3 perldelta - what's new for perl v5.6.0
7 This document describes differences between the 5.005 release and the 5.6.0
10 =head1 Core Enhancements
12 =head2 Interpreter cloning, threads, and concurrency
14 Perl 5.6.0 introduces the beginnings of support for running multiple
15 interpreters concurrently in different threads. In conjunction with
16 the perl_clone() API call, which can be used to selectively duplicate
17 the state of any given interpreter, it is possible to compile a
18 piece of code once in an interpreter, clone that interpreter
19 one or more times, and run all the resulting interpreters in distinct
22 On the Windows platform, this feature is used to emulate fork() at the
23 interpreter level. See L<perlfork> for details about that.
25 This feature is still in evolution. It is eventually meant to be used
26 to selectively clone a subroutine and data reachable from that
27 subroutine in a separate interpreter and run the cloned subroutine
28 in a separate thread. Since there is no shared data between the
29 interpreters, little or no locking will be needed (unless parts of
30 the symbol table are explicitly shared). This is obviously intended
31 to be an easy-to-use replacement for the existing threads support.
33 Support for cloning interpreters and interpreter concurrency can be
34 enabled using the -Dusethreads Configure option (see win32/Makefile for
35 how to enable it on Windows.) The resulting perl executable will be
36 functionally identical to one that was built with -Dmultiplicity, but
37 the perl_clone() API call will only be available in the former.
39 -Dusethreads enables the cpp macro USE_ITHREADS by default, which in turn
40 enables Perl source code changes that provide a clear separation between
41 the op tree and the data it operates with. The former is immutable, and
42 can therefore be shared between an interpreter and all of its clones,
43 while the latter is considered local to each interpreter, and is therefore
44 copied for each clone.
46 Note that building Perl with the -Dusemultiplicity Configure option
47 is adequate if you wish to run multiple B<independent> interpreters
48 concurrently in different threads. -Dusethreads only provides the
49 additional functionality of the perl_clone() API call and other
50 support for running B<cloned> interpreters concurrently.
52 NOTE: This is an experimental feature. Implementation details are
55 =head2 Lexically scoped warning categories
57 You can now control the granularity of warnings emitted by perl at a finer
58 level using the C<use warnings> pragma. L<warnings> and L<perllexwarn>
59 have copious documentation on this feature.
61 =head2 Unicode and UTF-8 support
63 Perl now uses UTF-8 as its internal representation for character
64 strings. The C<utf8> and C<bytes> pragmas are used to control this support
65 in the current lexical scope. See L<perlunicode>, L<utf8> and L<bytes> for
68 This feature is expected to evolve quickly to support some form of I/O
69 disciplines that can be used to specify the kind of input and output data
70 (bytes or characters). Until that happens, additional modules from CPAN
71 will be needed to complete the toolkit for dealing with Unicode.
73 NOTE: This should be considered an experimental feature. Implementation
74 details are subject to change.
76 =head2 Support for interpolating named characters
78 The new C<\N> escape interpolates named characters within strings.
79 For example, C<"Hi! \N{WHITE SMILING FACE}"> evaluates to a string
80 with a unicode smiley face at the end.
82 =head2 "our" declarations
84 An "our" declaration introduces a value that can be best understood
85 as a lexically scoped symbolic alias to a global variable in the
86 package that was current where the variable was declared. This is
87 mostly useful as an alternative to the C<vars> pragma, but also provides
88 the opportunity to introduce typing and other attributes for such
89 variables. See L<perlfunc/our>.
91 =head2 Support for strings represented as a vector of ordinals
93 Literals of the form C<v1.2.3.4> are now parsed as a string composed
94 of characters with the specified ordinals. This is an alternative, more
95 readable way to construct (possibly unicode) strings instead of
96 interpolating characters, as in C<"\x{1}\x{2}\x{3}\x{4}">. The leading
97 C<v> may be omitted if there are more than two ordinals, so C<1.2.3> is
98 parsed the same as C<v1.2.3>.
100 Strings written in this form are also useful to represent version "numbers".
101 It is easy to compare such version "numbers" (which are really just plain
102 strings) using any of the usual string comparison operators C<eq>, C<ne>,
103 C<lt>, C<gt>, etc., or perform bitwise string operations on them using C<|>,
106 In conjunction with the new C<$^V> magic variable (which contains
107 the perl version as a string), such literals can be used as a readable way
108 to check if you're running a particular version of Perl:
110 # this will parse in older versions of Perl also
111 if ($^V and $^V gt v5.6.0) {
112 # new features supported
115 C<require> and C<use> also have some special magic to support such literals.
116 They will be interpreted as a version rather than as a module name:
118 require v5.6.0; # croak if $^V lt v5.6.0
119 use v5.6.0; # same, but croaks at compile-time
121 Alternatively, the C<v> may be omitted if there is more than one dot:
126 Also, C<sprintf> and C<printf> support the Perl-specific format flag C<%v>
127 to print ordinals of characters in arbitrary strings:
129 printf "v%vd", $^V; # prints current version, such as "v5.5.650"
130 printf "%*vX", ":", $addr; # formats IPv6 address
131 printf "%*vb", " ", $bits; # displays bitstring
133 See L<perldata/"Scalar value constructors"> for additional information.
135 =head2 Improved Perl version numbering system
137 Beginning with Perl version 5.6.0, the version number convention has been
138 changed to a "dotted integer" scheme that is more commonly found in open
141 Maintenance versions of v5.6.0 will be released as v5.6.1, v5.6.2 etc.
142 The next development series following v5.6.0 will be numbered v5.7.x,
143 beginning with v5.7.0, and the next major production release following
144 v5.6.0 will be v5.8.0.
146 The English module now sets $PERL_VERSION to $^V (a string value) rather
147 than C<$]> (a numeric value). (This is a potential incompatibility.
148 Send us a report via perlbug if you are affected by this.)
150 The v1.2.3 syntax is also now legal in Perl.
151 See L<Support for strings represented as a vector of ordinals> for more on that.
153 To cope with the new versioning system's use of at least three significant
154 digits for each version component, the method used for incrementing the
155 subversion number has also changed slightly. We assume that versions older
156 than v5.6.0 have been incrementing the subversion component in multiples of
157 10. Versions after v5.6.0 will increment them by 1. Thus, using the new
158 notation, 5.005_03 is the "same" as v5.5.30, and the first maintenance
159 version following v5.6.0 will be v5.6.1 (which should be read as being
160 equivalent to a floating point value of 5.006_001 in the older format,
163 =head2 New syntax for declaring subroutine attributes
165 Formerly, if you wanted to mark a subroutine as being a method call or
166 as requiring an automatic lock() when it is entered, you had to declare
167 that with a C<use attrs> pragma in the body of the subroutine.
168 That can now be accomplished with declaration syntax, like this:
170 sub mymethod : locked method ;
172 sub mymethod : locked method {
176 sub othermethod :locked :method ;
178 sub othermethod :locked :method {
183 (Note how only the first C<:> is mandatory, and whitespace surrounding
184 the C<:> is optional.)
186 F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes
187 with the stubs they provide. See L<attributes>.
189 =head2 File and directory handles can be autovivified
191 Similar to how constructs such as C<< $x->[0] >> autovivify a reference,
192 handle constructors (open(), opendir(), pipe(), socketpair(), sysopen(),
193 socket(), and accept()) now autovivify a file or directory handle
194 if the handle passed to them is an uninitialized scalar variable. This
195 allows the constructs such as C<open(my $fh, ...)> and C<open(local $fh,...)>
196 to be used to create filehandles that will conveniently be closed
197 automatically when the scope ends, provided there are no other references
198 to them. This largely eliminates the need for typeglobs when opening
199 filehandles that must be passed around, as in the following example:
203 or die "Can't open '@_': $!";
208 my $f = myopen("</etc/motd");
210 # $f implicitly closed here
213 =head2 open() with more than two arguments
215 If open() is passed three arguments instead of two, the second argument
216 is used as the mode and the third argument is taken to be the file name.
217 This is primarily useful for protecting against unintended magic behavior
218 of the traditional two-argument form. See L<perlfunc/open>.
220 =head2 64-bit support
222 Any platform that has 64-bit integers either
224 (1) natively as longs or ints
225 (2) via special compiler flags
226 (3) using long long or int64_t
228 is able to use "quads" (64-bit integers) as follows:
234 constants (decimal, hexadecimal, octal, binary) in the code
238 arguments to oct() and hex()
242 arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)
250 pack() and unpack() "q" and "Q" formats
254 in basic arithmetics: + - * / % (NOTE: operating close to the limits
255 of the integer values may produce surprising results)
259 in bit arithmetics: & | ^ ~ << >> (NOTE: these used to be forced
260 to be 32 bits wide but now operate on the full native width.)
268 Note that unless you have the case (a) you will have to configure
269 and compile Perl using the -Duse64bitint Configure flag.
271 NOTE: The Configure flags -Duselonglong and -Duse64bits have been
272 deprecated. Use -Duse64bitint instead.
274 There are actually two modes of 64-bitness: the first one is achieved
275 using Configure -Duse64bitint and the second one using Configure
276 -Duse64bitall. The difference is that the first one is minimal and
277 the second one maximal. The first works in more places than the second.
279 The C<use64bitint> does only as much as is required to get 64-bit
280 integers into Perl (this may mean, for example, using "long longs")
281 while your memory may still be limited to 2 gigabytes (because your
282 pointers could still be 32-bit). Note that the name C<64bitint> does
283 not imply that your C compiler will be using 64-bit C<int>s (it might,
284 but it doesn't have to): the C<use64bitint> means that you will be
285 able to have 64 bits wide scalar values.
287 The C<use64bitall> goes all the way by attempting to switch also
288 integers (if it can), longs (and pointers) to being 64-bit. This may
289 create an even more binary incompatible Perl than -Duse64bitint: the
290 resulting executable may not run at all in a 32-bit box, or you may
291 have to reboot/reconfigure/rebuild your operating system to be 64-bit
294 Natively 64-bit systems like Alpha and Cray need neither -Duse64bitint
297 Last but not least: note that due to Perl's habit of always using
298 floating point numbers, the quads are still not true integers.
299 When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned,
300 -9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they
301 are silently promoted to floating point numbers, after which they will
302 start losing precision (in their lower digits).
304 NOTE: 64-bit support is still experimental on most platforms.
305 Existing support only covers the LP64 data model. In particular, the
306 LLP64 data model is not yet supported. 64-bit libraries and system
307 APIs on many platforms have not stabilized--your mileage may vary.
309 =head2 Large file support
311 If you have filesystems that support "large files" (files larger than
312 2 gigabytes), you may now also be able to create and access them from
315 NOTE: The default action is to enable large file support, if
316 available on the platform.
318 If the large file support is on, and you have a Fcntl constant
319 O_LARGEFILE, the O_LARGEFILE is automatically added to the flags
322 Beware that unless your filesystem also supports "sparse files" seeking
323 to umpteen petabytes may be inadvisable.
325 Note that in addition to requiring a proper file system to do large
326 files you may also need to adjust your per-process (or your
327 per-system, or per-process-group, or per-user-group) maximum filesize
328 limits before running Perl scripts that try to handle large files,
329 especially if you intend to write such files.
331 Finally, in addition to your process/process group maximum filesize
332 limits, you may have quota limits on your filesystems that stop you
333 (your user id or your user group id) from using large files.
335 Adjusting your process/user/group/file system/operating system limits
336 is outside the scope of Perl core language. For process limits, you
337 may try increasing the limits using your shell's limits/limit/ulimit
338 command before running Perl. The BSD::Resource extension (not
339 included with the standard Perl distribution) may also be of use, it
340 offers the getrlimit/setrlimit interface that can be used to adjust
341 process resource usage limits, including the maximum filesize limit.
345 In some systems you may be able to use long doubles to enhance the
346 range and precision of your double precision floating point numbers
347 (that is, Perl's numbers). Use Configure -Duselongdouble to enable
348 this support (if it is available).
352 You can "Configure -Dusemorebits" to turn on both the 64-bit support
353 and the long double support.
355 =head2 Enhanced support for sort() subroutines
357 Perl subroutines with a prototype of C<($$)>, and XSUBs in general, can
358 now be used as sort subroutines. In either case, the two elements to
359 be compared are passed as normal parameters in @_. See L<perlfunc/sort>.
