3 perldelta - what's new for perl v5.6.0
7 This document describes differences between the 5.005 release and this one.
9 =head1 Incompatible Changes
11 =head2 Perl Source Incompatibilities
13 Beware that any new warnings that have been added or old ones
14 that have been enhanced are B<not> considered incompatible changes.
16 Since all new warnings must be explicitly requested via the C<-w>
17 switch or the C<warnings> pragma, it is ultimately the programmer's
18 responsibility to ensure that warnings are enabled judiciously.
22 =item CHECK is a new keyword
24 In addition to C<BEGIN>, C<INIT>, C<END>, C<DESTROY> and C<AUTOLOAD>,
25 subroutines named C<CHECK> are now special. These are queued up during
26 compilation and behave similar to END blocks, except they are called at
27 the end of compilation rather than at the end of execution. They cannot
30 =item Treatment of list slices of undef has changed
32 When taking a slice of a literal list (as opposed to a slice of
33 an array or hash), Perl used to return an empty list if the
34 result happened to be composed of all undef values.
36 The new behavior is to produce an empty list if (and only if)
37 the original list was empty. Consider the following example:
39 @a = (1,undef,undef,2)[2,1,2];
41 The old behavior would have resulted in @a having no elements.
42 The new behavior ensures it has three undefined elements.
44 Note in particular that the behavior of slices of the following
45 cases remains unchanged:
49 @a = (anything_returning_empty_list())[2,1,2];
55 =head2 Perl's version numbering has changed
57 Beginning with Perl version 5.6.0, the version number convention has been
58 changed to a "dotted integer" scheme that is more commonly found in open
61 Maintenance versions of v5.6.0 will be released as v5.6.1, v5.6.2 etc.
62 The next development series following v5.6.0 will be numbered v5.7.x,
63 beginning with v5.7.0, and the next major production release following
64 v5.6.0 will be v5.8.0.
66 The English module now sets $PERL_VERSION to $^V (a string value) rather
67 than C<$]> (a numeric value). (This is a potential incompatibility.
68 Send us a report via perlbug if you are affected by this.)
70 The v1.2.3 syntax is also now legal in Perl.
71 See L<Support for strings represented as a vector of ordinals> for more on that.
73 To cope with the new versioning system's use of at least three significant
74 digits for each version component, the method used for incrementing the
75 subversion number has also changed slightly. We assume that versions older
76 than v5.6.0 have been incrementing the subversion component in multiples of
77 10. Versions after v5.6.0 will increment them by 1. Thus, using the new
78 notation, 5.005_03 is the "same" as v5.5.30, and the first maintenance
79 version following v5.6.0 will be v5.6.1 (which should be read as being
80 equivalent to a floating point value of 5.006_001 in the older format,
83 =item Literals of the form C<1.2.3> parse differently
85 Previously, numeric literals with more than one dot in them were
86 interpreted as a floating point number concatenated with one or more
87 numbers. Such "numbers" are now parsed as strings composed of the
90 For example, C<print 97.98.99> used to output C<97.9899> in earlier
91 versions, but now prints C<abc>.
93 See L<Support for strings represented as a vector of ordinals> below.
95 =item Possibly changed pseudo-random number generator
97 In 5.005_0x and earlier, perl's rand() function used the C library
98 rand(3) function. As of 5.005_52, Configure tests for drand48(),
99 random(), and rand() (in that order) and picks the first one it finds.
100 Perl programs that depend on reproducing a specific set of pseudo-random
101 numbers will now likely produce different output. You can use
102 C<sh Configure -Drandfunc=rand> to obtain the old behavior.
104 =item Hashing function for hash keys has changed
106 Perl hashes are not order preserving. The apparently random order
107 encountered when iterating on the contents of a hash is determined
108 by the hashing algorithm used. To improve the distribution of lower
109 bits in the hashed value, the algorithm has changed slightly as of
110 5.005_52. When iterating over hashes, this may yield a random order
111 that is B<different> from that of previous versions.
113 =item C<undef> fails on read only values
115 Using the C<undef> operator on a readonly value (such as $1) has
116 the same effect as assigning C<undef> to the readonly value--it
119 =item Close-on-exec bit may be set on pipe and socket handles
121 On systems that support a close-on-exec flag on filehandles, the
122 flag will be set for any handles created by pipe(), socketpair(),
123 socket(), and accept(), if that is warranted by the value of $^F
124 that may be in effect. Earlier versions neglected to set the flag
125 for handles created with these operators. See L<perlfunc/pipe>,
126 L<perlfunc/socketpair>, L<perlfunc/socket>, L<perlfunc/accept>,
129 =item Writing C<"$$1"> to mean C<"${$}1"> is unsupported
131 Perl 5.004 deprecated the interpretation of C<$$1> and
132 similar within interpolated strings to mean C<$$ . "1">,
133 but still allowed it.
135 In Perl 5.6.0 and later, C<"$$1"> always means C<"${$1}">.
137 =item delete(), values() and C<\(%h)> operate on aliases to values, not copies
139 delete(), each(), values() and hashes in a list context return the actual
140 values in the hash, instead of copies (as they used to in earlier
141 versions). Typical idioms for using these constructs copy the
142 returned values, but this can make a significant difference when
143 creating references to the returned values.
145 Keys in the hash are still returned as copies when iterating on
148 =item vec(EXPR,OFFSET,BITS) enforces powers-of-two BITS
150 vec() generates a run-time error if the BITS argument is not
151 a valid power-of-two integer.
153 =item Text of some diagnostic output has changed
155 Most references to internal Perl operations in diagnostics
156 have been changed to be more descriptive. This may be an
157 issue for programs that may incorrectly rely on the exact
158 text of diagnostics for proper functioning.
160 =item C<%@> has been removed
162 The undocumented special variable C<%@> that used to accumulate
163 "background" errors (such as those that happen in DESTROY())
164 has been removed, because it could potentially result in memory
167 =item Parenthesized not() behaves like a list operator
169 The C<not> operator now falls under the "if it looks like a function,
170 it behaves like a function" rule.
172 As a result, the parenthesized form can be used with C<grep> and C<map>.
173 The following construct used to be a syntax error before, but it works
176 grep not($_), @things;
178 On the other hand, using C<not> with a literal list slice may not
179 work. The following previously allowed construct:
181 print not (1,2,3)[0];
183 needs to be written with additional parentheses now:
185 print not((1,2,3)[0]);
187 The behavior remains unaffected when C<not> is not followed by parentheses.
189 =item Semantics of bareword prototype C<(*)> have changed
191 Arguments prototyped as C<*> will now be visible within the subroutine
192 as either a simple scalar or as a reference to a typeglob. Perl 5.005
193 always coerced simple scalar arguments to a typeglob, which wasn't useful
194 in situations where the subroutine must distinguish between a simple
195 scalar and a typeglob. See L<perlsub/Prototypes>.
197 =head2 On 64-bit platforms the semantics of bit operators have changed
199 If your platform is either natively 64-bit or your Perl has been
200 configured to used 64-bit integers, i.e., $Config{ivsize} is 8,
201 be warned that the semantics of all the bitwise numeric operators
202 (& | ^ ~ << >>) have been changed. These operators used to strictly
203 operate on the lower 32 bits of integers, but now operate over the
204 entire width of native integers. In particular, note that unary C<~>
205 will produce different results on platforms that have different
206 $Config{ivsize}. For portability, be sure to mask off the excess bits
207 in the result of unary C<~>, e.g., C<~$x & 0xffffffff>.
211 =head2 C Source Incompatibilities
215 =item C<PERL_POLLUTE>
217 Release 5.005 grandfathered old global symbol names by providing preprocessor
218 macros for extension source compatibility. As of release 5.6.0, these
219 preprocessor definitions are not available by default. You need to explicitly
220 compile perl with C<-DPERL_POLLUTE> to get these definitions. For
221 extensions still using the old symbols, this option can be
222 specified via MakeMaker:
224 perl Makefile.PL POLLUTE=1
226 =item C<PERL_IMPLICIT_CONTEXT>
228 NOTE: PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built
229 with one of -Dusethreads, -Dusemultiplicity, or both. It is not
230 intended to be enabled by users at this time.
232 This new build option provides a set of macros for all API functions
233 such that an implicit interpreter/thread context argument is passed to
234 every API function. As a result of this, something like C<sv_setsv(foo,bar)>
235 amounts to a macro invocation that actually translates to something like
236 C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
237 to not have any significant source compatibility issues, the difference
238 between a macro and a real function call will need to be considered.
240 This means that there B<is> a source compatibility issue as a result of
241 this if your extensions attempt to use pointers to any of the Perl API
244 Note that the above issue is not relevant to the default build of
245 Perl, whose interfaces continue to match those of prior versions
246 (but subject to the other options described here).
248 See L<perlguts/"The Perl API"> for detailed information on the
249 ramifications of building Perl with this option.
251 =item C<PERL_POLLUTE_MALLOC>
253 Enabling Perl's malloc in release 5.005 and earlier caused the namespace of
254 the system's malloc family of functions to be usurped by the Perl versions,
255 since by default they used the same names. Besides causing problems on
256 platforms that do not allow these functions to be cleanly replaced, this
257 also meant that the system versions could not be called in programs that
258 used Perl's malloc. Previous versions of Perl have allowed this behaviour
259 to be suppressed with the HIDEMYMALLOC and EMBEDMYMALLOC preprocessor
262 As of release 5.6.0, Perl's malloc family of functions have default names
263 distinct from the system versions. You need to explicitly compile perl with
264 C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
265 and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
268 Note that these functions do B<not> constitute Perl's memory allocation API.
269 See L<perlguts/"Memory Allocation"> for further information about that.
273 =head2 Compatible C Source API Changes
277 =item C<PATCHLEVEL> is now C<PERL_VERSION>
279 The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
280 are now available by default from perl.h, and reflect the base revision,
281 patchlevel, and subversion respectively. C<PERL_REVISION> had no
282 prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
283 previously available as C<PATCHLEVEL> and C<SUBVERSION>.
