
2771 run time opcodes are not exported by default. Perl 5.005 used to export
2772 all functions irrespective of whether they were considered part of the
2775 For the full list of public API functions, see L<perlapi>.
2777 =head1 Known Problems
2779 =head2 Thread test failures
2781 The subtests 19 and 20 of lib/thr5005.t test are known to fail due to
2782 fundamental problems in the 5.005 threading implementation. These are
2783 not new failures--Perl 5.005_0x has the same bugs, but didn't have these
2786 =head2 EBCDIC platforms not supported
2788 In earlier releases of Perl, EBCDIC environments like OS390 (also
2789 known as Open Edition MVS) and VM-ESA were supported. Due to changes
2790 required by the UTF-8 (Unicode) support, the EBCDIC platforms are not
2791 supported in Perl 5.6.0.
2793 =head2 In 64-bit HP-UX the lib/io_multihomed test may hang
2795 The lib/io_multihomed test may hang in HP-UX if Perl has been
2796 configured to be 64-bit. Because other 64-bit platforms do not
2797 hang in this test, HP-UX is suspect. All other tests pass
2798 in 64-bit HP-UX. The test attempts to create and connect to
2799 "multihomed" sockets (sockets which have multiple IP addresses).
2801 =head2 NEXTSTEP 3.3 POSIX test failure
2803 In NEXTSTEP 3.3p2 the implementation of the strftime(3) in the
2804 operating system libraries is buggy: the %j format numbers the days of
2805 a month starting from zero, which, while being logical to programmers,
2806 will cause the subtests 19 to 27 of the lib/posix test may fail.
2808 =head2 Tru64 (aka Digital UNIX, aka DEC OSF/1) lib/sdbm test failure with gcc
2810 If compiled with gcc 2.95 the lib/sdbm test will fail (dump core).
2811 The cure is to use the vendor cc, it comes with the operating system
2812 and produces good code.
2814 =head2 UNICOS/mk CC failures during Configure run
2816 In UNICOS/mk the following errors may appear during the Configure run:
2818 Guessing which symbols your C compiler and preprocessor define...
2819 CC-20 cc: ERROR File = try.c, Line = 3
2821 bad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K
2823 4 errors detected in the compilation of "try.c".
2825 The culprit is the broken awk of UNICOS/mk. The effect is fortunately
2826 rather mild: Perl itself is not adversely affected by the error, only
2827 the h2ph utility coming with Perl, and that is rather rarely needed
2830 =head2 Arrow operator and arrays
2832 When the left argument to the arrow operator C<< -> >> is an array, or
2833 the C<scalar> operator operating on an array, the result of the
2834 operation must be considered erroneous. For example:
2839 These expressions will get run-time errors in some future release of
2842 =head2 Experimental features
2844 As discussed above, many features are still experimental. Interfaces and
2845 implementation of these features are subject to change, and in extreme cases,
2846 even subject to removal in some future release of Perl. These features
2847 include the following:
2855 =item 64-bit support
2857 =item Lvalue subroutines
2859 =item Weak references
2861 =item The pseudo-hash data type
2863 =item The Compiler suite
2865 =item Internal implementation of file globbing
2869 =item The regular expression constructs C<(?{ code })> and C<(??{ code })>
2873 =head1 Obsolete Diagnostics
2877 =item Character class syntax [: :] is reserved for future extensions
2879 (W) Within regular expression character classes ([]) the syntax beginning
2880 with "[:" and ending with ":]" is reserved for future extensions.
2881 If you need to represent those character sequences inside a regular
2882 expression character class, just quote the square brackets with the
2883 backslash: "\[:" and ":\]".
2885 =item Ill-formed logical name |%s| in prime_env_iter
2887 (W) A warning peculiar to VMS. A logical name was encountered when preparing
2888 to iterate over %ENV which violates the syntactic rules governing logical
2889 names. Because it cannot be translated normally, it is skipped, and will not
2890 appear in %ENV. This may be a benign occurrence, as some software packages
2891 might directly modify logical name tables and introduce nonstandard names,
2892 or it may indicate that a logical name table has been corrupted.
2894 =item Probable precedence problem on %s
2896 (W) The compiler found a bareword where it expected a conditional,
2897 which often indicates that an || or && was parsed as part of the
2898 last argument of the previous construct, for example:
2902 =item regexp too big
2904 (F) The current implementation of regular expressions uses shorts as
2905 address offsets within a string. Unfortunately this means that if
2906 the regular expression compiles to longer than 32767, it'll blow up.
2907 Usually when you want a regular expression this big, there is a better
2908 way to do it with multiple statements. See L<perlre>.
2910 =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
2912 (D) Perl versions before 5.004 misinterpreted any type marker followed
2913 by "$" and a digit. For example, "$$0" was incorrectly taken to mean
2914 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
2916 However, the developers of Perl 5.004 could not fix this bug completely,
2917 because at least two widely-used modules depend on the old meaning of
2918 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
2919 old (broken) way inside strings; but it generates this message as a
2920 warning. And in Perl 5.005, this special treatment will cease.
2924 =head1 Reporting Bugs
2926 If you find what you think is a bug, you might check the
2927 articles recently posted to the comp.lang.perl.misc newsgroup.
2928 There may also be information at http://www.perl.com/perl/, the Perl
2931 If you believe you have an unreported bug, please run the B<perlbug>
2932 program included with your release. Be sure to trim your bug down
2933 to a tiny but sufficient test case. Your bug report, along with the
2934 output of C<perl -V>, will be sent off to perlbug@perl.com to be
2935 analysed by the Perl porting team.
2939 The F<Changes> file for exhaustive details on what changed.
2941 The F<INSTALL> file for how to build Perl.
2943 The F<README> file for general stuff.
2945 The F<Artistic> and F<Copying> files for copyright information.
2949 Written by Gurusamy Sarathy <F<gsar@activestate.com>>, with many
2950 contributions from The Perl Porters.
2952 Send omissions or corrections to <F<perlbug@perl.com>>.