361 For unprototyped sort subroutines, the historical behavior of passing
362 the elements to be compared as the global variables $a and $b remains
365 =head2 C<sort $coderef @foo> allowed
367 sort() did not accept a subroutine reference as the comparison
368 function in earlier versions. This is now permitted.
370 =head2 File globbing implemented internally
372 Perl now uses the File::Glob implementation of the glob() operator
373 automatically. This avoids using an external csh process and the
374 problems associated with it.
376 NOTE: This is currently an experimental feature. Interfaces and
377 implementation are subject to change.
379 =head2 Support for CHECK blocks
381 In addition to C<BEGIN>, C<INIT>, C<END>, C<DESTROY> and C<AUTOLOAD>,
382 subroutines named C<CHECK> are now special. These are queued up during
383 compilation and behave similar to END blocks, except they are called at
384 the end of compilation rather than at the end of execution. They cannot
387 =head2 POSIX character class syntax [: :] supported
389 For example to match alphabetic characters use /[[:alpha:]]/.
390 See L<perlre> for details.
392 =head2 Better pseudo-random number generator
394 In 5.005_0x and earlier, perl's rand() function used the C library
395 rand(3) function. As of 5.005_52, Configure tests for drand48(),
396 random(), and rand() (in that order) and picks the first one it finds.
398 These changes should result in better random numbers from rand().
400 =head2 Improved C<qw//> operator
402 The C<qw//> operator is now evaluated at compile time into a true list
403 instead of being replaced with a run time call to C<split()>. This
404 removes the confusing misbehaviour of C<qw//> in scalar context, which
405 had inherited that behaviour from split().
409 $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
411 now correctly prints "3|a", instead of "2|a".
413 =head2 Better worst-case behavior of hashes
415 Small changes in the hashing algorithm have been implemented in
416 order to improve the distribution of lower order bits in the
417 hashed value. This is expected to yield better performance on
418 keys that are repeated sequences.
420 =head2 pack() format 'Z' supported
422 The new format type 'Z' is useful for packing and unpacking null-terminated
423 strings. See L<perlfunc/"pack">.
425 =head2 pack() format modifier '!' supported
427 The new format type modifier '!' is useful for packing and unpacking
428 native shorts, ints, and longs. See L<perlfunc/"pack">.
430 =head2 pack() and unpack() support counted strings
432 The template character '/' can be used to specify a counted string
433 type to be packed or unpacked. See L<perlfunc/"pack">.
435 =head2 Comments in pack() templates
437 The '#' character in a template introduces a comment up to
438 end of the line. This facilitates documentation of pack()
441 =head2 Weak references
443 In previous versions of Perl, you couldn't cache objects so as
444 to allow them to be deleted if the last reference from outside
445 the cache is deleted. The reference in the cache would hold a
446 reference count on the object and the objects would never be
449 Another familiar problem is with circular references. When an
450 object references itself, its reference count would never go
451 down to zero, and it would not get destroyed until the program
454 Weak references solve this by allowing you to "weaken" any
455 reference, that is, make it not count towards the reference count.
456 When the last non-weak reference to an object is deleted, the object
457 is destroyed and all the weak references to the object are
458 automatically undef-ed.
460 To use this feature, you need the WeakRef package from CPAN, which
461 contains additional documentation.
463 NOTE: This is an experimental feature. Details are subject to change.
465 =head2 Binary numbers supported
467 Binary numbers are now supported as literals, in s?printf formats, and
471 printf "The answer is: %b\n", oct("0b101010");
473 =head2 Lvalue subroutines
475 Subroutines can now return modifiable lvalues.
476 See L<perlsub/"Lvalue subroutines">.
478 NOTE: This is an experimental feature. Details are subject to change.
480 =head2 Some arrows may be omitted in calls through references
482 Perl now allows the arrow to be omitted in many constructs
483 involving subroutine calls through references. For example,
484 C<< $foo[10]->('foo') >> may now be written C<$foo[10]('foo')>.
485 This is rather similar to how the arrow may be omitted from
486 C<< $foo[10]->{'foo'} >>. Note however, that the arrow is still
487 required for C<< foo(10)->('bar') >>.
489 =head2 Boolean assignment operators are legal lvalues
491 Constructs such as C<($a ||= 2) += 1> are now allowed.
493 =head2 exists() is supported on subroutine names
495 The exists() builtin now works on subroutine names. A subroutine
496 is considered to exist if it has been declared (even if implicitly).
497 See L<perlfunc/exists> for examples.
499 =head2 exists() and delete() are supported on array elements
501 The exists() and delete() builtins now work on simple arrays as well.
502 The behavior is similar to that on hash elements.
504 exists() can be used to check whether an array element has been
505 initialized. This avoids autovivifying array elements that don't exist.
506 If the array is tied, the EXISTS() method in the corresponding tied
507 package will be invoked.
509 delete() may be used to remove an element from the array and return
510 it. The array element at that position returns to its uninitialized
511 state, so that testing for the same element with exists() will return
512 false. If the element happens to be the one at the end, the size of
513 the array also shrinks up to the highest element that tests true for
514 exists(), or 0 if none such is found. If the array is tied, the DELETE()
515 method in the corresponding tied package will be invoked.
517 See L<perlfunc/exists> and L<perlfunc/delete> for examples.
519 =head2 Pseudo-hashes work better
521 Dereferencing some types of reference values in a pseudo-hash,
522 such as C<< $ph->{foo}[1] >>, was accidentally disallowed. This has
525 When applied to a pseudo-hash element, exists() now reports whether
526 the specified value exists, not merely if the key is valid.
528 delete() now works on pseudo-hashes. When given a pseudo-hash element
529 or slice it deletes the values corresponding to the keys (but not the keys
530 themselves). See L<perlref/"Pseudo-hashes: Using an array as a hash">.
532 Pseudo-hash slices with constant keys are now optimized to array lookups
535 List assignments to pseudo-hash slices are now supported.
537 The C<fields> pragma now provides ways to create pseudo-hashes, via
538 fields::new() and fields::phash(). See L<fields>.
540 NOTE: The pseudo-hash data type continues to be experimental.
541 Limiting oneself to the interface elements provided by the
542 fields pragma will provide protection from any future changes.
544 =head2 Automatic flushing of output buffers
546 fork(), exec(), system(), qx//, and pipe open()s now flush buffers
547 of all files opened for output when the operation was attempted. This
548 mostly eliminates confusing buffering mishaps suffered by users unaware
549 of how Perl internally handles I/O.
551 This is not supported on some platforms like Solaris where a suitably
552 correct implementation of fflush(NULL) isn't available.
554 =head2 Better diagnostics on meaningless filehandle operations
556 Constructs such as C<< open(<FH>) >> and C<< close(<FH>) >>
557 are compile time errors. Attempting to read from filehandles that
558 were opened only for writing will now produce warnings (just as
559 writing to read-only filehandles does).
561 =head2 Where possible, buffered data discarded from duped input filehandle
563 C<< open(NEW, "<&OLD") >> now attempts to discard any data that
564 was previously read and buffered in C<OLD> before duping the handle.
565 On platforms where doing this is allowed, the next read operation
566 on C<NEW> will return the same data as the corresponding operation
567 on C<OLD>. Formerly, it would have returned the data from the start
568 of the following disk block instead.
570 =head2 eof() has the same old magic as <>
572 C<eof()> would return true if no attempt to read from C<< <> >> had
573 yet been made. C<eof()> has been changed to have a little magic of its
574 own, it now opens the C<< <> >> files.
576 =head2 binmode() can be used to set :crlf and :raw modes
578 binmode() now accepts a second argument that specifies a discipline
579 for the handle in question. The two pseudo-disciplines ":raw" and
580 ":crlf" are currently supported on DOS-derivative platforms.
581 See L<perlfunc/"binmode"> and L<open>.
583 =head2 C<-T> filetest recognizes UTF-8 encoded files as "text"
585 The algorithm used for the C<-T> filetest has been enhanced to
586 correctly identify UTF-8 content as "text".
588 =head2 system(), backticks and pipe open now reflect exec() failure
590 On Unix and similar platforms, system(), qx() and open(FOO, "cmd |")
591 etc., are implemented via fork() and exec(). When the underlying
592 exec() fails, earlier versions did not report the error properly,
593 since the exec() happened to be in a different process.
595 The child process now communicates with the parent about the
596 error in launching the external command, which allows these
597 constructs to return with their usual error value and set $!.
599 =head2 Improved diagnostics
601 Line numbers are no longer suppressed (under most likely circumstances)
602 during the global destruction phase.
604 Diagnostics emitted from code running in threads other than the main
605 thread are now accompanied by the thread ID.
607 Embedded null characters in diagnostics now actually show up. They
608 used to truncate the message in prior versions.
610 $foo::a and $foo::b are now exempt from "possible typo" warnings only
611 if sort() is encountered in package C<foo>.
613 Unrecognized alphabetic escapes encountered when parsing quote
614 constructs now generate a warning, since they may take on new
615 semantics in later versions of Perl.
617 Many diagnostics now report the internal operation in which the warning
618 was provoked, like so:
620 Use of uninitialized value in concatenation (.) at (eval 1) line 1.
621 Use of uninitialized value in print at (eval 1) line 1.
623 Diagnostics that occur within eval may also report the file and line
624 number where the eval is located, in addition to the eval sequence
625 number and the line number within the evaluated text itself. For
628 Not enough arguments for scalar at (eval 4)[newlib/perl5db.pl:1411] line 2, at EOF
630 =head2 Diagnostics follow STDERR
632 Diagnostic output now goes to whichever file the C<STDERR> handle
633 is pointing at, instead of always going to the underlying C runtime
636 =head2 More consistent close-on-exec behavior
638 On systems that support a close-on-exec flag on filehandles, the
639 flag is now set for any handles created by pipe(), socketpair(),
640 socket(), and accept(), if that is warranted by the value of $^F
641 that may be in effect. Earlier versions neglected to set the flag
642 for handles created with these operators. See L<perlfunc/pipe>,
643 L<perlfunc/socketpair>, L<perlfunc/socket>, L<perlfunc/accept>,
646 =head2 syswrite() ease-of-use
648 The length argument of C<syswrite()> has become optional.
650 =head2 Better syntax checks on parenthesized unary operators
654 print defined(&foo,&bar,&baz);
655 print uc("foo","bar","baz");
658 used to be accidentally allowed in earlier versions, and produced
659 unpredictable behaviour. Some produced ancillary warnings
660 when used in this way; others silently did the wrong thing.
662 The parenthesized forms of most unary operators that expect a single
663 argument now ensure that they are not called with more than one
664 argument, making the cases shown above syntax errors. The usual
667 print defined &foo, &bar, &baz;
668 print uc "foo", "bar", "baz";
671 remains unchanged. See L<perlop>.
673 =head2 Bit operators support full native integer width
675 The bit operators (& | ^ ~ << >>) now operate on the full native
676 integral width (the exact size of which is available in $Config{ivsize}).
677 For example, if your platform is either natively 64-bit or if Perl
678 has been configured to use 64-bit integers, these operations apply
679 to 8 bytes (as opposed to 4 bytes on 32-bit platforms).
680 For portability, be sure to mask off the excess bits in the result of
681 unary C<~>, e.g., C<~$x & 0xffffffff>.
683 =head2 Improved security features
685 More potentially unsafe operations taint their results for improved
688 The C<passwd> and C<shell> fields returned by the getpwent(), getpwnam(),
689 and getpwuid() are now tainted, because the user can affect their own
690 encrypted password and login shell.
692 The variable modified by shmread(), and messages returned by msgrcv()
693 (and its object-oriented interface IPC::SysV::Msg::rcv) are also tainted,
694 because other untrusted processes can modify messages and shared memory
695 segments for their own nefarious purposes.
697 =head2 More functional bareword prototype (*)
699 Bareword prototypes have been rationalized to enable them to be used
700 to override builtins that accept barewords and interpret them in
701 a special way, such as C<require> or C<do>.
703 Arguments prototyped as C<*> will now be visible within the subroutine
704 as either a simple scalar or as a reference to a typeglob.
705 See L<perlsub/Prototypes>.
707 =head2 C<require> and C<do> may be overridden
709 C<require> and C<do 'file'> operations may be overridden locally
710 by importing subroutines of the same name into the current package
711 (or globally by importing them into the CORE::GLOBAL:: namespace).