285 The new names cause less pollution of the B<cpp> namespace and reflect what
286 the numbers have come to stand for in common practice. For compatibility,
287 the old names are still supported when F<patchlevel.h> is explicitly
288 included (as required before), so there is no source incompatibility
293 =head2 Binary Incompatibilities
295 In general, the default build of this release is expected to be binary
296 compatible for extensions built with the 5.005 release or its maintenance
297 versions. However, specific platforms may have broken binary compatibility
298 due to changes in the defaults used in hints files. Therefore, please be
299 sure to always check the platform-specific README files for any notes to
302 The usethreads or usemultiplicity builds are B<not> binary compatible
303 with the corresponding builds in 5.005.
305 On platforms that require an explicit list of exports (AIX, OS/2 and Windows,
306 among others), purely internal symbols such as parser functions and the
307 run time opcodes are not exported by default. Perl 5.005 used to export
308 all functions irrespective of whether they were considered part of the
311 For the full list of public API functions, see L<perlapi>.
313 =head1 Installation and Configuration Improvements
315 =head2 -Dusethreads means something different
317 WARNING: Support for threads continues to be an experimental feature.
318 Interfaces and implementation are subject to sudden and drastic changes.
320 The -Dusethreads flag now enables the experimental interpreter-based thread
321 support by default. To get the flavor of experimental threads that was in
322 5.005 instead, you need to run Configure with "-Dusethreads -Duse5005threads".
324 As of v5.6.0, interpreter-threads support is still lacking a way to
325 create new threads from Perl (i.e., C<use Thread;> will not work with
326 interpreter threads). C<use Thread;> continues to be available when you
327 specify the -Duse5005threads option to Configure, bugs and all.
329 =head2 New Configure flags
331 The following new flags may be enabled on the Configure command line
332 by running Configure with C<-Dflag>.
335 usethreads useithreads (new interpreter threads: no Perl API yet)
336 usethreads use5005threads (threads as they were in 5.005)
338 use64bitint (equal to now deprecated 'use64bits')
344 usesocks (only SOCKS v5 supported)
346 =head2 Threadedness and 64-bitness now more daring
348 The Configure options enabling the use of threads and the use of
349 64-bitness are now more daring in the sense that they no more have an
350 explicit list of operating systems of known threads/64-bit
351 capabilities. In other words: if your operating system has the
352 necessary APIs and datatypes, you should be able just to go ahead and
353 use them, for threads by Configure -Dusethreads, and for 64 bits
354 either explicitly by Configure -Duse64bitint or implicitly if your
355 system has 64-bit wide datatypes. See also L<"64-bit support">.
359 Some platforms have "long doubles", floating point numbers of even
360 larger range than ordinary "doubles". To enable using long doubles for
361 Perl's scalars, use -Duselongdouble.
365 You can enable both -Duse64bitint and -Duselongdouble with -Dusemorebits.
366 See also L<"64-bit support">.
368 =head2 -Duselargefiles
370 Some platforms support system APIs that are capable of handling large files
371 (typically, files larger than two gigabytes). Perl will try to use these
372 APIs if you ask for -Duselargefiles.
374 See L<"Large file support"> for more information.
376 =head2 installusrbinperl
378 You can use "Configure -Uinstallusrbinperl" which causes installperl
379 to skip installing perl also as /usr/bin/perl. This is useful if you
380 prefer not to modify /usr/bin for some reason or another but harmful
381 because many scripts assume to find Perl in /usr/bin/perl.
385 You can use "Configure -Dusesocks" which causes Perl to probe
386 for the SOCKS proxy protocol library (v5, not v4). For more information
389 http://www.socks.nec.com/
393 You can "post-edit" the Configure variables using the Configure C<-A>
394 switch. The editing happens immediately after the platform specific
395 hints files have been processed but before the actual configuration
396 process starts. Run C<Configure -h> to find out the full C<-A> syntax.
398 =head2 Enhanced Installation Directories
400 The installation structure has been enriched to improve the support
401 for maintaining multiple versions of perl, to provide locations for
402 vendor-supplied modules, scripts, and manpages, and to ease maintenance
403 of locally-added modules, scripts, and manpages. See the section on
404 Installation Directories in the INSTALL file for complete details.
405 For most users building and installing from source, the defaults should
408 If you previously used C<Configure -Dsitelib> or C<-Dsitearch> to set
409 special values for library directories, you might wish to consider using
410 the new C<-Dsiteprefix> setting instead. Also, if you wish to re-use a
411 config.sh file from an earlier version of perl, you should be sure to
412 check that Configure makes sensible choices for the new directories.
413 See INSTALL for complete details.
417 =head2 Unicode and UTF-8 support
419 WARNING: This is an experimental feature. Implementation details are
422 Perl now uses UTF-8 as its internal representation for character
423 strings. The C<utf8> and C<bytes> pragmas are used to control this support
424 in the current lexical scope. See L<perlunicode>, L<utf8> and L<bytes> for
427 =head2 Interpreter cloning, threads, and concurrency
429 WARNING: This is an experimental feature. Implementation details are
432 Perl 5.005_63 introduces the beginnings of support for running multiple
433 interpreters concurrently in different threads. In conjunction with
434 the perl_clone() API call, which can be used to selectively duplicate
435 the state of any given interpreter, it is possible to compile a
436 piece of code once in an interpreter, clone that interpreter
437 one or more times, and run all the resulting interpreters in distinct
440 On Windows, this feature is used to emulate fork() at the interpreter
441 level. See L<perlfork>.
443 This feature is still in evolution. It is eventually meant to be used
444 to selectively clone a subroutine and data reachable from that
445 subroutine in a separate interpreter and run the cloned subroutine
446 in a separate thread. Since there is no shared data between the
447 interpreters, little or no locking will be needed (unless parts of
448 the symbol table are explicitly shared). This is obviously intended
449 to be an easy-to-use replacement for the existing threads support.
451 Support for cloning interpreters and interpreter concurrency can be
452 enabled using the -Dusethreads Configure option (see win32/Makefile for
453 how to enable it on Windows.) The resulting perl executable will be
454 functionally identical to one that was built with -Dmultiplicity, but
455 the perl_clone() API call will only be available in the former.
457 -Dusethreads enables the cpp macro USE_ITHREADS by default, which in turn
458 enables Perl source code changes that provide a clear separation between
459 the op tree and the data it operates with. The former is immutable, and
460 can therefore be shared between an interpreter and all of its clones,
461 while the latter is considered local to each interpreter, and is therefore
462 copied for each clone.
464 Note that building Perl with the -Dusemultiplicity Configure option
465 is adequate if you wish to run multiple B<independent> interpreters
466 concurrently in different threads. -Dusethreads only provides the
467 additional functionality of the perl_clone() API call and other
468 support for running B<cloned> interpreters concurrently.
470 =head2 Lexically scoped warning categories
472 You can now control the granularity of warnings emitted by perl at a finer
473 level using the C<use warnings> pragma. See L<warnings> and L<perllexwarn>
476 =head2 Lvalue subroutines
478 WARNING: This is an experimental feature. Details are subject to change.
480 Subroutines can now return modifiable lvalues.
481 See L<perlsub/"Lvalue subroutines">.
483 =head2 "our" declarations
485 An "our" declaration introduces a value that can be best understood
486 as a lexically scoped symbolic alias to a global variable in the
487 package that was current where the variable was declared. This is
488 mostly useful as an alternative to the C<vars> pragma, but also provides
489 the opportunity to introduce typing and other attributes for such
490 variables. See L<perlfunc/our>.
492 =head2 Support for strings represented as a vector of ordinals
494 Literals of the form C<v1.2.3.4> are now parsed as a string composed of
495 of characters with the specified ordinals. This is an alternative, more
496 readable way to construct (possibly unicode) strings instead of
497 interpolating characters, as in C<"\x{1}\x{2}\x{3}\x{4}">. The leading
498 C<v> may be omitted if there are more than two ordinals, so C<1.2.3> is
499 parsed the same as C<v1.2.3>.
501 Strings written in this form are also useful to represent version "numbers".
502 It is easy to compare such version "numbers" (which are really just plain
503 strings) using any of the usual string comparison operators C<eq>, C<ne>,
504 C<lt>, C<gt>, etc., or perform bitwise string operations on them using C<|>,
507 In conjunction with the new C<$^V> magic variable (which contains
508 the perl version as a string), such literals can be used as a readable way
509 to check if you're running a particular version of Perl:
511 # this will parse in older versions of Perl also
512 if ($^V and $^V gt v5.6.0) {
513 # new features supported
516 C<require> and C<use> also have some special magic to support such literals.
517 They will be interpreted as a version rather than as a module name:
519 require v5.6.0; # croak if $^V lt v5.6.0
520 use v5.6.0; # same, but croaks at compile-time
522 Alternatively, the C<v> may be omitted if there is more than one dot:
527 Also, C<sprintf> and C<printf> support the Perl-specific format flag C<%v>
528 to print ordinals of characters in arbitrary strings:
530 printf "v%vd", $^V; # prints current version, such as "v5.5.650"
531 printf "%*vX", ":", $addr; # formats IPv6 address
532 printf "%*vb", " ", $bits; # displays bitstring
534 See L<perldata/"Scalar value constructors"> for additional information.
536 =head2 Weak references
538 WARNING: This is an experimental feature.
540 In previous versions of Perl, you couldn't cache objects so as
541 to allow them to be deleted if the last reference from outside
542 the cache is deleted. The reference in the cache would hold a
543 reference count on the object and the objects would never be
546 Another familiar problem is with circular references. When an
547 object references itself, its reference count would never go
548 down to zero, and it would not get destroyed until the program
551 Weak references solve this by allowing you to "weaken" any
552 reference, that is, make it not count towards the reference count.
553 When the last non-weak reference to an object is deleted, the object
554 is destroyed and all the weak references to the object are
555 automatically undef-ed.