712 Overriding C<require> will also affect C<use>, provided the override
713 is visible at compile-time.
714 See L<perlsub/"Overriding Built-in Functions">.
716 =head2 $^X variables may now have names longer than one character
718 Formerly, $^X was synonymous with ${"\cX"}, but $^XY was a syntax
719 error. Now variable names that begin with a control character may be
720 arbitrarily long. However, for compatibility reasons, these variables
721 I<must> be written with explicit braces, as C<${^XY}> for example.
722 C<${^XYZ}> is synonymous with ${"\cXYZ"}. Variable names with more
723 than one control character, such as C<${^XY^Z}>, are illegal.
725 The old syntax has not changed. As before, `^X' may be either a
726 literal control-X character or the two-character sequence `caret' plus
727 `X'. When braces are omitted, the variable name stops after the
728 control character. Thus C<"$^XYZ"> continues to be synonymous with
729 C<$^X . "YZ"> as before.
731 As before, lexical variables may not have names beginning with control
732 characters. As before, variables whose names begin with a control
733 character are always forced to be in package `main'. All such variables
734 are reserved for future extensions, except those that begin with
735 C<^_>, which may be used by user programs and are guaranteed not to
736 acquire special meaning in any future version of Perl.
738 =head2 New variable $^C reflects C<-c> switch
740 C<$^C> has a boolean value that reflects whether perl is being run
741 in compile-only mode (i.e. via the C<-c> switch). Since
742 BEGIN blocks are executed under such conditions, this variable
743 enables perl code to determine whether actions that make sense
744 only during normal running are warranted. See L<perlvar>.
746 =head2 New variable $^V contains Perl version as a string
748 C<$^V> contains the Perl version number as a string composed of
749 characters whose ordinals match the version numbers, i.e. v5.6.0.
750 This may be used in string comparisons.
752 See C<Support for strings represented as a vector of ordinals> for an
755 =head2 Optional Y2K warnings
757 If Perl is built with the cpp macro C<PERL_Y2KWARN> defined,
758 it emits optional warnings when concatenating the number 19
761 This behavior must be specifically enabled when running Configure.
762 See F<INSTALL> and F<README.Y2K>.
764 =head2 Arrays now always interpolate into double-quoted strings
766 In double-quoted strings, arrays now interpolate, no matter what. The
767 behavior in earlier versions of perl 5 was that arrays would interpolate
768 into strings if the array had been mentioned before the string was
769 compiled, and otherwise Perl would raise a fatal compile-time error.
770 In versions 5.000 through 5.003, the error was
772 Literal @example now requires backslash
774 In versions 5.004_01 through 5.6.0, the error was
776 In string, @example now must be written as \@example
778 The idea here was to get people into the habit of writing
779 C<"fred\@example.com"> when they wanted a literal C<@> sign, just as
780 they have always written C<"Give me back my \$5"> when they wanted a
783 Starting with 5.6.1, when Perl now sees an C<@> sign in a
784 double-quoted string, it I<always> attempts to interpolate an array,
785 regardless of whether or not the array has been used or declared
786 already. The fatal error has been downgraded to an optional warning:
788 Possible unintended interpolation of @example in string
790 This warns you that C<"fred@example.com"> is going to turn into
791 C<fred.com> if you don't backslash the C<@>.
792 See http://www.plover.com/~mjd/perl/at-error.html for more details
793 about the history here.
795 =head1 Modules and Pragmata
803 While used internally by Perl as a pragma, this module also
804 provides a way to fetch subroutine and variable attributes.
809 The Perl Compiler suite has been extensively reworked for this
810 release. More of the standard Perl testsuite passes when run
811 under the Compiler, but there is still a significant way to
812 go to achieve production quality compiled executables.
814 NOTE: The Compiler suite remains highly experimental. The
815 generated code may not be correct, even when it manages to execute
820 Overall, Benchmark results exhibit lower average error and better timing
823 You can now run tests for I<n> seconds instead of guessing the right
824 number of tests to run: e.g., timethese(-5, ...) will run each
825 code for at least 5 CPU seconds. Zero as the "number of repetitions"
826 means "for at least 3 CPU seconds". The output format has also
827 changed. For example:
829 use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
831 will now output something like this:
833 Benchmark: running a, b, each for at least 5 CPU seconds...
834 a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
835 b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
837 New features: "each for at least N CPU seconds...", "wallclock secs",
838 and the "@ operations/CPU second (n=operations)".
840 timethese() now returns a reference to a hash of Benchmark objects containing
841 the test results, keyed on the names of the tests.
843 timethis() now returns the iterations field in the Benchmark result object
846 timethese(), timethis(), and the new cmpthese() (see below) can also take
847 a format specifier of 'none' to suppress output.
849 A new function countit() is just like timeit() except that it takes a
850 TIME instead of a COUNT.
852 A new function cmpthese() prints a chart comparing the results of each test
853 returned from a timethese() call. For each possible pair of tests, the
854 percentage speed difference (iters/sec or seconds/iter) is shown.
856 For other details, see L<Benchmark>.
860 The ByteLoader is a dedicated extension to generate and run
861 Perl bytecode. See L<ByteLoader>.
865 References can now be used.
867 The new version also allows a leading underscore in constant names, but
868 disallows a double leading underscore (as in "__LINE__"). Some other names
869 are disallowed or warned against, including BEGIN, END, etc. Some names
870 which were forced into main:: used to fail silently in some cases; now they're
871 fatal (outside of main::) and an optional warning (inside of main::).
872 The ability to detect whether a constant had been set with a given name has
879 This pragma implements the C<\N> string escape. See L<charnames>.
883 A C<Maxdepth> setting can be specified to avoid venturing
884 too deeply into deep data structures. See L<Data::Dumper>.
886 The XSUB implementation of Dump() is now automatically called if the
887 C<Useqq> setting is not in use.
889 Dumping C<qr//> objects works correctly.
893 C<DB> is an experimental module that exposes a clean abstraction
894 to Perl's debugging API.
898 DB_File can now be built with Berkeley DB versions 1, 2 or 3.
899 See C<ext/DB_File/Changes>.
903 Devel::DProf, a Perl source code profiler has been added. See
904 L<Devel::DProf> and L<dprofpp>.
908 The Devel::Peek module provides access to the internal representation
909 of Perl variables and data. It is a data debugging tool for the XS programmer.
913 The Dumpvalue module provides screen dumps of Perl data.
917 DynaLoader now supports a dl_unload_file() function on platforms that
918 support unloading shared objects using dlclose().
920 Perl can also optionally arrange to unload all extension shared objects
921 loaded by Perl. To enable this, build Perl with the Configure option
922 C<-Accflags=-DDL_UNLOAD_ALL_AT_EXIT>. (This maybe useful if you are
923 using Apache with mod_perl.)
927 $PERL_VERSION now stands for C<$^V> (a string value) rather than for C<$]>
932 Env now supports accessing environment variables like PATH as array
937 More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
938 large file (more than 4GB) access (NOTE: the O_LARGEFILE is
939 automatically added to sysopen() flags if large file support has been
940 configured, as is the default), Free/Net/OpenBSD locking behaviour
941 flags F_FLOCK, F_POSIX, Linux F_SHLCK, and O_ACCMODE: the combined
942 mask of O_RDONLY, O_WRONLY, and O_RDWR. The seek()/sysseek()
943 constants SEEK_SET, SEEK_CUR, and SEEK_END are available via the
944 C<:seek> tag. The chmod()/stat() S_IF* constants and S_IS* functions
945 are available via the C<:mode> tag.
949 A compare_text() function has been added, which allows custom
950 comparison functions. See L<File::Compare>.
954 File::Find now works correctly when the wanted() function is either
955 autoloaded or is a symbolic reference.
957 A bug that caused File::Find to lose track of the working directory
958 when pruning top-level directories has been fixed.
960 File::Find now also supports several other options to control its
961 behavior. It can follow symbolic links if the C<follow> option is
962 specified. Enabling the C<no_chdir> option will make File::Find skip
963 changing the current directory when walking directories. The C<untaint>
964 flag can be useful when running with taint checks enabled.
970 This extension implements BSD-style file globbing. By default,
971 it will also be used for the internal implementation of the glob()
972 operator. See L<File::Glob>.
976 New methods have been added to the File::Spec module: devnull() returns
977 the name of the null device (/dev/null on Unix) and tmpdir() the name of
978 the temp directory (normally /tmp on Unix). There are now also methods
979 to convert between absolute and relative filenames: abs2rel() and
980 rel2abs(). For compatibility with operating systems that specify volume
981 names in file paths, the splitpath(), splitdir(), and catdir() methods
984 =item File::Spec::Functions
986 The new File::Spec::Functions modules provides a function interface
987 to the File::Spec module. Allows shorthand
989 $fullname = catfile($dir1, $dir2, $file);
993 $fullname = File::Spec->catfile($dir1, $dir2, $file);
997 Getopt::Long licensing has changed to allow the Perl Artistic License
998 as well as the GPL. It used to be GPL only, which got in the way of
999 non-GPL applications that wanted to use Getopt::Long.
1001 Getopt::Long encourages the use of Pod::Usage to produce help
1002 messages. For example:
1008 GetOptions('help|?' => \$help, man => \$man) or pod2usage(2);
1009 pod2usage(1) if $help;
1010 pod2usage(-exitstatus => 0, -verbose => 2) if $man;
1016 sample - Using GetOpt::Long and Pod::Usage
1020 sample [options] [file ...]
1023 -help brief help message
1024 -man full documentation
1032 Print a brief help message and exits.
1036 Prints the manual page and exits.
1042 B<This program> will read the given input file(s) and do something
1043 useful with the contents thereof.
1047 See L<Pod::Usage> for details.
1049 A bug that prevented the non-option call-back <> from being
1050 specified as the first argument has been fixed.
1052 To specify the characters < and > as option starters, use ><. Note,
1053 however, that changing option starters is strongly deprecated.
1057 write() and syswrite() will now accept a single-argument
1058 form of the call, for consistency with Perl's syswrite().
1060 You can now create a TCP-based IO::Socket::INET without forcing
1061 a connect attempt. This allows you to configure its options
1062 (like making it non-blocking) and then call connect() manually.
1064 A bug that prevented the IO::Socket::protocol() accessor
1065 from ever returning the correct value has been corrected.
1067 IO::Socket::connect now uses non-blocking IO instead of alarm()
1068 to do connect timeouts.
1070 IO::Socket::accept now uses select() instead of alarm() for doing
1073 IO::Socket::INET->new now sets $! correctly on failure. $@ is
1074 still set for backwards compatibility.
1078 Java Perl Lingo is now distributed with Perl. See jpl/README
1079 for more information.
1083 C<use lib> now weeds out any trailing duplicate entries.
1084 C<no lib> removes all named entries.
1088 The bitwise operations C<<< << >>>, C<<< >> >>>, C<&>, C<|>,
1089 and C<~> are now supported on bigints.
1093 The accessor methods Re, Im, arg, abs, rho, and theta can now also
1094 act as mutators (accessor $z->Re(), mutator $z->Re(3)).
1096 The class method C<display_format> and the corresponding object method
1097 C<display_format>, in addition to accepting just one argument, now can
1098 also accept a parameter hash. Recognized keys of a parameter hash are
1099 C<"style">, which corresponds to the old one parameter case, and two
1100 new parameters: C<"format">, which is a printf()-style format string
1101 (defaults usually to C<"%.15g">, you can revert to the default by
1102 setting the format string to C<undef>) used for both parts of a
1103 complex number, and C<"polar_pretty_print"> (defaults to true),
1104 which controls whether an attempt is made to try to recognize small
1105 multiples and rationals of pi (2pi, pi/2) at the argument (angle) of a
1106 polar complex number.
1108 The potentially disruptive change is that in list context both methods
1109 now I<return the parameter hash>, instead of only the value of the
1110 C<"style"> parameter.
1114 A little bit of radial trigonometry (cylindrical and spherical),
1115 radial coordinate conversions, and the great circle distance were added.
1117 =item Pod::Parser, Pod::InputObjects
1119 Pod::Parser is a base class for parsing and selecting sections of
1120 pod documentation from an input stream. This module takes care of
1121 identifying pod paragraphs and commands in the input and hands off the
1122 parsed paragraphs and commands to user-defined methods which are free
1123 to interpret or translate them as they see fit.
1125 Pod::InputObjects defines some input objects needed by Pod::Parser, and
1126 for advanced users of Pod::Parser that need more about a command besides
1129 As of release 5.6.0 of Perl, Pod::Parser is now the officially sanctioned
1130 "base parser code" recommended for use by all pod2xxx translators.
1131 Pod::Text (pod2text) and Pod::Man (pod2man) have already been converted
1132 to use Pod::Parser and efforts to convert Pod::HTML (pod2html) are already
1133 underway. For any questions or comments about pod parsing and translating
1134 issues and utilities, please use the pod-people@perl.org mailing list.
1136 For further information, please see L<Pod::Parser> and L<Pod::InputObjects>.
1138 =item Pod::Checker, podchecker
1140 This utility checks pod files for correct syntax, according to
1141 L<perlpod>. Obvious errors are flagged as such, while warnings are
1142 printed for mistakes that can be handled gracefully. The checklist is
1143 not complete yet. See L<Pod::Checker>.
1145 =item Pod::ParseUtils, Pod::Find
1147 These modules provide a set of gizmos that are useful mainly for pod
1148 translators. L<Pod::Find|Pod::Find> traverses directory structures and
1149 returns found pod files, along with their canonical names (like
1150 C<File::Spec::Unix>). L<Pod::ParseUtils|Pod::ParseUtils> contains
1151 B<Pod::List> (useful for storing pod list information), B<Pod::Hyperlink>
1152 (for parsing the contents of C<LE<lt>E<gt>> sequences) and B<Pod::Cache>
1153 (for caching information about pod files, e.g., link nodes).
1155 =item Pod::Select, podselect
1157 Pod::Select is a subclass of Pod::Parser which provides a function
1158 named "podselect()" to filter out user-specified sections of raw pod
1159 documentation from an input stream. podselect is a script that provides
1160 access to Pod::Select from other scripts to be used as a filter.
1163 =item Pod::Usage, pod2usage
1165 Pod::Usage provides the function "pod2usage()" to print usage messages for
1166 a Perl script based on its embedded pod documentation. The pod2usage()
1167 function is generally useful to all script authors since it lets them
1168 write and maintain a single source (the pods) for documentation, thus
1169 removing the need to create and maintain redundant usage message text
1170 consisting of information already in the pods.
1172 There is also a pod2usage script which can be used from other kinds of
1173 scripts to print usage messages from pods (even for non-Perl scripts
1174 with pods embedded in comments).
1176 For details and examples, please see L<Pod::Usage>.
1178 =item Pod::Text and Pod::Man
1180 Pod::Text has been rewritten to use Pod::Parser. While pod2text() is
1181 still available for backwards compatibility, the module now has a new
1182 preferred interface. See L<Pod::Text> for the details. The new Pod::Text
1183 module is easily subclassed for tweaks to the output, and two such
1184 subclasses (Pod::Text::Termcap for man-page-style bold and underlining
1185 using termcap information, and Pod::Text::Color for markup with ANSI color
1186 sequences) are now standard.
1188 pod2man has been turned into a module, Pod::Man, which also uses
1189 Pod::Parser. In the process, several outstanding bugs related to quotes
1190 in section headers, quoting of code escapes, and nested lists have been
1191 fixed. pod2man is now a wrapper script around this module.
1195 An EXISTS method has been added to this module (and sdbm_exists() has
1196 been added to the underlying sdbm library), so one can now call exists
1197 on an SDBM_File tied hash and get the correct result, rather than a
1200 A bug that may have caused data loss when more than one disk block
1201 happens to be read from the database in a single FETCH() has been
1206 Sys::Syslog now uses XSUBs to access facilities from syslog.h so it
1207 no longer requires syslog.ph to exist.
1211 Sys::Hostname now uses XSUBs to call the C library's gethostname() or
1212 uname() if they exist.
1214 =item Term::ANSIColor
1216 Term::ANSIColor is a very simple module to provide easy and readable
1217 access to the ANSI color and highlighting escape sequences, supported by
1218 most ANSI terminal emulators. It is now included standard.
1222 The timelocal() and timegm() functions used to silently return bogus
1223 results when the date fell outside the machine's integer range. They
1224 now consistently croak() if the date falls in an unsupported range.
1228 The error return value in list context has been changed for all functions
1229 that return a list of values. Previously these functions returned a list
1230 with a single element C<undef> if an error occurred. Now these functions
1231 return the empty list in these situations. This applies to the following
1237 The remaining functions are unchanged and continue to return C<undef> on
1238 error even in list context.
1240 The Win32::SetLastError(ERROR) function has been added as a complement
1241 to the Win32::GetLastError() function.
1243 The new Win32::GetFullPathName(FILENAME) returns the full absolute
1244 pathname for FILENAME in scalar context. In list context it returns
1245 a two-element list containing the fully qualified directory name and
1246 the filename. See L<Win32>.
1250 The XSLoader extension is a simpler alternative to DynaLoader.
1255 A new feature called "DBM Filters" has been added to all the
1256 DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
1257 DBM Filters add four new methods to each DBM module:
1264 These can be used to filter key-value pairs before the pairs are
1265 written to the database or just after they are read from the database.
1266 See L<perldbmfilter> for further information.
1272 C<use attrs> is now obsolete, and is only provided for
1273 backward-compatibility. It's been replaced by the C<sub : attributes>
1274 syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>.
1276 Lexical warnings pragma, C<use warnings;>, to control optional warnings.
1279 C<use filetest> to control the behaviour of filetests (C<-r> C<-w>
1280 ...). Currently only one subpragma implemented, "use filetest
1281 'access';", that uses access(2) or equivalent to check permissions
1282 instead of using stat(2) as usual. This matters in filesystems
1283 where there are ACLs (access control lists): the stat(2) might lie,
1284 but access(2) knows better.
1286 The C<open> pragma can be used to specify default disciplines for
1287 handle constructors (e.g. open()) and for qx//. The two
1288 pseudo-disciplines C<:raw> and C<:crlf> are currently supported on
1289 DOS-derivative platforms (i.e. where binmode is not a no-op).
1290 See also L</"binmode() can be used to set :crlf and :raw modes">.
1292 =head1 Utility Changes
1296 C<dprofpp> is used to display profile data generated using C<Devel::DProf>.
1301 The C<find2perl> utility now uses the enhanced features of the File::Find
1302 module. The -depth and -follow options are supported. Pod documentation
1303 is also included in the script.
1307 The C<h2xs> tool can now work in conjunction with C<C::Scan> (available
1308 from CPAN) to automatically parse real-life header files. The C<-M>,
1309 C<-a>, C<-k>, and C<-o> options are new.
1313 C<perlcc> now supports the C and Bytecode backends. By default,
1314 it generates output from the simple C backend rather than the
1315 optimized C backend.
1317 Support for non-Unix platforms has been improved.
1321 C<perldoc> has been reworked to avoid possible security holes.
1322 It will not by default let itself be run as the superuser, but you
1323 may still use the B<-U> switch to try to make it drop privileges
1326 =head2 The Perl Debugger
1328 Many bug fixes and enhancements were added to F<perl5db.pl>, the
1329 Perl debugger. The help documentation was rearranged. New commands
1330 include C<< < ? >>, C<< > ? >>, and C<< { ? >> to list out current
1331 actions, C<man I<docpage>> to run your doc viewer on some perl
1332 docset, and support for quoted options. The help information was
1333 rearranged, and should be viewable once again if you're using B<less>
1334 as your pager. A serious security hole was plugged--you should
1335 immediately remove all older versions of the Perl debugger as
1336 installed in previous releases, all the way back to perl3, from
1337 your system to avoid being bitten by this.
1339 =head1 Improved Documentation
1341 Many of the platform-specific README files are now part of the perl
1342 installation. See L<perl> for the complete list.
1348 The official list of public Perl API functions.
1352 A tutorial for beginners on object-oriented Perl.
1354 =item perlcompile.pod
1356 An introduction to using the Perl Compiler suite.
1358 =item perldbmfilter.pod
1360 A howto document on using the DBM filter facility.
1364 All material unrelated to running the Perl debugger, plus all
1365 low-level guts-like details that risked crushing the casual user
1366 of the debugger, have been relocated from the old manpage to the
1369 =item perldebguts.pod
1371 This new manpage contains excessively low-level material not related
1372 to the Perl debugger, but slightly related to debugging Perl itself.
1373 It also contains some arcane internal details of how the debugging
1374 process works that may only be of interest to developers of Perl
1379 Notes on the fork() emulation currently available for the Windows platform.
1381 =item perlfilter.pod
1383 An introduction to writing Perl source filters.
1387 Some guidelines for hacking the Perl source code.
1389 =item perlintern.pod
1391 A list of internal functions in the Perl source code.
1392 (List is currently empty.)
1394 =item perllexwarn.pod
1396 Introduction and reference information about lexically scoped
1399 =item perlnumber.pod
1401 Detailed information about numbers as they are represented in Perl.
1403 =item perlopentut.pod
1405 A tutorial on using open() effectively.
1407 =item perlreftut.pod
1409 A tutorial that introduces the essentials of references.
1413 A tutorial on managing class data for object modules.
1417 Discussion of the most often wanted features that may someday be
1420 =item perlunicode.pod
1422 An introduction to Unicode support features in Perl.
1426 =head1 Performance enhancements
1428 =head2 Simple sort() using { $a <=> $b } and the like are optimized
1430 Many common sort() operations using a simple inlined block are now
1431 optimized for faster performance.
1433 =head2 Optimized assignments to lexical variables
1435 Certain operations in the RHS of assignment statements have been
1436 optimized to directly set the lexical variable on the LHS,
1437 eliminating redundant copying overheads.
1439 =head2 Faster subroutine calls
1441 Minor changes in how subroutine calls are handled internally
1442 provide marginal improvements in performance.
1444 =head2 delete(), each(), values() and hash iteration are faster
1446 The hash values returned by delete(), each(), values() and hashes in a
1447 list context are the actual values in the hash, instead of copies.
1448 This results in significantly better performance, because it eliminates
1449 needless copying in most situations.
1451 =head1 Installation and Configuration Improvements
1453 =head2 -Dusethreads means something different
1455 The -Dusethreads flag now enables the experimental interpreter-based thread
1456 support by default. To get the flavor of experimental threads that was in
1457 5.005 instead, you need to run Configure with "-Dusethreads -Duse5005threads".
1459 As of v5.6.0, interpreter-threads support is still lacking a way to
1460 create new threads from Perl (i.e., C<use Thread;> will not work with
1461 interpreter threads). C<use Thread;> continues to be available when you
1462 specify the -Duse5005threads option to Configure, bugs and all.
1464 NOTE: Support for threads continues to be an experimental feature.
1465 Interfaces and implementation are subject to sudden and drastic changes.
1467 =head2 New Configure flags
1469 The following new flags may be enabled on the Configure command line
1470 by running Configure with C<-Dflag>.
1473 usethreads useithreads (new interpreter threads: no Perl API yet)
1474 usethreads use5005threads (threads as they were in 5.005)
1476 use64bitint (equal to now deprecated 'use64bits')
1482 usesocks (only SOCKS v5 supported)
1484 =head2 Threadedness and 64-bitness now more daring
1486 The Configure options enabling the use of threads and the use of
1487 64-bitness are now more daring in the sense that they no more have an
1488 explicit list of operating systems of known threads/64-bit
1489 capabilities. In other words: if your operating system has the
1490 necessary APIs and datatypes, you should be able just to go ahead and
1491 use them, for threads by Configure -Dusethreads, and for 64 bits
1492 either explicitly by Configure -Duse64bitint or implicitly if your
1493 system has 64-bit wide datatypes. See also L<"64-bit support">.