557 To use this feature, you need the WeakRef package from CPAN, which
558 contains additional documentation.
560 =head2 File globbing implemented internally
562 WARNING: This is currently an experimental feature. Interfaces and
563 implementation are likely to change.
565 Perl now uses the File::Glob implementation of the glob() operator
566 automatically. This avoids using an external csh process and the
567 problems associated with it.
569 =head2 Binary numbers supported
571 Binary numbers are now supported as literals, in s?printf formats, and
575 printf "The answer is: %b\n", oct("0b101010");
577 =head2 Some arrows may be omitted in calls through references
579 Perl now allows the arrow to be omitted in many constructs
580 involving subroutine calls through references. For example,
581 C<< $foo[10]->('foo') >> may now be written C<$foo[10]('foo')>.
582 This is rather similar to how the arrow may be omitted from
583 C<< $foo[10]->{'foo'} >>. Note however, that the arrow is still
584 required for C<< foo(10)->('bar') >>.
586 =head2 exists() is supported on subroutine names
588 The exists() builtin now works on subroutine names. A subroutine
589 is considered to exist if it has been declared (even if implicitly).
590 See L<perlfunc/exists> for examples.
592 =head2 exists() and delete() are supported on array elements
594 The exists() and delete() builtins now work on simple arrays as well.
595 The behavior is similar to that on hash elements.
597 exists() can be used to check whether an array element has been
598 initialized. This avoids autovivifying array elements that don't exist.
599 If the array is tied, the EXISTS() method in the corresponding tied
600 package will be invoked.
602 delete() may be used to remove an element from the array and return
603 it. The array element at that position returns to its unintialized
604 state, so that testing for the same element with exists() will return
605 false. If the element happens to be the one at the end, the size of
606 the array also shrinks up to the highest element that tests true for
607 exists(), or 0 if none such is found. If the array is tied, the DELETE()
608 method in the corresponding tied package will be invoked.
610 See L<perlfunc/exists> and L<perlfunc/delete> for examples.
612 =head2 syswrite() ease-of-use
614 The length argument of C<syswrite()> has become optional.
616 =head2 File and directory handles can be autovivified
618 Similar to how constructs such as C<< $x->[0] >> autovivify a reference,
619 handle constructors (open(), opendir(), pipe(), socketpair(), sysopen(),
620 socket(), and accept()) now autovivify a file or directory handle
621 if the handle passed to them is an uninitialized scalar variable. This
622 allows the constructs such as C<open(my $fh, ...)> and C<open(local $fh,...)>
623 to be used to create filehandles that will conveniently be closed
624 automatically when the scope ends, provided there are no other references
625 to them. This largely eliminates the need for typeglobs when opening
626 filehandles that must be passed around, as in the following example:
630 or die "Can't open '@_': $!";
635 my $f = myopen("</etc/motd");
637 # $f implicitly closed here
640 =head2 open() with more than two arguments
642 If open() is passed three arguments instead of two, the second arguments
643 is used as the mode and the third argument is taken to be the file name.
644 This is primarily useful for protecting against unintended magic behavior
645 of the traditional two-argument form. See L<perlfunc/open>.
647 =head2 64-bit support
649 NOTE: The Configure flags -Duselonglong and -Duse64bits have been
650 deprecated. Use -Duse64bitint instead.
652 Any platform that has 64-bit integers either
654 (1) natively as longs or ints
655 (2) via special compiler flags
656 (3) using long long or int64_t
658 are able to use "quads" (64-bit integers) as follows:
664 constants (decimal, hexadecimal, octal, binary) in the code
668 arguments to oct() and hex()
672 arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)
680 pack() and unpack() "q" and "Q" formats
684 in basic arithmetics: + - * / % (NOTE: operating close to the limits
685 of the integer values may produce surprising results)
689 in bit arithmetics: & | ^ ~ << >> (NOTE: these used to be forced
690 to be 32 bits wide but now operate on the full native width.)
698 Note that unless you have the case (a) you will have to configure
699 and compile Perl using the -Duse64bitint Configure flag.
701 There are actually two modes of 64-bitness: the first one is achieved
702 using Configure -Duse64bitint and the second one using Configure
703 -Duse64bitall. The difference is that the first one is minimal and
704 the second one maximal.
706 The C<use64bitint> does only as much as is required to get 64-bit
707 integers into Perl (this may mean, for example, using "long longs")
708 while your memory may still be limited to 2 gigabytes (because your
709 pointers could still be 32-bit). Note that the name C<64bitint> does
710 not imply that your C compiler will be using 64-bit C<int>s (it might,
711 but it doesn't have to): the C<use64bitint> means that you will be
712 able to have 64 bits wide scalar values.
714 The C<use64bitall> goes all the way by attempting to switch also
715 integers (if it can), longs (and pointers) to being 64-bit. This may
716 create an even more binary incompatible Perl than -Duse64bitint: the
717 resulting executable may not run at all in a 32-bit box, or you may
718 have to reboot/reconfigure/rebuild your operating system to be 64-bit
721 Natively 64-bit systems like Alpha and Cray need neither -Duse64bitint
724 Last but not least: note that due to Perl's habit of always using
725 floating point numbers, the quads are still not true integers.
726 When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned,
727 -9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they
728 are silently promoted to floating point numbers, after which they will
729 start losing precision (in their lower digits).
731 =head2 Large file support
733 If you have filesystems that support "large files" (files larger than
734 2 gigabytes), you may now also be able to create and access them from
735 Perl. NOTE: the default action is to use the large file support, if
736 available on the platform.
738 If the large file support is on, and you have a Fcntl constant
739 O_LARGEFILE, the O_LARGEFILE is automatically added to the flags
742 Beware: unless your filesystem also supports "sparse files" seeking to
743 umpteen petabytes may be unadvisable.
745 Note that in addition to requiring a proper file system to do large
746 files you may also need to adjust your per-process (or your
747 per-system, or per-process-group, or per-user-group) maximum filesize
748 limits before running Perl scripts that try to handle large files,
749 especially if you intend to write such files.
751 Finally, in addition to your process/process group maximum filesize
752 limits, you may have quota limits on your filesystems that stop you
753 (your user id or your user group id) from using large files.
755 Adjusting your process/user/group/file system/operating system limits
756 is outside the scope of Perl core language. For process limits, you
757 may try increasing the limits using your shell's limits/limit/ulimit
758 command before running Perl. The BSD::Resource extension (not
759 included with the standard Perl distribution) may also be of use, it
760 offers the getrlimit/setrlimit interface that can be used to adjust
761 process resource usage limits, including the maximum filesize limit.
765 In some systems you may be able to use long doubles to enhance the
766 range and precision of your double precision floating point numbers
767 (that is, Perl's numbers). Use Configure -Duselongdouble to enable
768 this support (if it is available).
772 You can "Configure -Dusemorebits" to turn on both the 64-bit support
773 and the long double support.
775 =head2 Enhanced support for sort() subroutines
777 Perl subroutines with a prototype of C<($$)>, and XSUBs in general, can
778 now be used as sort subroutines. In either case, the two elements to
779 be compared are passed as normal parameters in @_. See L<perlfunc/sort>.
781 For unprototyped sort subroutines, the historical behavior of passing
782 the elements to be compared as the global variables $a and $b remains
785 =head2 Better syntax checks on parenthesized unary operators
789 print defined(&foo,&bar,&baz);
790 print uc("foo","bar","baz");
793 used to be accidentally allowed in earlier versions, and produced
794 unpredictable behaviour. Some produced ancillary warnings
795 when used in this way; others silently did the wrong thing.
797 The parenthesized forms of most unary operators that expect a single
798 argument now ensure that they are not called with more than one
799 argument, making the cases shown above syntax errors. The usual
802 print defined &foo, &bar, &baz;
803 print uc "foo", "bar", "baz";
806 remains unchanged. See L<perlop>.
808 =head2 POSIX character class syntax [: :] supported
810 For example to match alphabetic characters use /[[:alpha:]]/.
811 See L<perlre> for details.
813 =head2 Improved C<qw//> operator
815 The C<qw//> operator is now evaluated at compile time into a true list
816 instead of being replaced with a run time call to C<split()>. This
817 removes the confusing misbehaviour of C<qw//> in scalar context, which
818 had inherited that behaviour from split().
822 $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
824 now correctly prints "3|a", instead of "2|a".
826 =head2 pack() format 'Z' supported
828 The new format type 'Z' is useful for packing and unpacking null-terminated
829 strings. See L<perlfunc/"pack">.
831 =head2 pack() format modifier '!' supported
833 The new format type modifier '!' is useful for packing and unpacking
834 native shorts, ints, and longs. See L<perlfunc/"pack">.
836 =head2 pack() and unpack() support counted strings
838 The template character '/' can be used to specify a counted string
839 type to be packed or unpacked. See L<perlfunc/"pack">.
841 =head2 Comments in pack() templates
843 The '#' character in a template introduces a comment up to
844 end of the line. This facilitates documentation of pack()
847 =head2 $^X variables may now have names longer than one character
849 Formerly, $^X was synonymous with ${"\cX"}, but $^XY was a syntax
850 error. Now variable names that begin with a control character may be
851 arbitrarily long. However, for compatibility reasons, these variables
852 I<must> be written with explicit braces, as C<${^XY}> for example.
853 C<${^XYZ}> is synonymous with ${"\cXYZ"}. Variable names with more
854 than one control character, such as C<${^XY^Z}>, are illegal.
856 The old syntax has not changed. As before, `^X' may be either a
857 literal control-X character or the two-character sequence `caret' plus
858 `X'. When braces are omitted, the variable name stops after the
859 control character. Thus C<"$^XYZ"> continues to be synonymous with
860 C<$^X . "YZ"> as before.
862 As before, lexical variables may not have names beginning with control
863 characters. As before, variables whose names begin with a control
864 character are always forced to be in package `main'. All such variables
865 are reserved for future extensions, except those that begin with
866 C<^_>, which may be used by user programs and are guaranteed not to
867 acquire special meaning in any future version of Perl.
869 =head2 C<use attrs> implicit in subroutine attributes
871 Formerly, if you wanted to mark a subroutine as being a method call or
872 as requiring an automatic lock() when it is entered, you had to declare
873 that with a C<use attrs> pragma in the body of the subroutine.
874 That can now be accomplished with declaration syntax, like this:
876 sub mymethod : locked method ;
878 sub mymethod : locked method {
882 sub othermethod :locked :method ;
884 sub othermethod :locked :method {
889 (Note how only the first C<:> is mandatory, and whitespace surrounding
890 the C<:> is optional.)
892 F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes
893 with the stubs they provide. See L<attributes>.
895 =head2 Support for interpolating named characters
897 The new C<\N> escape interpolates named characters within strings.
898 For example, C<"Hi! \N{WHITE SMILING FACE}"> evaluates to a string
899 with a unicode smiley face at the end.
901 =head2 C<require> and C<do> may be overridden
903 C<require> and C<do 'file'> operations may be overridden locally
904 by importing subroutines of the same name into the current package
905 (or globally by importing them into the CORE::GLOBAL:: namespace).
906 Overriding C<require> will also affect C<use>, provided the override
907 is visible at compile-time.
908 See L<perlsub/"Overriding Built-in Functions">.
910 =head2 New variable $^C reflects C<-c> switch
912 C<$^C> has a boolean value that reflects whether perl is being run
913 in compile-only mode (i.e. via the C<-c> switch). Since
914 BEGIN blocks are executed under such conditions, this variable
915 enables perl code to determine whether actions that make sense
916 only during normal running are warranted. See L<perlvar>.
918 =head2 New variable $^V contains Perl version as a string
920 C<$^V> contains the Perl version number as a string composed of
921 characters whose ordinals match the version numbers, i.e. v5.6.0.
922 This may be used in string comparisons.
924 See C<Support for strings represented as a vector of ordinals> for an
927 =head2 Optional Y2K warnings
929 If Perl is built with the cpp macro C<PERL_Y2KWARN> defined,
930 it emits optional warnings when concatenating the number 19
933 This behavior must be specifically enabled when running Configure.
934 See F<INSTALL> and F<README.Y2K>.
936 =head1 Significant bug fixes
938 =head2 <HANDLE> on empty files
940 With C<$/> set to C<undef>, "slurping" an empty file returns a string of
941 zero length (instead of C<undef>, as it used to) the first time the
942 HANDLE is read after C<$/> is set to C<undef>. Further reads yield
945 This means that the following will append "foo" to an empty file (it used
948 perl -0777 -pi -e 's/^/foo/' empty_file
952 perl -pi -e 's/^/foo/' empty_file
954 is unchanged (it continues to leave the file empty).
956 =head2 C<eval '...'> improvements
958 Line numbers (as reflected by caller() and most diagnostics) within
959 C<eval '...'> were often incorrect where here documents were involved.
960 This has been corrected.
962 Lexical lookups for variables appearing in C<eval '...'> within
963 functions that were themselves called within an C<eval '...'> were
964 searching the wrong place for lexicals. The lexical search now
965 correctly ends at the subroutine's block boundary.
967 Parsing of here documents used to be flawed when they appeared as
968 the replacement expression in C<eval 's/.../.../e'>. This has
971 =head2 All compilation errors are true errors
973 Some "errors" encountered at compile time were by neccessity
974 generated as warnings followed by eventual termination of the
975 program. This enabled more such errors to be reported in a
976 single run, rather than causing a hard stop at the first error
977 that was encountered.
979 The mechanism for reporting such errors has been reimplemented
980 to queue compile-time errors and report them at the end of the
981 compilation as true errors rather than as warnings. This fixes
982 cases where error messages leaked through in the form of warnings
983 when code was compiled at run time using C<eval STRING>, and
984 also allows such errors to be reliably trapped using C<eval "...">.
986 =head2 Automatic flushing of output buffers
988 fork(), exec(), system(), qx//, and pipe open()s now flush buffers
989 of all files opened for output when the operation was attempted. This
990 mostly eliminates confusing buffering mishaps suffered by users unaware
991 of how Perl internally handles I/O.
993 This is not supported on some platforms like Solaris where a suitably
994 correct implementation of fflush(NULL) isn't available.
996 =head2 Better diagnostics on meaningless filehandle operations
998 Constructs such as C<< open(<FH>) >> and C<< close(<FH>) >>
999 are compile time errors. Attempting to read from filehandles that
1000 were opened only for writing will now produce warnings (just as
1001 writing to read-only filehandles does).
1003 =head2 Where possible, buffered data discarded from duped input filehandle
1005 C<< open(NEW, "<&OLD") >> now attempts to discard any data that
1006 was previously read and buffered in C<OLD> before duping the handle.
1007 On platforms where doing this is allowed, the next read operation
1008 on C<NEW> will return the same data as the corresponding operation
1009 on C<OLD>. Formerly, it would have returned the data from the start
1010 of the following disk block instead.
1012 =head2 eof() has the same old magic as <>
1014 C<eof()> would return true if no attempt to read from C<< <> >> had
1015 yet been made. C<eof()> has been changed to have a little magic of its
1016 own, it now opens the C<< <> >> files.
1018 =head2 system(), backticks and pipe open now reflect exec() failure
1020 On Unix and similar platforms, system(), qx() and open(FOO, "cmd |")
1021 etc., are implemented via fork() and exec(). When the underlying
1022 exec() fails, earlier versions did not report the error properly,
1023 since the exec() happened to be in a different process.
1025 The child process now communicates with the parent about the
1026 error in launching the external command, which allows these
1027 constructs to return with their usual error value and set $!.
1029 =head2 Implicitly closed filehandles are safer
1031 Sometimes implicitly closed filehandles (as when they are localized,
1032 and Perl automatically closes them on exiting the scope) could
1033 inadvertently set $? or $!. This has been corrected.
1035 =head2 C<(\$)> prototype and C<$foo{a}>
1037 A scalar reference prototype now correctly allows a hash or
1038 array element in that slot.
1040 =head2 Pseudo-hashes work better
1042 Dereferencing some types of reference values in a pseudo-hash,
1043 such as C<< $ph->{foo}[1] >>, was accidentally disallowed. This has
1046 When applied to a pseudo-hash element, exists() now reports whether
1047 the specified value exists, not merely if the key is valid.
1049 delete() now works on pseudo-hashes. When given a pseudo-hash element
1050 or slice it deletes the values corresponding to the keys (but not the keys
1051 themselves). See L<perlref/"Pseudo-hashes: Using an array as a hash">.
1053 Pseudo-hash slices with constant keys are now optimized to array lookups
1056 The C<fields> pragma now provides ways to create pseudo-hashes, via
1057 fields::new() and fields::phash(). See L<fields>.
1059 =head2 C<goto &sub> and AUTOLOAD
1061 The C<goto &sub> construct works correctly when C<&sub> happens
1064 =head2 C<-bareword> allowed under C<use integer>
1066 The autoquoting of barewords preceded by C<-> did not work
1067 in prior versions when the C<integer> pragma was enabled.
1068 This has been fixed.
1070 =head2 Boolean assignment operators are legal lvalues
1072 Constructs such as C<($a ||= 2) += 1> are now allowed.
1074 =head2 C<sort $coderef @foo> allowed
1076 sort() did not accept a subroutine reference as the comparison
1077 function in earlier versions. This is now permitted.
1079 =head2 Failures in DESTROY()
1081 When code in a destructor threw an exception, it went unnoticed
1082 in earlier versions of Perl, unless someone happened to be
1083 looking in $@ just after the point the destructor happened to
1084 run. Such failures are now visible as warnings when warnings are
1087 =head2 Locale bugs fixed
1089 printf() and sprintf() previously reset the numeric locale
1090 back to the default "C" locale. This has been fixed.
1092 Numbers formatted according to the local numeric locale
1093 (such as using a decimal comma instead of a decimal dot) caused
1094 "isn't numeric" warnings, even while the operations accessing
1095 those numbers produced correct results. These warnings have been
1100 The C<eval 'return sub {...}'> construct could sometimes leak
1101 memory. This has been fixed.
1103 Operations that aren't filehandle constructors used to leak memory
1104 when used on invalid filehandles. This has been fixed.
1106 Constructs that modified C<@_> could fail to deallocate values
1107 in C<@_> and thus leak memory. This has been corrected.
1109 =head2 Spurious subroutine stubs after failed subroutine calls
1111 Perl could sometimes create empty subroutine stubs when a
1112 subroutine was not found in the package. Such cases stopped
1113 later method lookups from progressing into base packages.
1114 This has been corrected.
1116 =head2 Taint failures under C<-U>
1118 When running in unsafe mode, taint violations could sometimes
1119 cause silent failures. This has been fixed.
1121 =head2 END blocks and the C<-c> switch
1123 Prior versions used to run BEGIN B<and> END blocks when Perl was
1124 run in compile-only mode. Since this is typically not the expected
1125 behavior, END blocks are not executed anymore when the C<-c> switch
1128 See L<CHECK blocks> for how to run things when the compile phase ends.
1130 =head2 Potential to leak DATA filehandles
1132 Using the C<__DATA__> token creates an implicit filehandle to
1133 the file that contains the token. It is the program's
1134 responsibility to close it when it is done reading from it.
1136 This caveat is now better explained in the documentation.
1139 =head2 Diagnostics follow STDERR
1141 Diagnostic output now goes to whichever file the C<STDERR> handle
1142 is pointing at, instead of always going to the underlying C runtime
1143 library's C<stderr>.
1145 =head2 Other fixes for better diagnostics
1147 Line numbers are no longer suppressed (under most likely circumstances)
1148 during the global destruction phase.
1150 Diagnostics emitted from code running in threads other than the main
1151 thread are now accompanied by the thread ID.