1497 Some platforms have "long doubles", floating point numbers of even
1498 larger range than ordinary "doubles". To enable using long doubles for
1499 Perl's scalars, use -Duselongdouble.
1501 =head2 -Dusemorebits
1503 You can enable both -Duse64bitint and -Duselongdouble with -Dusemorebits.
1504 See also L<"64-bit support">.
1506 =head2 -Duselargefiles
1508 Some platforms support system APIs that are capable of handling large files
1509 (typically, files larger than two gigabytes). Perl will try to use these
1510 APIs if you ask for -Duselargefiles.
1512 See L<"Large file support"> for more information.
1514 =head2 installusrbinperl
1516 You can use "Configure -Uinstallusrbinperl" which causes installperl
1517 to skip installing perl also as /usr/bin/perl. This is useful if you
1518 prefer not to modify /usr/bin for some reason or another but harmful
1519 because many scripts assume to find Perl in /usr/bin/perl.
1521 =head2 SOCKS support
1523 You can use "Configure -Dusesocks" which causes Perl to probe
1524 for the SOCKS proxy protocol library (v5, not v4). For more information
1527 http://www.socks.nec.com/
1531 You can "post-edit" the Configure variables using the Configure C<-A>
1532 switch. The editing happens immediately after the platform specific
1533 hints files have been processed but before the actual configuration
1534 process starts. Run C<Configure -h> to find out the full C<-A> syntax.
1536 =head2 Enhanced Installation Directories
1538 The installation structure has been enriched to improve the support
1539 for maintaining multiple versions of perl, to provide locations for
1540 vendor-supplied modules, scripts, and manpages, and to ease maintenance
1541 of locally-added modules, scripts, and manpages. See the section on
1542 Installation Directories in the INSTALL file for complete details.
1543 For most users building and installing from source, the defaults should
1546 If you previously used C<Configure -Dsitelib> or C<-Dsitearch> to set
1547 special values for library directories, you might wish to consider using
1548 the new C<-Dsiteprefix> setting instead. Also, if you wish to re-use a
1549 config.sh file from an earlier version of perl, you should be sure to
1550 check that Configure makes sensible choices for the new directories.
1551 See INSTALL for complete details.
1553 =head1 Platform specific changes
1555 =head2 Supported platforms
1561 The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
1566 GNU/Hurd is now supported.
1570 Rhapsody/Darwin is now supported.
1574 EPOC is now supported (on Psion 5).
1578 The cygwin port (formerly cygwin32) has been greatly improved.
1588 Perl now works with djgpp 2.02 (and 2.03 alpha).
1592 Environment variable names are not converted to uppercase any more.
1596 Incorrect exit codes from backticks have been fixed.
1600 This port continues to use its own builtin globbing (not File::Glob).
1604 =head2 OS390 (OpenEdition MVS)
1606 Support for this EBCDIC platform has not been renewed in this release.
1607 There are difficulties in reconciling Perl's standardization on UTF-8
1608 as its internal representation for characters with the EBCDIC character
1609 set, because the two are incompatible.
1611 It is unclear whether future versions will renew support for this
1612 platform, but the possibility exists.
1616 Numerous revisions and extensions to configuration, build, testing, and
1617 installation process to accommodate core changes and VMS-specific options.
1619 Expand %ENV-handling code to allow runtime mapping to logical names,
1620 CLI symbols, and CRTL environ array.
1622 Extension of subprocess invocation code to accept filespecs as command
1625 Add to Perl command line processing the ability to use default file types and
1626 to recognize Unix-style C<2E<gt>&1>.
1628 Expansion of File::Spec::VMS routines, and integration into ExtUtils::MM_VMS.
1630 Extension of ExtUtils::MM_VMS to handle complex extensions more flexibly.
1632 Barewords at start of Unix-syntax paths may be treated as text rather than
1633 only as logical names.
1635 Optional secure translation of several logical names used internally by Perl.
1637 Miscellaneous bugfixing and porting of new core code to VMS.
1639 Thanks are gladly extended to the many people who have contributed VMS
1640 patches, testing, and ideas.
1644 Perl can now emulate fork() internally, using multiple interpreters running
1645 in different concurrent threads. This support must be enabled at build
1646 time. See L<perlfork> for detailed information.
1648 When given a pathname that consists only of a drivename, such as C<A:>,
1649 opendir() and stat() now use the current working directory for the drive
1650 rather than the drive root.
1652 The builtin XSUB functions in the Win32:: namespace are documented. See
1655 $^X now contains the full path name of the running executable.
1657 A Win32::GetLongPathName() function is provided to complement
1658 Win32::GetFullPathName() and Win32::GetShortPathName(). See L<Win32>.
1660 POSIX::uname() is supported.
1662 system(1,...) now returns true process IDs rather than process
1663 handles. kill() accepts any real process id, rather than strictly
1664 return values from system(1,...).
1666 For better compatibility with Unix, C<kill(0, $pid)> can now be used to
1667 test whether a process exists.
1669 The C<Shell> module is supported.
1671 Better support for building Perl under command.com in Windows 95
1674 Scripts are read in binary mode by default to allow ByteLoader (and
1675 the filter mechanism in general) to work properly. For compatibility,
1676 the DATA filehandle will be set to text mode if a carriage return is
1677 detected at the end of the line containing the __END__ or __DATA__
1678 token; if not, the DATA filehandle will be left open in binary mode.
1679 Earlier versions always opened the DATA filehandle in text mode.
1681 The glob() operator is implemented via the C<File::Glob> extension,
1682 which supports glob syntax of the C shell. This increases the flexibility
1683 of the glob() operator, but there may be compatibility issues for
1684 programs that relied on the older globbing syntax. If you want to
1685 preserve compatibility with the older syntax, you might want to run
1686 perl with C<-MFile::DosGlob>. For details and compatibility information,
1689 =head1 Significant bug fixes
1691 =head2 <HANDLE> on empty files
1693 With C<$/> set to C<undef>, "slurping" an empty file returns a string of
1694 zero length (instead of C<undef>, as it used to) the first time the
1695 HANDLE is read after C<$/> is set to C<undef>. Further reads yield
1698 This means that the following will append "foo" to an empty file (it used
1701 perl -0777 -pi -e 's/^/foo/' empty_file
1705 perl -pi -e 's/^/foo/' empty_file
1707 is unchanged (it continues to leave the file empty).
1709 =head2 C<eval '...'> improvements
1711 Line numbers (as reflected by caller() and most diagnostics) within
1712 C<eval '...'> were often incorrect where here documents were involved.
1713 This has been corrected.
1715 Lexical lookups for variables appearing in C<eval '...'> within
1716 functions that were themselves called within an C<eval '...'> were
1717 searching the wrong place for lexicals. The lexical search now
1718 correctly ends at the subroutine's block boundary.
1720 The use of C<return> within C<eval {...}> caused $@ not to be reset
1721 correctly when no exception occurred within the eval. This has
1724 Parsing of here documents used to be flawed when they appeared as
1725 the replacement expression in C<eval 's/.../.../e'>. This has
1728 =head2 All compilation errors are true errors
1730 Some "errors" encountered at compile time were by necessity
1731 generated as warnings followed by eventual termination of the
1732 program. This enabled more such errors to be reported in a
1733 single run, rather than causing a hard stop at the first error
1734 that was encountered.
1736 The mechanism for reporting such errors has been reimplemented
1737 to queue compile-time errors and report them at the end of the
1738 compilation as true errors rather than as warnings. This fixes
1739 cases where error messages leaked through in the form of warnings
1740 when code was compiled at run time using C<eval STRING>, and
1741 also allows such errors to be reliably trapped using C<eval "...">.
1743 =head2 Implicitly closed filehandles are safer
1745 Sometimes implicitly closed filehandles (as when they are localized,
1746 and Perl automatically closes them on exiting the scope) could
1747 inadvertently set $? or $!. This has been corrected.
1750 =head2 Behavior of list slices is more consistent
1752 When taking a slice of a literal list (as opposed to a slice of
1753 an array or hash), Perl used to return an empty list if the
1754 result happened to be composed of all undef values.
1756 The new behavior is to produce an empty list if (and only if)
1757 the original list was empty. Consider the following example:
1759 @a = (1,undef,undef,2)[2,1,2];
1761 The old behavior would have resulted in @a having no elements.
1762 The new behavior ensures it has three undefined elements.
1764 Note in particular that the behavior of slices of the following
1765 cases remains unchanged:
1768 @a = (getpwent)[7,0];
1769 @a = (anything_returning_empty_list())[2,1,2];
1771 @a = @c{'a','b','c'};
1775 =head2 C<(\$)> prototype and C<$foo{a}>
1777 A scalar reference prototype now correctly allows a hash or
1778 array element in that slot.
1780 =head2 C<goto &sub> and AUTOLOAD
1782 The C<goto &sub> construct works correctly when C<&sub> happens
1785 =head2 C<-bareword> allowed under C<use integer>
1787 The autoquoting of barewords preceded by C<-> did not work
1788 in prior versions when the C<integer> pragma was enabled.
1789 This has been fixed.
1791 =head2 Failures in DESTROY()
1793 When code in a destructor threw an exception, it went unnoticed
1794 in earlier versions of Perl, unless someone happened to be
1795 looking in $@ just after the point the destructor happened to
1796 run. Such failures are now visible as warnings when warnings are
1799 =head2 Locale bugs fixed
1801 printf() and sprintf() previously reset the numeric locale
1802 back to the default "C" locale. This has been fixed.
1804 Numbers formatted according to the local numeric locale
1805 (such as using a decimal comma instead of a decimal dot) caused
1806 "isn't numeric" warnings, even while the operations accessing
1807 those numbers produced correct results. These warnings have been
1812 The C<eval 'return sub {...}'> construct could sometimes leak
1813 memory. This has been fixed.
1815 Operations that aren't filehandle constructors used to leak memory
1816 when used on invalid filehandles. This has been fixed.
1818 Constructs that modified C<@_> could fail to deallocate values
1819 in C<@_> and thus leak memory. This has been corrected.
1821 =head2 Spurious subroutine stubs after failed subroutine calls
1823 Perl could sometimes create empty subroutine stubs when a
1824 subroutine was not found in the package. Such cases stopped
1825 later method lookups from progressing into base packages.
1826 This has been corrected.
1828 =head2 Taint failures under C<-U>
1830 When running in unsafe mode, taint violations could sometimes
1831 cause silent failures. This has been fixed.
1833 =head2 END blocks and the C<-c> switch
1835 Prior versions used to run BEGIN B<and> END blocks when Perl was
1836 run in compile-only mode. Since this is typically not the expected
1837 behavior, END blocks are not executed anymore when the C<-c> switch
1838 is used, or if compilation fails.
1840 See L</"Support for CHECK blocks"> for how to run things when the compile
1843 =head2 Potential to leak DATA filehandles
1845 Using the C<__DATA__> token creates an implicit filehandle to
1846 the file that contains the token. It is the program's
1847 responsibility to close it when it is done reading from it.
1849 This caveat is now better explained in the documentation.
1852 =head1 New or Changed Diagnostics
1856 =item "%s" variable %s masks earlier declaration in same %s
1858 (W misc) A "my" or "our" variable has been redeclared in the current scope or statement,
1859 effectively eliminating all access to the previous instance. This is almost
1860 always a typographical error. Note that the earlier variable will still exist
1861 until the end of the scope or until all closure referents to it are
1864 =item "my sub" not yet implemented
1866 (F) Lexically scoped subroutines are not yet implemented. Don't try that
1869 =item "our" variable %s redeclared
1871 (W misc) You seem to have already declared the same global once before in the
1872 current lexical scope.
1874 =item '!' allowed only after types %s
1876 (F) The '!' is allowed in pack() and unpack() only after certain types.
1877 See L<perlfunc/pack>.
1879 =item / cannot take a count
1881 (F) You had an unpack template indicating a counted-length string,
1882 but you have also specified an explicit size for the string.
1883 See L<perlfunc/pack>.
1885 =item / must be followed by a, A or Z
1887 (F) You had an unpack template indicating a counted-length string,
1888 which must be followed by one of the letters a, A or Z
1889 to indicate what sort of string is to be unpacked.
1890 See L<perlfunc/pack>.
1892 =item / must be followed by a*, A* or Z*
1894 (F) You had a pack template indicating a counted-length string,
1895 Currently the only things that can have their length counted are a*, A* or Z*.
1896 See L<perlfunc/pack>.
1898 =item / must follow a numeric type
1900 (F) You had an unpack template that contained a '#',
1901 but this did not follow some numeric unpack specification.