1153 Embedded null characters in diagnostics now actually show up. They
1154 used to truncate the message in prior versions.
1156 $foo::a and $foo::b are now exempt from "possible typo" warnings only
1157 if sort() is encountered in package C<foo>.
1159 Unrecognized alphabetic escapes encountered when parsing quote
1160 constructs now generate a warning, since they may take on new
1161 semantics in later versions of Perl.
1163 Many diagnostics now report the internal operation in which the warning
1164 was provoked, like so:
1166 Use of uninitialized value in concatenation (.) at (eval 1) line 1.
1167 Use of uninitialized value in print at (eval 1) line 1.
1169 Diagnostics that occur within eval may also report the file and line
1170 number where the eval is located, in addition to the eval sequence
1171 number and the line number within the evaluated text itself. For
1174 Not enough arguments for scalar at (eval 4)[newlib/perl5db.pl:1411] line 2, at EOF
1176 =head1 Performance enhancements
1178 =head2 Simple sort() using { $a <=> $b } and the like are optimized
1180 Many common sort() operations using a simple inlined block are now
1181 optimized for faster performance.
1183 =head2 Optimized assignments to lexical variables
1185 Certain operations in the RHS of assignment statements have been
1186 optimized to directly set the lexical variable on the LHS,
1187 eliminating redundant copying overheads.
1189 =head2 Faster subroutine calls
1191 Minor changes in how subroutine calls are handled internally
1192 provide marginal improvements in performance.
1194 =head1 Platform specific changes
1196 =head2 Supported platforms
1202 VM/ESA is now supported.
1206 Siemens BS2000 is now supported under the POSIX Shell.
1210 The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
1215 GNU/Hurd is now supported.
1219 Rhapsody/Darwin is now supported.
1223 EPOC is is now supported (on Psion 5).
1233 Perl now works with djgpp 2.02 (and 2.03 alpha).
1237 Environment variable names are not converted to uppercase any more.
1241 Incorrect exit codes from backticks have been fixed.
1245 This port continues to use its own builtin globbing (not File::Glob).
1249 =head2 OS390 (OpenEdition MVS)
1251 Support for this EBCDIC platform has not been renewed in this release.
1252 There are difficulties in reconciling Perl's standardization on UTF-8
1253 as its internal representation for characters with the EBCDIC character
1254 set, because the two are incompatible.
1256 It is unclear whether future versions will renew support for this
1257 platform, but the possibility exists.
1261 Numerous revisions and extensions to configuration, build, testing, and
1262 installation process to accomodate core changes and VMS-specific options.
1264 Expand %ENV-handling code to allow runtime mapping to logical names,
1265 CLI symbols, and CRTL environ array.
1267 Extension of subprocess invocation code to accept filespecs as command
1270 Add to Perl command line processing the ability to use default file types and
1271 to recognize Unix-style C<2E<gt>&1>.
1273 Expansion of File::Spec::VMS routines, and integration into ExtUtils::MM_VMS.
1275 Extension of ExtUtils::MM_VMS to handle complex extensions more flexibly.
1277 Barewords at start of Unix-syntax paths may be treated as text rather than
1278 only as logical names.
1280 Optional secure translation of several logical names used internally by Perl.
1282 Miscellaneous bugfixing and porting of new core code to VMS.
1284 Thanks are gladly extended to the many people who have contributed VMS
1285 patches, testing, and ideas.
1289 Perl can now emulate fork() internally, using multiple interpreters running
1290 in different concurrent threads. This support must be enabled at build
1291 time. See L<perlfork> for detailed information.
1293 When given a pathname that consists only of a drivename, such as C<A:>,
1294 opendir() and stat() now use the current working directory for the drive
1295 rather than the drive root.
1297 The builtin XSUB functions in the Win32:: namespace are documented. See
1300 $^X now contains the full path name of the running executable.
1302 A Win32::GetLongPathName() function is provided to complement
1303 Win32::GetFullPathName() and Win32::GetShortPathName(). See L<Win32>.
1305 POSIX::uname() is supported.
1307 system(1,...) now returns true process IDs rather than process
1308 handles. kill() accepts any real process id, rather than strictly
1309 return values from system(1,...).
1311 For better compatibility with Unix, C<kill(0, $pid)> can now be used to
1312 test whether a process exists.
1314 The C<Shell> module is supported.
1316 Better support for building Perl under command.com in Windows 95
1319 Scripts are read in binary mode by default to allow ByteLoader (and
1320 the filter mechanism in general) to work properly. For compatibility,
1321 the DATA filehandle will be set to text mode if a carriage return is
1322 detected at the end of the line containing the __END__ or __DATA__
1323 token; if not, the DATA filehandle will be left open in binary mode.
1324 Earlier versions always opened the DATA filehandle in text mode.
1326 The glob() operator is implemented via the C<File::Glob> extension,
1327 which supports glob syntax of the C shell. This increases the flexibility
1328 of the glob() operator, but there may be compatibility issues for
1329 programs that relied on the older globbing syntax. If you want to
1330 preserve compatibility with the older syntax, you might want to run
1331 perl with C<-MFile::DosGlob>. For details and compatibility information,
1340 Compatibility tests for C<sub : attrs> vs the older C<use attrs>.
1344 Tests for new environment scalar capability (e.g., C<use Env qw($BAR);>).
1348 Tests for new environment array capability (e.g., C<use Env qw(@PATH);>).
1352 IO constants (SEEK_*, _IO*).
1356 Directory-related IO methods (new, read, close, rewind, tied delete).
1358 =item lib/io_multihomed
1360 INET sockets with multi-homed hosts.
1372 Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>.
1376 File test operators.
1380 Verify operations that access pad objects (lexicals and temporaries).
1384 Verify C<exists &sub> operations.
1388 =head1 Modules and Pragmata
1396 While used internally by Perl as a pragma, this module also
1397 provides a way to fetch subroutine and variable attributes.
1402 WARNING: The Compiler suite remains highly experimental. The
1403 generated code may not be correct, even it manages to execute
1406 The Perl Compiler suite has been extensively reworked for this
1407 release. More of the standard Perl testsuite passes when run
1408 under the Compiler, but there is still a significant way to
1409 go to achieve production quality compiled executables.
1413 The ByteLoader is a dedicated extension to generate and run
1414 Perl bytecode. See L<ByteLoader>.
1418 References can now be used.
1420 The new version also allows a leading underscore in constant names, but
1421 disallows a double leading underscore (as in "__LINE__"). Some other names
1422 are disallowed or warned against, including BEGIN, END, etc. Some names
1423 which were forced into main:: used to fail silently in some cases; now they're
1424 fatal (outside of main::) and an optional warning (inside of main::).
1425 The ability to detect whether a constant had been set with a given name has
1432 This pragma implements the C<\N> string escape. See L<charnames>.
1436 A C<Maxdepth> setting can be specified to avoid venturing
1437 too deeply into deep data structures. See L<Data::Dumper>.
1439 The XSUB implementation of Dump() is now automatically called if the
1440 C<Useqq> setting is not in use.
1442 Dumping C<qr//> objects works correctly.
1446 C<DB> is an experimental module that exposes a clean abstraction
1447 to Perl's debugging API.
1451 DB_File can now be built with Berkeley DB versions 1, 2 or 3.
1452 See C<ext/DB_File/Changes>.
1456 Devel::DProf, a Perl source code profiler has been added. See
1457 L<Devel::DProf> and L<dprofpp>.
1461 The Dumpvalue module provides screen dumps of Perl data.
1465 Overall, Benchmark results exhibit lower average error and better timing
1468 You can now run tests for I<n> seconds instead of guessing the right
1469 number of tests to run: e.g., timethese(-5, ...) will run each
1470 code for at least 5 CPU seconds. Zero as the "number of repetitions"
1471 means "for at least 3 CPU seconds". The output format has also
1472 changed. For example:
1474 use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
1476 will now output something like this:
1478 Benchmark: running a, b, each for at least 5 CPU seconds...
1479 a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
1480 b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
1482 New features: "each for at least N CPU seconds...", "wallclock secs",
1483 and the "@ operations/CPU second (n=operations)".
1485 timethese() now returns a reference to a hash of Benchmark objects containing
1486 the test results, keyed on the names of the tests.
1488 timethis() now returns the iterations field in the Benchmark result object
1491 timethese(), timethis(), and the new cmpthese() (see below) can also take
1492 a format specifier of 'none' to suppress output.
1494 A new function countit() is just like timeit() except that it takes a
1495 TIME instead of a COUNT.
1497 A new function cmpthese() prints a chart comparing the results of each test
1498 returned from a timethese() call. For each possible pair of tests, the
1499 percentage speed difference (iters/sec or seconds/iter) is shown.
1501 For other details, see L<Benchmark>.
1505 The Devel::Peek module provides access to the internal representation
1506 of Perl variables and data. It is a data debugging tool for the XS programmer.
1510 $PERL_VERSION now stands for C<$^V> (a string value) rather than for C<$]>
1515 Env now supports accessing environment variables like PATH as array
1520 More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
1521 large file (more than 4GB) access (NOTE: the O_LARGEFILE is
1522 automatically added to sysopen() flags if large file support has been
1523 configured, as is the default), Free/Net/OpenBSD locking behaviour
1524 flags F_FLOCK, F_POSIX, Linux F_SHLCK, and O_ACCMODE: the combined
1525 mask of O_RDONLY, O_WRONLY, and O_RDWR. The seek()/sysseek()
1526 constants SEEK_SET, SEEK_CUR, and SEEK_END are available via the
1527 C<:seek> tag. The chmod()/stat() S_IF* constants and S_IS* functions
1528 are available via the C<:mode> tag.
1532 A compare_text() function has been added, which allows custom
1533 comparison functions. See L<File::Compare>.
1537 File::Find now works correctly when the wanted() function is either
1538 autoloaded or is a symbolic reference.
1540 A bug that caused File::Find to lose track of the working directory
1541 when pruning top-level directories has been fixed.