1902 See L<perlfunc/pack>.
1904 =item /%s/: Unrecognized escape \\%c passed through
1906 (W regexp) You used a backslash-character combination which is not recognized
1907 by Perl. This combination appears in an interpolated variable or a
1908 C<'>-delimited regular expression. The character was understood literally.
1910 =item /%s/: Unrecognized escape \\%c in character class passed through
1912 (W regexp) You used a backslash-character combination which is not recognized
1913 by Perl inside character classes. The character was understood literally.
1915 =item /%s/ should probably be written as "%s"
1917 (W syntax) You have used a pattern where Perl expected to find a string,
1918 as in the first argument to C<join>. Perl will treat the true
1919 or false result of matching the pattern against $_ as the string,
1920 which is probably not what you had in mind.
1922 =item %s() called too early to check prototype
1924 (W prototype) You've called a function that has a prototype before the parser saw a
1925 definition or declaration for it, and Perl could not check that the call
1926 conforms to the prototype. You need to either add an early prototype
1927 declaration for the subroutine in question, or move the subroutine
1928 definition ahead of the call to get proper prototype checking. Alternatively,
1929 if you are certain that you're calling the function correctly, you may put
1930 an ampersand before the name to avoid the warning. See L<perlsub>.
1932 =item %s argument is not a HASH or ARRAY element
1934 (F) The argument to exists() must be a hash or array element, such as:
1939 =item %s argument is not a HASH or ARRAY element or slice
1941 (F) The argument to delete() must be either a hash or array element, such as:
1946 or a hash or array slice, such as:
1948 @foo[$bar, $baz, $xyzzy]
1949 @{$ref->[12]}{"susie", "queue"}
1951 =item %s argument is not a subroutine name
1953 (F) The argument to exists() for C<exists &sub> must be a subroutine
1954 name, and not a subroutine call. C<exists &sub()> will generate this error.
1956 =item %s package attribute may clash with future reserved word: %s
1958 (W reserved) A lowercase attribute name was used that had a package-specific handler.
1959 That name might have a meaning to Perl itself some day, even though it
1960 doesn't yet. Perhaps you should use a mixed-case attribute name, instead.
1963 =item (in cleanup) %s
1965 (W misc) This prefix usually indicates that a DESTROY() method raised
1966 the indicated exception. Since destructors are usually called by
1967 the system at arbitrary points during execution, and often a vast
1968 number of times, the warning is issued only once for any number
1969 of failures that would otherwise result in the same message being
1972 Failure of user callbacks dispatched using the C<G_KEEPERR> flag
1973 could also result in this warning. See L<perlcall/G_KEEPERR>.
1975 =item <> should be quotes
1977 (F) You wrote C<< require <file> >> when you should have written
1980 =item Attempt to join self
1982 (F) You tried to join a thread from within itself, which is an
1983 impossible task. You may be joining the wrong thread, or you may
1984 need to move the join() to some other thread.
1986 =item Bad evalled substitution pattern
1988 (F) You've used the /e switch to evaluate the replacement for a
1989 substitution, but perl found a syntax error in the code to evaluate,
1990 most likely an unexpected right brace '}'.
1992 =item Bad realloc() ignored
1994 (S) An internal routine called realloc() on something that had never been
1995 malloc()ed in the first place. Mandatory, but can be disabled by
1996 setting environment variable C<PERL_BADFREE> to 1.
1998 =item Bareword found in conditional
2000 (W bareword) The compiler found a bareword where it expected a conditional,
2001 which often indicates that an || or && was parsed as part of the
2002 last argument of the previous construct, for example:
2006 It may also indicate a misspelled constant that has been interpreted
2009 use constant TYPO => 1;
2010 if (TYOP) { print "foo" }
2012 The C<strict> pragma is useful in avoiding such errors.
2014 =item Binary number > 0b11111111111111111111111111111111 non-portable
2016 (W portable) The binary number you specified is larger than 2**32-1
2017 (4294967295) and therefore non-portable between systems. See
2018 L<perlport> for more on portability concerns.
2020 =item Bit vector size > 32 non-portable
2022 (W portable) Using bit vector sizes larger than 32 is non-portable.
2024 =item Buffer overflow in prime_env_iter: %s
2026 (W internal) A warning peculiar to VMS. While Perl was preparing to iterate over
2027 %ENV, it encountered a logical name or symbol definition which was too long,
2028 so it was truncated to the string shown.
2030 =item Can't check filesystem of script "%s"
2032 (P) For some reason you can't check the filesystem of the script for nosuid.
2034 =item Can't declare class for non-scalar %s in "%s"
2036 (S) Currently, only scalar variables can declared with a specific class
2037 qualifier in a "my" or "our" declaration. The semantics may be extended
2038 for other types of variables in future.
2040 =item Can't declare %s in "%s"
2042 (F) Only scalar, array, and hash variables may be declared as "my" or
2043 "our" variables. They must have ordinary identifiers as names.
2045 =item Can't ignore signal CHLD, forcing to default
2047 (W signal) Perl has detected that it is being run with the SIGCHLD signal
2048 (sometimes known as SIGCLD) disabled. Since disabling this signal
2049 will interfere with proper determination of exit status of child
2050 processes, Perl has reset the signal to its default value.
2051 This situation typically indicates that the parent program under
2052 which Perl may be running (e.g., cron) is being very careless.
2054 =item Can't modify non-lvalue subroutine call
2056 (F) Subroutines meant to be used in lvalue context should be declared as
2057 such, see L<perlsub/"Lvalue subroutines">.
2059 =item Can't read CRTL environ
2061 (S) A warning peculiar to VMS. Perl tried to read an element of %ENV
2062 from the CRTL's internal environment array and discovered the array was
2063 missing. You need to figure out where your CRTL misplaced its environ
2064 or define F<PERL_ENV_TABLES> (see L<perlvms>) so that environ is not searched.
2066 =item Can't remove %s: %s, skipping file
2068 (S) You requested an inplace edit without creating a backup file. Perl
2069 was unable to remove the original file to replace it with the modified
2070 file. The file was left unmodified.
2072 =item Can't return %s from lvalue subroutine
2074 (F) Perl detected an attempt to return illegal lvalues (such
2075 as temporary or readonly values) from a subroutine used as an lvalue.
2076 This is not allowed.
2078 =item Can't weaken a nonreference
2080 (F) You attempted to weaken something that was not a reference. Only
2081 references can be weakened.
2083 =item Character class [:%s:] unknown
2085 (F) The class in the character class [: :] syntax is unknown.
2088 =item Character class syntax [%s] belongs inside character classes
2090 (W unsafe) The character class constructs [: :], [= =], and [. .] go
2091 I<inside> character classes, the [] are part of the construct,
2092 for example: /[012[:alpha:]345]/. Note that [= =] and [. .]
2093 are not currently implemented; they are simply placeholders for
2096 =item Constant is not %s reference
2098 (F) A constant value (perhaps declared using the C<use constant> pragma)
2099 is being dereferenced, but it amounts to the wrong type of reference. The
2100 message indicates the type of reference that was expected. This usually
2101 indicates a syntax error in dereferencing the constant value.
2102 See L<perlsub/"Constant Functions"> and L<constant>.
2104 =item constant(%s): %s
2106 (F) The parser found inconsistencies either while attempting to define an
2107 overloaded constant, or when trying to find the character name specified
2108 in the C<\N{...}> escape. Perhaps you forgot to load the corresponding
2109 C<overload> or C<charnames> pragma? See L<charnames> and L<overload>.
2111 =item CORE::%s is not a keyword
2113 (F) The CORE:: namespace is reserved for Perl keywords.
2115 =item defined(@array) is deprecated
2117 (D) defined() is not usually useful on arrays because it checks for an
2118 undefined I<scalar> value. If you want to see if the array is empty,
2119 just use C<if (@array) { # not empty }> for example.
2121 =item defined(%hash) is deprecated
2123 (D) defined() is not usually useful on hashes because it checks for an
2124 undefined I<scalar> value. If you want to see if the hash is empty,
2125 just use C<if (%hash) { # not empty }> for example.
2127 =item Did not produce a valid header
2131 =item (Did you mean "local" instead of "our"?)
2133 (W misc) Remember that "our" does not localize the declared global variable.
2134 You have declared it again in the same lexical scope, which seems superfluous.
2136 =item Document contains no data
2140 =item entering effective %s failed
2142 (F) While under the C<use filetest> pragma, switching the real and
2143 effective uids or gids failed.
2145 =item false [] range "%s" in regexp
2147 (W regexp) A character class range must start and end at a literal character, not
2148 another character class like C<\d> or C<[:alpha:]>. The "-" in your false
2149 range is interpreted as a literal "-". Consider quoting the "-", "\-".
2152 =item Filehandle %s opened only for output
2154 (W io) You tried to read from a filehandle opened only for writing. If you
2155 intended it to be a read/write filehandle, you needed to open it with
2156 "+<" or "+>" or "+>>" instead of with "<" or nothing. If
2157 you intended only to read from the file, use "<". See
2160 =item flock() on closed filehandle %s
2162 (W closed) The filehandle you're attempting to flock() got itself closed some
2163 time before now. Check your logic flow. flock() operates on filehandles.
2164 Are you attempting to call flock() on a dirhandle by the same name?
2166 =item Global symbol "%s" requires explicit package name
2168 (F) You've said "use strict vars", which indicates that all variables
2169 must either be lexically scoped (using "my"), declared beforehand using
2170 "our", or explicitly qualified to say which package the global variable
2173 =item Hexadecimal number > 0xffffffff non-portable
2175 (W portable) The hexadecimal number you specified is larger than 2**32-1
2176 (4294967295) and therefore non-portable between systems. See
2177 L<perlport> for more on portability concerns.
2179 =item Ill-formed CRTL environ value "%s"
2181 (W internal) A warning peculiar to VMS. Perl tried to read the CRTL's internal
2182 environ array, and encountered an element without the C<=> delimiter
2183 used to separate keys from values. The element is ignored.
2185 =item Ill-formed message in prime_env_iter: |%s|
2187 (W internal) A warning peculiar to VMS. Perl tried to read a logical name
2188 or CLI symbol definition when preparing to iterate over %ENV, and
2189 didn't see the expected delimiter between key and value, so the
2192 =item Illegal binary digit %s
2194 (F) You used a digit other than 0 or 1 in a binary number.
2196 =item Illegal binary digit %s ignored
2198 (W digit) You may have tried to use a digit other than 0 or 1 in a binary number.
2199 Interpretation of the binary number stopped before the offending digit.
2201 =item Illegal number of bits in vec
2203 (F) The number of bits in vec() (the third argument) must be a power of
2204 two from 1 to 32 (or 64, if your platform supports that).
2206 =item Integer overflow in %s number
2208 (W overflow) The hexadecimal, octal or binary number you have specified either
2209 as a literal or as an argument to hex() or oct() is too big for your
2210 architecture, and has been converted to a floating point number. On a
2211 32-bit architecture the largest hexadecimal, octal or binary number
2212 representable without overflow is 0xFFFFFFFF, 037777777777, or
2213 0b11111111111111111111111111111111 respectively. Note that Perl
2214 transparently promotes all numbers to a floating point representation
2215 internally--subject to loss of precision errors in subsequent
2218 =item Invalid %s attribute: %s
2220 The indicated attribute for a subroutine or variable was not recognized
2221 by Perl or by a user-supplied handler. See L<attributes>.
2223 =item Invalid %s attributes: %s
2225 The indicated attributes for a subroutine or variable were not recognized
2226 by Perl or by a user-supplied handler. See L<attributes>.
2228 =item invalid [] range "%s" in regexp
2230 The offending range is now explicitly displayed.
2232 =item Invalid separator character %s in attribute list
2234 (F) Something other than a colon or whitespace was seen between the
2235 elements of an attribute list. If the previous attribute
2236 had a parenthesised parameter list, perhaps that list was terminated
2237 too soon. See L<attributes>.
2239 =item Invalid separator character %s in subroutine attribute list
2241 (F) Something other than a colon or whitespace was seen between the
2242 elements of a subroutine attribute list. If the previous attribute
2243 had a parenthesised parameter list, perhaps that list was terminated
2246 =item leaving effective %s failed
2248 (F) While under the C<use filetest> pragma, switching the real and
2249 effective uids or gids failed.
2251 =item Lvalue subs returning %s not implemented yet
2253 (F) Due to limitations in the current implementation, array and hash
2254 values cannot be returned in subroutines used in lvalue context.