1543 File::Find now also supports several other options to control its
1544 behavior. It can follow symbolic links if the C<follow> option is
1545 specified. Enabling the C<no_chdir> option will make File::Find skip
1546 changing the current directory when walking directories. The C<untaint>
1547 flag can be useful when running with taint checks enabled.
1553 This extension implements BSD-style file globbing. By default,
1554 it will also be used for the internal implementation of the glob()
1555 operator. See L<File::Glob>.
1559 New methods have been added to the File::Spec module: devnull() returns
1560 the name of the null device (/dev/null on Unix) and tmpdir() the name of
1561 the temp directory (normally /tmp on Unix). There are now also methods
1562 to convert between absolute and relative filenames: abs2rel() and
1563 rel2abs(). For compatibility with operating systems that specify volume
1564 names in file paths, the splitpath(), splitdir(), and catdir() methods
1567 =item File::Spec::Functions
1569 The new File::Spec::Functions modules provides a function interface
1570 to the File::Spec module. Allows shorthand
1572 $fullname = catfile($dir1, $dir2, $file);
1576 $fullname = File::Spec->catfile($dir1, $dir2, $file);
1580 Getopt::Long licensing has changed to allow the Perl Artistic License
1581 as well as the GPL. It used to be GPL only, which got in the way of
1582 non-GPL applications that wanted to use Getopt::Long.
1584 Getopt::Long encourages the use of Pod::Usage to produce help
1585 messages. For example:
1591 GetOptions('help|?' => \$help, man => \$man) or pod2usage(2);
1592 pod2usage(1) if $help;
1593 pod2usage(-exitstatus => 0, -verbose => 2) if $man;
1599 sample - Using GetOpt::Long and Pod::Usage
1603 sample [options] [file ...]
1606 -help brief help message
1607 -man full documentation
1615 Print a brief help message and exits.
1619 Prints the manual page and exits.
1625 B<This program> will read the given input file(s) and do someting
1626 useful with the contents thereof.
1630 See L<Pod::Usage> for details.
1632 A bug that prevented the non-option call-back <> from being
1633 specified as the first argument has been fixed.
1635 To specify the characters < and > as option starters, use ><. Note,
1636 however, that changing option starters is strongly deprecated.
1640 write() and syswrite() will now accept a single-argument
1641 form of the call, for consistency with Perl's syswrite().
1643 You can now create a TCP-based IO::Socket::INET without forcing
1644 a connect attempt. This allows you to configure its options
1645 (like making it non-blocking) and then call connect() manually.
1647 A bug that prevented the IO::Socket::protocol() accessor
1648 from ever returning the correct value has been corrected.
1650 IO::Socket::connect now uses non-blocking IO instead of alarm()
1651 to do connect timeouts.
1653 IO::Socket::accept now uses select() instead of alarm() for doing
1656 IO::Socket::INET->new now sets $! correctly on failure. $@ is
1657 still set for backwards compatability.
1661 Java Perl Lingo is now distributed with Perl. See jpl/README
1662 for more information.
1666 C<use lib> now weeds out any trailing duplicate entries.
1667 C<no lib> removes all named entries.
1671 The bitwise operations C<<< << >>>, C<<< >> >>>, C<&>, C<|>,
1672 and C<~> are now supported on bigints.
1676 The accessor methods Re, Im, arg, abs, rho, and theta can now also
1677 act as mutators (accessor $z->Re(), mutator $z->Re(3)).
1679 The class method C<display_format> and the corresponding object method
1680 C<display_format>, in addition to accepting just one argument, now can
1681 also accept a parameter hash. Recognized keys of a parameter hash are
1682 C<"style">, which corresponds to the old one parameter case, and two
1683 new parameters: C<"format">, which is a printf()-style format string
1684 (defaults usually to C<"%.15g">, you can revert to the default by
1685 setting the format string to C<undef>) used for both parts of a
1686 complex number, and C<"polar_pretty_print"> (defaults to true),
1687 which controls whether an attempt is made to try to recognize small
1688 multiples and rationals of pi (2pi, pi/2) at the argument (angle) of a
1689 polar complex number.
1691 The potentially disruptive change is that in list context both methods
1692 now I<return the parameter hash>, instead of only the value of the
1693 C<"style"> parameter.
1697 A little bit of radial trigonometry (cylindrical and spherical),
1698 radial coordinate conversions, and the great circle distance were added.
1700 =item Pod::Parser, Pod::InputObjects
1702 Pod::Parser is a base class for parsing and selecting sections of
1703 pod documentation from an input stream. This module takes care of
1704 identifying pod paragraphs and commands in the input and hands off the
1705 parsed paragraphs and commands to user-defined methods which are free
1706 to interpret or translate them as they see fit.
1708 Pod::InputObjects defines some input objects needed by Pod::Parser, and
1709 for advanced users of Pod::Parser that need more about a command besides
1712 As of release 5.6.0 of Perl, Pod::Parser is now the officially sanctioned
1713 "base parser code" recommended for use by all pod2xxx translators.
1714 Pod::Text (pod2text) and Pod::Man (pod2man) have already been converted
1715 to use Pod::Parser and efforts to convert Pod::HTML (pod2html) are already
1716 underway. For any questions or comments about pod parsing and translating
1717 issues and utilities, please use the pod-people@perl.org mailing list.
1719 For further information, please see L<Pod::Parser> and L<Pod::InputObjects>.
1721 =item Pod::Checker, podchecker
1723 This utility checks pod files for correct syntax, according to
1724 L<perlpod>. Obvious errors are flagged as such, while warnings are
1725 printed for mistakes that can be handled gracefully. The checklist is
1726 not complete yet. See L<Pod::Checker>.
1728 =item Pod::ParseUtils, Pod::Find
1730 These modules provide a set of gizmos that are useful mainly for pod
1731 translators. L<Pod::Find|Pod::Find> traverses directory structures and
1732 returns found pod files, along with their canonical names (like
1733 C<File::Spec::Unix>). L<Pod::ParseUtils|Pod::ParseUtils> contains
1734 B<Pod::List> (useful for storing pod list information), B<Pod::Hyperlink>
1735 (for parsing the contents of C<LE<lt>E<gt>> sequences) and B<Pod::Cache>
1736 (for caching information about pod files, e.g., link nodes).
1738 =item Pod::Select, podselect
1740 Pod::Select is a subclass of Pod::Parser which provides a function
1741 named "podselect()" to filter out user-specified sections of raw pod
1742 documentation from an input stream. podselect is a script that provides
1743 access to Pod::Select from other scripts to be used as a filter.
1746 =item Pod::Usage, pod2usage
1748 Pod::Usage provides the function "pod2usage()" to print usage messages for
1749 a Perl script based on its embedded pod documentation. The pod2usage()
1750 function is generally useful to all script authors since it lets them
1751 write and maintain a single source (the pods) for documentation, thus
1752 removing the need to create and maintain redundant usage message text
1753 consisting of information already in the pods.
1755 There is also a pod2usage script which can be used from other kinds of
1756 scripts to print usage messages from pods (even for non-Perl scripts
1757 with pods embedded in comments).
1759 For details and examples, please see L<Pod::Usage>.
1761 =item Pod::Text and Pod::Man
1763 Pod::Text has been rewritten to use Pod::Parser. While pod2text() is
1764 still available for backwards compatibility, the module now has a new
1765 preferred interface. See L<Pod::Text> for the details. The new Pod::Text
1766 module is easily subclassed for tweaks to the output, and two such
1767 subclasses (Pod::Text::Termcap for man-page-style bold and underlining
1768 using termcap information, and Pod::Text::Color for markup with ANSI color
1769 sequences) are now standard.
1771 pod2man has been turned into a module, Pod::Man, which also uses
1772 Pod::Parser. In the process, several outstanding bugs related to quotes
1773 in section headers, quoting of code escapes, and nested lists have been
1774 fixed. pod2man is now a wrapper script around this module.
1778 An EXISTS method has been added to this module (and sdbm_exists() has
1779 been added to the underlying sdbm library), so one can now call exists
1780 on an SDBM_File tied hash and get the correct result, rather than a
1783 A bug that may have caused data loss when more than one disk block
1784 happens to be read from the database in a single FETCH() has been
1789 Sys::Syslog now uses XSUBs to access facilities from syslog.h so it
1790 no longer requires syslog.ph to exist.
1794 Sys::Hostname now uses XSUBs to call the C library's gethostname() or
1795 uname() if they exist.
1797 =item Term::ANSIColor
1799 Term::ANSIColor is a very simple module to provide easy and readable
1800 access to the ANSI color and highlighting escape sequences, supported by
1801 most ANSI terminal emulators. It is now included standard.
1805 The timelocal() and timegm() functions used to silently return bogus
1806 results when the date fell outside the machine's integer range. They
1807 now consistently croak() if the date falls in an unsupported range.
1811 The error return value in list context has been changed for all functions
1812 that return a list of values. Previously these functions returned a list
1813 with a single element C<undef> if an error occurred. Now these functions
1814 return the empty list in these situations. This applies to the following
1820 The remaining functions are unchanged and continue to return C<undef> on
1821 error even in list context.
1823 The Win32::SetLastError(ERROR) function has been added as a complement
1824 to the Win32::GetLastError() function.
1826 The new Win32::GetFullPathName(FILENAME) returns the full absolute
1827 pathname for FILENAME in scalar context. In list context it returns
1828 a two-element list containing the fully qualified directory name and
1829 the filename. See L<Win32>.
1833 A new feature called "DBM Filters" has been added to all the
1834 DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
1835 DBM Filters add four new methods to each DBM module:
1842 These can be used to filter key-value pairs before the pairs are
1843 written to the database or just after they are read from the database.
1844 See L<perldbmfilter> for further information.
1850 C<use attrs> is now obsolete, and is only provided for
1851 backward-compatibility. It's been replaced by the C<sub : attributes>
1852 syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>.