2255 See L<perlsub/"Lvalue subroutines">.
2257 =item Method %s not permitted
2261 =item Missing %sbrace%s on \N{}
2263 (F) Wrong syntax of character name literal C<\N{charname}> within
2264 double-quotish context.
2266 =item Missing command in piped open
2268 (W pipe) You used the C<open(FH, "| command")> or C<open(FH, "command |")>
2269 construction, but the command was missing or blank.
2271 =item Missing name in "my sub"
2273 (F) The reserved syntax for lexically scoped subroutines requires that they
2274 have a name with which they can be found.
2276 =item No %s specified for -%c
2278 (F) The indicated command line switch needs a mandatory argument, but
2279 you haven't specified one.
2281 =item No package name allowed for variable %s in "our"
2283 (F) Fully qualified variable names are not allowed in "our" declarations,
2284 because that doesn't make much sense under existing semantics. Such
2285 syntax is reserved for future extensions.
2287 =item No space allowed after -%c
2289 (F) The argument to the indicated command line switch must follow immediately
2290 after the switch, without intervening spaces.
2292 =item no UTC offset information; assuming local time is UTC
2294 (S) A warning peculiar to VMS. Perl was unable to find the local
2295 timezone offset, so it's assuming that local system time is equivalent
2296 to UTC. If it's not, define the logical name F<SYS$TIMEZONE_DIFFERENTIAL>
2297 to translate to the number of seconds which need to be added to UTC to
2300 =item Octal number > 037777777777 non-portable
2302 (W portable) The octal number you specified is larger than 2**32-1 (4294967295)
2303 and therefore non-portable between systems. See L<perlport> for more
2304 on portability concerns.
2306 See also L<perlport> for writing portable code.
2308 =item panic: del_backref
2310 (P) Failed an internal consistency check while trying to reset a weak
2313 =item panic: kid popen errno read
2315 (F) forked child returned an incomprehensible message about its errno.
2317 =item panic: magic_killbackrefs
2319 (P) Failed an internal consistency check while trying to reset all weak
2320 references to an object.
2322 =item Parentheses missing around "%s" list
2324 (W parenthesis) You said something like
2330 my ($foo, $bar) = @_;
2332 Remember that "my", "our", and "local" bind tighter than comma.
2334 =item Possible unintended interpolation of %s in string
2336 (W ambiguous) It used to be that Perl would try to guess whether you
2337 wanted an array interpolated or a literal @. It no longer does this;
2338 arrays are now I<always> interpolated into strings. This means that
2339 if you try something like:
2341 print "fred@example.com";
2343 and the array C<@example> doesn't exist, Perl is going to print
2344 C<fred.com>, which is probably not what you wanted. To get a literal
2345 C<@> sign in a string, put a backslash before it, just as you would
2346 to get a literal C<$> sign.
2348 =item Possible Y2K bug: %s
2350 (W y2k) You are concatenating the number 19 with another number, which
2351 could be a potential Year 2000 problem.
2353 =item pragma "attrs" is deprecated, use "sub NAME : ATTRS" instead
2355 (W deprecated) You have written something like this:
2359 use attrs qw(locked);
2362 You should use the new declaration syntax instead.
2368 The C<use attrs> pragma is now obsolete, and is only provided for
2369 backward-compatibility. See L<perlsub/"Subroutine Attributes">.
2372 =item Premature end of script headers
2376 =item Repeat count in pack overflows
2378 (F) You can't specify a repeat count so large that it overflows
2379 your signed integers. See L<perlfunc/pack>.
2381 =item Repeat count in unpack overflows
2383 (F) You can't specify a repeat count so large that it overflows
2384 your signed integers. See L<perlfunc/unpack>.
2386 =item realloc() of freed memory ignored
2388 (S) An internal routine called realloc() on something that had already
2391 =item Reference is already weak
2393 (W misc) You have attempted to weaken a reference that is already weak.
2394 Doing so has no effect.
2396 =item setpgrp can't take arguments
2398 (F) Your system has the setpgrp() from BSD 4.2, which takes no arguments,
2399 unlike POSIX setpgid(), which takes a process ID and process group ID.
2401 =item Strange *+?{} on zero-length expression
2403 (W regexp) You applied a regular expression quantifier in a place where it
2404 makes no sense, such as on a zero-width assertion.
2405 Try putting the quantifier inside the assertion instead. For example,
2406 the way to match "abc" provided that it is followed by three
2407 repetitions of "xyz" is C</abc(?=(?:xyz){3})/>, not C</abc(?=xyz){3}/>.
2409 =item switching effective %s is not implemented
2411 (F) While under the C<use filetest> pragma, we cannot switch the
2412 real and effective uids or gids.
2414 =item This Perl can't reset CRTL environ elements (%s)
2416 =item This Perl can't set CRTL environ elements (%s=%s)
2418 (W internal) Warnings peculiar to VMS. You tried to change or delete an element
2419 of the CRTL's internal environ array, but your copy of Perl wasn't
2420 built with a CRTL that contained the setenv() function. You'll need to
2421 rebuild Perl with a CRTL that does, or redefine F<PERL_ENV_TABLES> (see
2422 L<perlvms>) so that the environ array isn't the target of the change to
2423 %ENV which produced the warning.
2425 =item Too late to run %s block
2427 (W void) A CHECK or INIT block is being defined during run time proper,
2428 when the opportunity to run them has already passed. Perhaps you are
2429 loading a file with C<require> or C<do> when you should be using
2430 C<use> instead. Or perhaps you should put the C<require> or C<do>
2431 inside a BEGIN block.
2433 =item Unknown open() mode '%s'
2435 (F) The second argument of 3-argument open() is not among the list
2436 of valid modes: C<< < >>, C<< > >>, C<<< >> >>>, C<< +< >>,
2437 C<< +> >>, C<<< +>> >>>, C<-|>, C<|->.
2439 =item Unknown process %x sent message to prime_env_iter: %s
2441 (P) An error peculiar to VMS. Perl was reading values for %ENV before
2442 iterating over it, and someone else stuck a message in the stream of
2443 data Perl expected. Someone's very confused, or perhaps trying to
2444 subvert Perl's population of %ENV for nefarious purposes.
2446 =item Unrecognized escape \\%c passed through
2448 (W misc) You used a backslash-character combination which is not recognized
2449 by Perl. The character was understood literally.
2451 =item Unterminated attribute parameter in attribute list
2453 (F) The lexer saw an opening (left) parenthesis character while parsing an
2454 attribute list, but the matching closing (right) parenthesis
2455 character was not found. You may need to add (or remove) a backslash
2456 character to get your parentheses to balance. See L<attributes>.
2458 =item Unterminated attribute list
2460 (F) The lexer found something other than a simple identifier at the start
2461 of an attribute, and it wasn't a semicolon or the start of a
2462 block. Perhaps you terminated the parameter list of the previous attribute
2463 too soon. See L<attributes>.
2465 =item Unterminated attribute parameter in subroutine attribute list
2467 (F) The lexer saw an opening (left) parenthesis character while parsing a
2468 subroutine attribute list, but the matching closing (right) parenthesis
2469 character was not found. You may need to add (or remove) a backslash
2470 character to get your parentheses to balance.
2472 =item Unterminated subroutine attribute list
2474 (F) The lexer found something other than a simple identifier at the start
2475 of a subroutine attribute, and it wasn't a semicolon or the start of a
2476 block. Perhaps you terminated the parameter list of the previous attribute
2479 =item Value of CLI symbol "%s" too long
2481 (W misc) A warning peculiar to VMS. Perl tried to read the value of an %ENV
2482 element from a CLI symbol table, and found a resultant string longer
2483 than 1024 characters. The return value has been truncated to 1024
2486 =item Version number must be a constant number
2488 (P) The attempt to translate a C<use Module n.n LIST> statement into
2489 its equivalent C<BEGIN> block found an internal inconsistency with
2500 Compatibility tests for C<sub : attrs> vs the older C<use attrs>.
2504 Tests for new environment scalar capability (e.g., C<use Env qw($BAR);>).
2508 Tests for new environment array capability (e.g., C<use Env qw(@PATH);>).
2512 IO constants (SEEK_*, _IO*).
2516 Directory-related IO methods (new, read, close, rewind, tied delete).
2518 =item lib/io_multihomed
2520 INET sockets with multi-homed hosts.
2532 Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>.
2536 File test operators.
2540 Verify operations that access pad objects (lexicals and temporaries).
2544 Verify C<exists &sub> operations.
2548 =head1 Incompatible Changes
2550 =head2 Perl Source Incompatibilities
2552 Beware that any new warnings that have been added or old ones
2553 that have been enhanced are B<not> considered incompatible changes.
2555 Since all new warnings must be explicitly requested via the C<-w>
2556 switch or the C<warnings> pragma, it is ultimately the programmer's
2557 responsibility to ensure that warnings are enabled judiciously.
2561 =item CHECK is a new keyword
2563 All subroutine definitions named CHECK are now special. See
2564 C</"Support for CHECK blocks"> for more information.
2566 =item Treatment of list slices of undef has changed
2568 There is a potential incompatibility in the behavior of list slices
2569 that are comprised entirely of undefined values.
2570 See L</"Behavior of list slices is more consistent">.
2572 =item Format of $English::PERL_VERSION is different
2574 The English module now sets $PERL_VERSION to $^V (a string value) rather
2575 than C<$]> (a numeric value). This is a potential incompatibility.
2576 Send us a report via perlbug if you are affected by this.
2578 See L</"Improved Perl version numbering system"> for the reasons for
2581 =item Literals of the form C<1.2.3> parse differently
2583 Previously, numeric literals with more than one dot in them were
2584 interpreted as a floating point number concatenated with one or more
2585 numbers. Such "numbers" are now parsed as strings composed of the
2588 For example, C<print 97.98.99> used to output C<97.9899> in earlier
2589 versions, but now prints C<abc>.
2591 See L</"Support for strings represented as a vector of ordinals">.
2593 =item Possibly changed pseudo-random number generator
2595 Perl programs that depend on reproducing a specific set of pseudo-random
2596 numbers may now produce different output due to improvements made to the
2597 rand() builtin. You can use C<sh Configure -Drandfunc=rand> to obtain
2600 See L</"Better pseudo-random number generator">.
2602 =item Hashing function for hash keys has changed
2604 Even though Perl hashes are not order preserving, the apparently
2605 random order encountered when iterating on the contents of a hash
2606 is actually determined by the hashing algorithm used. Improvements
2607 in the algorithm may yield a random order that is B<different> from
2608 that of previous versions, especially when iterating on hashes.
2610 See L</"Better worst-case behavior of hashes"> for additional
2613 =item C<undef> fails on read only values
2615 Using the C<undef> operator on a readonly value (such as $1) has
2616 the same effect as assigning C<undef> to the readonly value--it
2617 throws an exception.
2619 =item Close-on-exec bit may be set on pipe and socket handles
2621 Pipe and socket handles are also now subject to the close-on-exec
2622 behavior determined by the special variable $^F.
2624 See L</"More consistent close-on-exec behavior">.
2626 =item Writing C<"$$1"> to mean C<"${$}1"> is unsupported
2628 Perl 5.004 deprecated the interpretation of C<$$1> and
2629 similar within interpolated strings to mean C<$$ . "1">,
2630 but still allowed it.
2632 In Perl 5.6.0 and later, C<"$$1"> always means C<"${$1}">.
2634 =item delete(), values() and C<\(%h)> operate on aliases to values, not copies
2636 delete(), each(), values() and hashes in a list context return the actual
2637 values in the hash, instead of copies (as they used to in earlier
2638 versions). Typical idioms for using these constructs copy the
2639 returned values, but this can make a significant difference when
2640 creating references to the returned values. Keys in the hash are still
2641 returned as copies when iterating on a hash.
2643 See also L</"delete(), each(), values() and hash iteration are faster">.
2645 =item vec(EXPR,OFFSET,BITS) enforces powers-of-two BITS
2647 vec() generates a run-time error if the BITS argument is not
2648 a valid power-of-two integer.
2650 =item Text of some diagnostic output has changed
2652 Most references to internal Perl operations in diagnostics
2653 have been changed to be more descriptive. This may be an
2654 issue for programs that may incorrectly rely on the exact
2655 text of diagnostics for proper functioning.