1854 Lexical warnings pragma, C<use warnings;>, to control optional warnings.
1857 C<use filetest> to control the behaviour of filetests (C<-r> C<-w>
1858 ...). Currently only one subpragma implemented, "use filetest
1859 'access';", that uses access(2) or equivalent to check permissions
1860 instead of using stat(2) as usual. This matters in filesystems
1861 where there are ACLs (access control lists): the stat(2) might lie,
1862 but access(2) knows better.
1864 =head1 Utility Changes
1868 C<perlcc> now supports the C and Bytecode backends. By default,
1869 it generates output from the simple C backend rather than the
1870 optimized C backend.
1872 Support for non-Unix platforms has been improved.
1874 =head1 Documentation Changes
1880 The official list of public Perl API functions.
1882 =item perlcompile.pod
1884 An introduction to using the Perl Compiler suite.
1886 =item perlfilter.pod
1888 An introduction to writing Perl source filters.
1892 Some guidelines for hacking the Perl source code.
1894 =item perlintern.pod
1896 A list of internal functions in the Perl source code.
1897 (List is currently empty.)
1899 =item perlopentut.pod
1901 A tutorial on using open() effectively.
1903 =item perlreftut.pod
1905 A tutorial that introduces the essentials of references.
1909 A tutorial for beginners on object-oriented Perl.
1913 A tutorial on managing class data for object modules.
1915 =item perlunicode.pod
1917 An introduction to Unicode support features in Perl.
1921 =head1 New or Changed Diagnostics
1925 =item "%s" variable %s masks earlier declaration in same %s
1927 (W) A "my" or "our" variable has been redeclared in the current scope or statement,
1928 effectively eliminating all access to the previous instance. This is almost
1929 always a typographical error. Note that the earlier variable will still exist
1930 until the end of the scope or until all closure referents to it are
1933 =item "my sub" not yet implemented
1935 (F) Lexically scoped subroutines are not yet implemented. Don't try that
1938 =item "our" variable %s redeclared
1940 (W) You seem to have already declared the same global once before in the
1941 current lexical scope.
1943 =item '!' allowed only after types %s
1945 (F) The '!' is allowed in pack() and unpack() only after certain types.
1946 See L<perlfunc/pack>.
1948 =item / cannot take a count
1950 (F) You had an unpack template indicating a counted-length string,
1951 but you have also specified an explicit size for the string.
1952 See L<perlfunc/pack>.
1954 =item / must be followed by a, A or Z
1956 (F) You had an unpack template indicating a counted-length string,
1957 which must be followed by one of the letters a, A or Z
1958 to indicate what sort of string is to be unpacked.
1959 See L<perlfunc/pack>.
1961 =item / must be followed by a*, A* or Z*
1963 (F) You had a pack template indicating a counted-length string,
1964 Currently the only things that can have their length counted are a*, A* or Z*.
1965 See L<perlfunc/pack>.
1967 =item / must follow a numeric type
1969 (F) You had an unpack template that contained a '#',
1970 but this did not follow some numeric unpack specification.
1971 See L<perlfunc/pack>.
1973 =item /%s/: Unrecognized escape \\%c passed through
1975 (W) You used a backslash-character combination which is not recognized
1976 by Perl. This combination appears in an interpolated variable or a
1977 C<'>-delimited regular expression. The character was understood literally.
1979 =item /%s/: Unrecognized escape \\%c in character class passed through
1981 (W) You used a backslash-character combination which is not recognized
1982 by Perl inside character classes. The character was understood literally.
1984 =item /%s/ should probably be written as "%s"
1986 (W) You have used a pattern where Perl expected to find a string,
1987 as in the first argument to C<join>. Perl will treat the true
1988 or false result of matching the pattern against $_ as the string,
1989 which is probably not what you had in mind.
1991 =item %s() called too early to check prototype
1993 (W) You've called a function that has a prototype before the parser saw a
1994 definition or declaration for it, and Perl could not check that the call
1995 conforms to the prototype. You need to either add an early prototype
1996 declaration for the subroutine in question, or move the subroutine
1997 definition ahead of the call to get proper prototype checking. Alternatively,
1998 if you are certain that you're calling the function correctly, you may put
1999 an ampersand before the name to avoid the warning. See L<perlsub>.
2001 =item %s argument is not a HASH or ARRAY element
2003 (F) The argument to exists() must be a hash or array element, such as:
2006 $ref->[12]->["susie"]
2008 =item %s argument is not a HASH or ARRAY element or slice
2010 (F) The argument to delete() must be either a hash or array element, such as:
2013 $ref->[12]->["susie"]
2015 or a hash or array slice, such as:
2017 @foo[$bar, $baz, $xyzzy]
2018 @{$ref->[12]}{"susie", "queue"}
2020 =item %s argument is not a subroutine name
2022 (F) The argument to exists() for C<exists &sub> must be a subroutine
2023 name, and not a subroutine call. C<exists &sub()> will generate this error.
2025 =item %s package attribute may clash with future reserved word: %s
2027 (W) A lowercase attribute name was used that had a package-specific handler.
2028 That name might have a meaning to Perl itself some day, even though it
2029 doesn't yet. Perhaps you should use a mixed-case attribute name, instead.
2032 =item (in cleanup) %s
2034 (W) This prefix usually indicates that a DESTROY() method raised
2035 the indicated exception. Since destructors are usually called by
2036 the system at arbitrary points during execution, and often a vast
2037 number of times, the warning is issued only once for any number
2038 of failures that would otherwise result in the same message being
2041 Failure of user callbacks dispatched using the C<G_KEEPERR> flag
2042 could also result in this warning. See L<perlcall/G_KEEPERR>.
2044 =item <> should be quotes
2046 (F) You wrote C<< require <file> >> when you should have written
2049 =item Attempt to join self
2051 (F) You tried to join a thread from within itself, which is an
2052 impossible task. You may be joining the wrong thread, or you may
2053 need to move the join() to some other thread.
2055 =item Bad evalled substitution pattern
2057 (F) You've used the /e switch to evaluate the replacement for a
2058 substitution, but perl found a syntax error in the code to evaluate,
2059 most likely an unexpected right brace '}'.
2061 =item Bad realloc() ignored
2063 (S) An internal routine called realloc() on something that had never been
2064 malloc()ed in the first place. Mandatory, but can be disabled by
2065 setting environment variable C<PERL_BADFREE> to 1.
2067 =item Bareword found in conditional
2069 (W) The compiler found a bareword where it expected a conditional,
2070 which often indicates that an || or && was parsed as part of the
2071 last argument of the previous construct, for example:
2075 It may also indicate a misspelled constant that has been interpreted
2078 use constant TYPO => 1;
2079 if (TYOP) { print "foo" }
2081 The C<strict> pragma is useful in avoiding such errors.
2083 =item Binary number > 0b11111111111111111111111111111111 non-portable
2085 (W) The binary number you specified is larger than 2**32-1
2086 (4294967295) and therefore non-portable between systems. See
2087 L<perlport> for more on portability concerns.
2089 =item Bit vector size > 32 non-portable
2091 (W) Using bit vector sizes larger than 32 is non-portable.
2093 =item Buffer overflow in prime_env_iter: %s
2095 (W) A warning peculiar to VMS. While Perl was preparing to iterate over
2096 %ENV, it encountered a logical name or symbol definition which was too long,
2097 so it was truncated to the string shown.
2099 =item Can't check filesystem of script "%s"
2101 (P) For some reason you can't check the filesystem of the script for nosuid.
2103 =item Can't declare class for non-scalar %s in "%s"
2105 (S) Currently, only scalar variables can declared with a specific class
2106 qualifier in a "my" or "our" declaration. The semantics may be extended
2107 for other types of variables in future.
2109 =item Can't declare %s in "%s"
2111 (F) Only scalar, array, and hash variables may be declared as "my" or
2112 "our" variables. They must have ordinary identifiers as names.
2114 =item Can't ignore signal CHLD, forcing to default
2116 (W) Perl has detected that it is being run with the SIGCHLD signal
2117 (sometimes known as SIGCLD) disabled. Since disabling this signal
2118 will interfere with proper determination of exit status of child
2119 processes, Perl has reset the signal to its default value.
2120 This situation typically indicates that the parent program under
2121 which Perl may be running (e.g., cron) is being very careless.
2123 =item Can't modify non-lvalue subroutine call
2125 (F) Subroutines meant to be used in lvalue context should be declared as
2126 such, see L<perlsub/"Lvalue subroutines">.
2128 =item Can't read CRTL environ
2130 (S) A warning peculiar to VMS. Perl tried to read an element of %ENV
2131 from the CRTL's internal environment array and discovered the array was
2132 missing. You need to figure out where your CRTL misplaced its environ
2133 or define F<PERL_ENV_TABLES> (see L<perlvms>) so that environ is not searched.
2135 =item Can't remove %s: %s, skipping file
2137 (S) You requested an inplace edit without creating a backup file. Perl
2138 was unable to remove the original file to replace it with the modified
2139 file. The file was left unmodified.
2141 =item Can't return %s from lvalue subroutine
2143 (F) Perl detected an attempt to return illegal lvalues (such
2144 as temporary or readonly values) from a subroutine used as an lvalue.
2145 This is not allowed.
2147 =item Can't weaken a nonreference
2149 (F) You attempted to weaken something that was not a reference. Only
2150 references can be weakened.
2152 =item Character class [:%s:] unknown
2154 (F) The class in the character class [: :] syntax is unknown.
2157 =item Character class syntax [%s] belongs inside character classes
2159 (W) The character class constructs [: :], [= =], and [. .] go
2160 I<inside> character classes, the [] are part of the construct,
2161 for example: /[012[:alpha:]345]/. Note that [= =] and [. .]
2162 are not currently implemented; they are simply placeholders for
2165 =item Constant is not %s reference
2167 (F) A constant value (perhaps declared using the C<use constant> pragma)
2168 is being dereferenced, but it amounts to the wrong type of reference. The
2169 message indicates the type of reference that was expected. This usually
2170 indicates a syntax error in dereferencing the constant value.