2657 =item C<%@> has been removed
2659 The undocumented special variable C<%@> that used to accumulate
2660 "background" errors (such as those that happen in DESTROY())
2661 has been removed, because it could potentially result in memory
2664 =item Parenthesized not() behaves like a list operator
2666 The C<not> operator now falls under the "if it looks like a function,
2667 it behaves like a function" rule.
2669 As a result, the parenthesized form can be used with C<grep> and C<map>.
2670 The following construct used to be a syntax error before, but it works
2673 grep not($_), @things;
2675 On the other hand, using C<not> with a literal list slice may not
2676 work. The following previously allowed construct:
2678 print not (1,2,3)[0];
2680 needs to be written with additional parentheses now:
2682 print not((1,2,3)[0]);
2684 The behavior remains unaffected when C<not> is not followed by parentheses.
2686 =item Semantics of bareword prototype C<(*)> have changed
2688 The semantics of the bareword prototype C<*> have changed. Perl 5.005
2689 always coerced simple scalar arguments to a typeglob, which wasn't useful
2690 in situations where the subroutine must distinguish between a simple
2691 scalar and a typeglob. The new behavior is to not coerce bareword
2692 arguments to a typeglob. The value will always be visible as either
2693 a simple scalar or as a reference to a typeglob.
2695 See L</"More functional bareword prototype (*)">.
2697 =item Semantics of bit operators may have changed on 64-bit platforms
2699 If your platform is either natively 64-bit or if Perl has been
2700 configured to used 64-bit integers, i.e., $Config{ivsize} is 8,
2701 there may be a potential incompatibility in the behavior of bitwise
2702 numeric operators (& | ^ ~ << >>). These operators used to strictly
2703 operate on the lower 32 bits of integers in previous versions, but now
2704 operate over the entire native integral width. In particular, note
2705 that unary C<~> will produce different results on platforms that have
2706 different $Config{ivsize}. For portability, be sure to mask off
2707 the excess bits in the result of unary C<~>, e.g., C<~$x & 0xffffffff>.
2709 See L</"Bit operators support full native integer width">.
2711 =item More builtins taint their results
2713 As described in L</"Improved security features">, there may be more
2714 sources of taint in a Perl program.
2716 To avoid these new tainting behaviors, you can build Perl with the
2717 Configure option C<-Accflags=-DINCOMPLETE_TAINTS>. Beware that the
2718 ensuing perl binary may be insecure.
2722 =head2 C Source Incompatibilities
2726 =item C<PERL_POLLUTE>
2728 Release 5.005 grandfathered old global symbol names by providing preprocessor
2729 macros for extension source compatibility. As of release 5.6.0, these
2730 preprocessor definitions are not available by default. You need to explicitly
2731 compile perl with C<-DPERL_POLLUTE> to get these definitions. For
2732 extensions still using the old symbols, this option can be
2733 specified via MakeMaker:
2735 perl Makefile.PL POLLUTE=1
2737 =item C<PERL_IMPLICIT_CONTEXT>
2739 This new build option provides a set of macros for all API functions
2740 such that an implicit interpreter/thread context argument is passed to
2741 every API function. As a result of this, something like C<sv_setsv(foo,bar)>
2742 amounts to a macro invocation that actually translates to something like
2743 C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
2744 to not have any significant source compatibility issues, the difference
2745 between a macro and a real function call will need to be considered.
2747 This means that there B<is> a source compatibility issue as a result of
2748 this if your extensions attempt to use pointers to any of the Perl API
2751 Note that the above issue is not relevant to the default build of
2752 Perl, whose interfaces continue to match those of prior versions
2753 (but subject to the other options described here).
2755 See L<perlguts/"The Perl API"> for detailed information on the
2756 ramifications of building Perl with this option.
2758 NOTE: PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built
2759 with one of -Dusethreads, -Dusemultiplicity, or both. It is not
2760 intended to be enabled by users at this time.
2762 =item C<PERL_POLLUTE_MALLOC>
2764 Enabling Perl's malloc in release 5.005 and earlier caused the namespace of
2765 the system's malloc family of functions to be usurped by the Perl versions,
2766 since by default they used the same names. Besides causing problems on
2767 platforms that do not allow these functions to be cleanly replaced, this
2768 also meant that the system versions could not be called in programs that
2769 used Perl's malloc. Previous versions of Perl have allowed this behaviour
2770 to be suppressed with the HIDEMYMALLOC and EMBEDMYMALLOC preprocessor
2773 As of release 5.6.0, Perl's malloc family of functions have default names
2774 distinct from the system versions. You need to explicitly compile perl with
2775 C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
2776 and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
2779 Note that these functions do B<not> constitute Perl's memory allocation API.
2780 See L<perlguts/"Memory Allocation"> for further information about that.
2784 =head2 Compatible C Source API Changes
2788 =item C<PATCHLEVEL> is now C<PERL_VERSION>
2790 The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
2791 are now available by default from perl.h, and reflect the base revision,
2792 patchlevel, and subversion respectively. C<PERL_REVISION> had no
2793 prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
2794 previously available as C<PATCHLEVEL> and C<SUBVERSION>.
2796 The new names cause less pollution of the B<cpp> namespace and reflect what
2797 the numbers have come to stand for in common practice. For compatibility,
2798 the old names are still supported when F<patchlevel.h> is explicitly
2799 included (as required before), so there is no source incompatibility
2804 =head2 Binary Incompatibilities
2806 In general, the default build of this release is expected to be binary
2807 compatible for extensions built with the 5.005 release or its maintenance
2808 versions. However, specific platforms may have broken binary compatibility
2809 due to changes in the defaults used in hints files. Therefore, please be
2810 sure to always check the platform-specific README files for any notes to
2813 The usethreads or usemultiplicity builds are B<not> binary compatible
2814 with the corresponding builds in 5.005.
2816 On platforms that require an explicit list of exports (AIX, OS/2 and Windows,
2817 among others), purely internal symbols such as parser functions and the
2818 run time opcodes are not exported by default. Perl 5.005 used to export
2819 all functions irrespective of whether they were considered part of the
2822 For the full list of public API functions, see L<perlapi>.
2824 =head1 Known Problems
2826 =head2 Thread test failures
2828 The subtests 19 and 20 of lib/thr5005.t test are known to fail due to
2829 fundamental problems in the 5.005 threading implementation. These are
2830 not new failures--Perl 5.005_0x has the same bugs, but didn't have these
2833 =head2 EBCDIC platforms not supported
2835 In earlier releases of Perl, EBCDIC environments like OS390 (also
2836 known as Open Edition MVS) and VM-ESA were supported. Due to changes
2837 required by the UTF-8 (Unicode) support, the EBCDIC platforms are not
2838 supported in Perl 5.6.0.
2840 =head2 In 64-bit HP-UX the lib/io_multihomed test may hang
2842 The lib/io_multihomed test may hang in HP-UX if Perl has been
2843 configured to be 64-bit. Because other 64-bit platforms do not
2844 hang in this test, HP-UX is suspect. All other tests pass
2845 in 64-bit HP-UX. The test attempts to create and connect to
2846 "multihomed" sockets (sockets which have multiple IP addresses).
2848 =head2 NEXTSTEP 3.3 POSIX test failure
2850 In NEXTSTEP 3.3p2 the implementation of the strftime(3) in the
2851 operating system libraries is buggy: the %j format numbers the days of
2852 a month starting from zero, which, while being logical to programmers,
2853 will cause the subtests 19 to 27 of the lib/posix test may fail.
2855 =head2 Tru64 (aka Digital UNIX, aka DEC OSF/1) lib/sdbm test failure with gcc
2857 If compiled with gcc 2.95 the lib/sdbm test will fail (dump core).
2858 The cure is to use the vendor cc, it comes with the operating system
2859 and produces good code.
2861 =head2 UNICOS/mk CC failures during Configure run
2863 In UNICOS/mk the following errors may appear during the Configure run:
2865 Guessing which symbols your C compiler and preprocessor define...
2866 CC-20 cc: ERROR File = try.c, Line = 3
2868 bad switch yylook 79bad switch yylook 79bad switch yylook 79bad switch yylook 79#ifdef A29K
2870 4 errors detected in the compilation of "try.c".
2872 The culprit is the broken awk of UNICOS/mk. The effect is fortunately
2873 rather mild: Perl itself is not adversely affected by the error, only
2874 the h2ph utility coming with Perl, and that is rather rarely needed
2877 =head2 Arrow operator and arrays
2879 When the left argument to the arrow operator C<< -> >> is an array, or
2880 the C<scalar> operator operating on an array, the result of the
2881 operation must be considered erroneous. For example:
2886 These expressions will get run-time errors in some future release of
2889 =head2 Experimental features
2891 As discussed above, many features are still experimental. Interfaces and
2892 implementation of these features are subject to change, and in extreme cases,
2893 even subject to removal in some future release of Perl. These features
2894 include the following:
2902 =item 64-bit support
2904 =item Lvalue subroutines
2906 =item Weak references
2908 =item The pseudo-hash data type
2910 =item The Compiler suite
2912 =item Internal implementation of file globbing
2916 =item The regular expression constructs C<(?{ code })> and C<(??{ code })>
2920 =head1 Obsolete Diagnostics
2924 =item Character class syntax [: :] is reserved for future extensions
2926 (W) Within regular expression character classes ([]) the syntax beginning
2927 with "[:" and ending with ":]" is reserved for future extensions.
2928 If you need to represent those character sequences inside a regular
2929 expression character class, just quote the square brackets with the
2930 backslash: "\[:" and ":\]".
2932 =item Ill-formed logical name |%s| in prime_env_iter
2934 (W) A warning peculiar to VMS. A logical name was encountered when preparing
2935 to iterate over %ENV which violates the syntactic rules governing logical
2936 names. Because it cannot be translated normally, it is skipped, and will not
2937 appear in %ENV. This may be a benign occurrence, as some software packages
2938 might directly modify logical name tables and introduce nonstandard names,
2939 or it may indicate that a logical name table has been corrupted.
2941 =item In string, @%s now must be written as \@%s
2943 The description of this error used to say:
2945 (Someday it will simply assume that an unbackslashed @
2946 interpolates an array.)
2948 That day has come, and this fatal error has been removed. It has been
2949 replaced by a non-fatal warning instead.
2950 See L</Arrays now always interpolate into double-quoted strings> for
2953 =item Probable precedence problem on %s
2955 (W) The compiler found a bareword where it expected a conditional,
2956 which often indicates that an || or && was parsed as part of the
2957 last argument of the previous construct, for example:
2961 =item regexp too big
2963 (F) The current implementation of regular expressions uses shorts as
2964 address offsets within a string. Unfortunately this means that if
2965 the regular expression compiles to longer than 32767, it'll blow up.
2966 Usually when you want a regular expression this big, there is a better
2967 way to do it with multiple statements. See L<perlre>.
2969 =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
2971 (D) Perl versions before 5.004 misinterpreted any type marker followed
2972 by "$" and a digit. For example, "$$0" was incorrectly taken to mean
2973 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
2975 However, the developers of Perl 5.004 could not fix this bug completely,
2976 because at least two widely-used modules depend on the old meaning of
2977 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
2978 old (broken) way inside strings; but it generates this message as a
2979 warning. And in Perl 5.005, this special treatment will cease.
2983 =head1 Reporting Bugs
2985 If you find what you think is a bug, you might check the
2986 articles recently posted to the comp.lang.perl.misc newsgroup.
2987 There may also be information at http://www.perl.com/perl/, the Perl
2990 If you believe you have an unreported bug, please run the B<perlbug>
2991 program included with your release. Be sure to trim your bug down
2992 to a tiny but sufficient test case. Your bug report, along with the
2993 output of C<perl -V>, will be sent off to perlbug@perl.org to be
2994 analysed by the Perl porting team.
2998 The F<Changes> file for exhaustive details on what changed.
3000 The F<INSTALL> file for how to build Perl.
3002 The F<README> file for general stuff.
3004 The F<Artistic> and F<Copying> files for copyright information.
3008 Written by Gurusamy Sarathy <F<gsar@activestate.com>>, with many
3009 contributions from The Perl Porters.
3011 Send omissions or corrections to <F<perlbug@perl.org>>.