2171 See L<perlsub/"Constant Functions"> and L<constant>.
2173 =item constant(%s): %%^H is not localized
2175 (F) When setting compile-time-lexicalized hash %^H one should set the
2176 corresponding bit of $^H as well.
2178 =item constant(%s): %s
2180 (F) Compile-time-substitutions (such as overloaded constants and
2181 character names) were not correctly set up.
2183 =item CORE::%s is not a keyword
2185 (F) The CORE:: namespace is reserved for Perl keywords.
2187 =item defined(@array) is deprecated
2189 (D) defined() is not usually useful on arrays because it checks for an
2190 undefined I<scalar> value. If you want to see if the array is empty,
2191 just use C<if (@array) { # not empty }> for example.
2193 =item defined(%hash) is deprecated
2195 (D) defined() is not usually useful on hashes because it checks for an
2196 undefined I<scalar> value. If you want to see if the hash is empty,
2197 just use C<if (%hash) { # not empty }> for example.
2199 =item Did not produce a valid header
2203 =item Did you mean "local" instead of "our"?
2205 (W) Remember that "our" does not localize the declared global variable.
2206 You have declared it again in the same lexical scope, which seems superfluous.
2208 =item Document contains no data
2212 =item entering effective %s failed
2214 (F) While under the C<use filetest> pragma, switching the real and
2215 effective uids or gids failed.
2217 =item false [] range "%s" in regexp
2219 (W) A character class range must start and end at a literal character, not
2220 another character class like C<\d> or C<[:alpha:]>. The "-" in your false
2221 range is interpreted as a literal "-". Consider quoting the "-", "\-".
2224 =item Filehandle %s opened only for output
2226 (W) You tried to read from a filehandle opened only for writing. If you
2227 intended it to be a read/write filehandle, you needed to open it with
2228 "+<" or "+>" or "+>>" instead of with "<" or nothing. If
2229 you intended only to read from the file, use "<". See
2232 =item flock() on closed filehandle %s
2234 (W) The filehandle you're attempting to flock() got itself closed some
2235 time before now. Check your logic flow. flock() operates on filehandles.
2236 Are you attempting to call flock() on a dirhandle by the same name?
2238 =item Global symbol "%s" requires explicit package name
2240 (F) You've said "use strict vars", which indicates that all variables
2241 must either be lexically scoped (using "my"), declared beforehand using
2242 "our", or explicitly qualified to say which package the global variable
2245 =item Hexadecimal number > 0xffffffff non-portable
2247 (W) The hexadecimal number you specified is larger than 2**32-1
2248 (4294967295) and therefore non-portable between systems. See
2249 L<perlport> for more on portability concerns.
2251 =item Ill-formed CRTL environ value "%s"
2253 (W) A warning peculiar to VMS. Perl tried to read the CRTL's internal
2254 environ array, and encountered an element without the C<=> delimiter
2255 used to spearate keys from values. The element is ignored.
2257 =item Ill-formed message in prime_env_iter: |%s|
2259 (W) A warning peculiar to VMS. Perl tried to read a logical name
2260 or CLI symbol definition when preparing to iterate over %ENV, and
2261 didn't see the expected delimiter between key and value, so the
2264 =item Illegal binary digit %s
2266 (F) You used a digit other than 0 or 1 in a binary number.
2268 =item Illegal binary digit %s ignored
2270 (W) You may have tried to use a digit other than 0 or 1 in a binary number.
2271 Interpretation of the binary number stopped before the offending digit.
2273 =item Illegal number of bits in vec
2275 (F) The number of bits in vec() (the third argument) must be a power of
2276 two from 1 to 32 (or 64, if your platform supports that).
2278 =item Integer overflow in %s number
2280 (W) The hexadecimal, octal or binary number you have specified either
2281 as a literal or as an argument to hex() or oct() is too big for your
2282 architecture, and has been converted to a floating point number. On a
2283 32-bit architecture the largest hexadecimal, octal or binary number
2284 representable without overflow is 0xFFFFFFFF, 037777777777, or
2285 0b11111111111111111111111111111111 respectively. Note that Perl
2286 transparently promotes all numbers to a floating point representation
2287 internally--subject to loss of precision errors in subsequent
2290 =item Invalid %s attribute: %s
2292 The indicated attribute for a subroutine or variable was not recognized
2293 by Perl or by a user-supplied handler. See L<attributes>.
2295 =item Invalid %s attributes: %s
2297 The indicated attributes for a subroutine or variable were not recognized
2298 by Perl or by a user-supplied handler. See L<attributes>.
2300 =item invalid [] range "%s" in regexp
2302 The offending range is now explicitly displayed.
2304 =item Invalid separator character %s in attribute list
2306 (F) Something other than a colon or whitespace was seen between the
2307 elements of an attribute list. If the previous attribute
2308 had a parenthesised parameter list, perhaps that list was terminated
2309 too soon. See L<attributes>.
2311 =item Invalid separator character %s in subroutine attribute list
2313 (F) Something other than a colon or whitespace was seen between the
2314 elements of a subroutine attribute list. If the previous attribute
2315 had a parenthesised parameter list, perhaps that list was terminated
2318 =item leaving effective %s failed
2320 (F) While under the C<use filetest> pragma, switching the real and
2321 effective uids or gids failed.
2323 =item Lvalue subs returning %s not implemented yet
2325 (F) Due to limitations in the current implementation, array and hash
2326 values cannot be returned in subroutines used in lvalue context.
2327 See L<perlsub/"Lvalue subroutines">.
2329 =item Method %s not permitted
2333 =item Missing %sbrace%s on \N{}
2335 (F) Wrong syntax of character name literal C<\N{charname}> within
2336 double-quotish context.
2338 =item Missing command in piped open
2340 (W) You used the C<open(FH, "| command")> or C<open(FH, "command |")>
2341 construction, but the command was missing or blank.
2343 =item Missing name in "my sub"
2345 (F) The reserved syntax for lexically scoped subroutines requires that they
2346 have a name with which they can be found.
2348 =item No %s specified for -%c
2350 (F) The indicated command line switch needs a mandatory argument, but
2351 you haven't specified one.
2353 =item No package name allowed for variable %s in "our"
2355 (F) Fully qualified variable names are not allowed in "our" declarations,
2356 because that doesn't make much sense under existing semantics. Such
2357 syntax is reserved for future extensions.
2359 =item No space allowed after -%c
2361 (F) The argument to the indicated command line switch must follow immediately
2362 after the switch, without intervening spaces.
2364 =item no UTC offset information; assuming local time is UTC
2366 (S) A warning peculiar to VMS. Perl was unable to find the local
2367 timezone offset, so it's assuming that local system time is equivalent
2368 to UTC. If it's not, define the logical name F<SYS$TIMEZONE_DIFFERENTIAL>
2369 to translate to the number of seconds which need to be added to UTC to
2372 =item Octal number > 037777777777 non-portable
2374 (W) The octal number you specified is larger than 2**32-1 (4294967295)
2375 and therefore non-portable between systems. See L<perlport> for more
2376 on portability concerns.
2378 See also L<perlport> for writing portable code.
2380 =item panic: del_backref
2382 (P) Failed an internal consistency check while trying to reset a weak
2385 =item panic: kid popen errno read
2387 (F) forked child returned an incomprehensible message about its errno.
2389 =item panic: magic_killbackrefs
2391 (P) Failed an internal consistency check while trying to reset all weak
2392 references to an object.
2394 =item Parentheses missing around "%s" list
2396 (W) You said something like
2402 my ($foo, $bar) = @_;
2404 Remember that "my", "our", and "local" bind tighter than comma.
2406 =item Possible Y2K bug: %s
2408 (W) You are concatenating the number 19 with another number, which
2409 could be a potential Year 2000 problem.
2411 =item Premature end of script headers
2415 =item Repeat count in pack overflows
2417 (F) You can't specify a repeat count so large that it overflows
2418 your signed integers. See L<perlfunc/pack>.
2420 =item Repeat count in unpack overflows
2422 (F) You can't specify a repeat count so large that it overflows
2423 your signed integers. See L<perlfunc/unpack>.
2425 =item realloc() of freed memory ignored
2427 (S) An internal routine called realloc() on something that had already
2430 =item Reference is already weak
2432 (W) You have attempted to weaken a reference that is already weak.
2433 Doing so has no effect.
2435 =item setpgrp can't take arguments
2437 (F) Your system has the setpgrp() from BSD 4.2, which takes no arguments,
2438 unlike POSIX setpgid(), which takes a process ID and process group ID.
2440 =item Strange *+?{} on zero-length expression
2442 (W) You applied a regular expression quantifier in a place where it
2443 makes no sense, such as on a zero-width assertion.
2444 Try putting the quantifier inside the assertion instead. For example,
2445 the way to match "abc" provided that it is followed by three
2446 repetitions of "xyz" is C</abc(?=(?:xyz){3})/>, not C</abc(?=xyz){3}/>.
2448 =item switching effective %s is not implemented
2450 (F) While under the C<use filetest> pragma, we cannot switch the
2451 real and effective uids or gids.
2453 =item This Perl can't reset CRTL environ elements (%s)
2455 =item This Perl can't set CRTL environ elements (%s=%s)
2457 (W) Warnings peculiar to VMS. You tried to change or delete an element
2458 of the CRTL's internal environ array, but your copy of Perl wasn't
2459 built with a CRTL that contained the setenv() function. You'll need to
2460 rebuild Perl with a CRTL that does, or redefine F<PERL_ENV_TABLES> (see
2461 L<perlvms>) so that the environ array isn't the target of the change to
2462 %ENV which produced the warning.
2464 =item Unknown open() mode '%s'
2466 (F) The second argument of 3-argument open() is not among the list
2467 of valid modes: C<< < >>, C<< > >>, C<<< >> >>>, C<< +< >>,
2468 C<< +> >>, C<<< +>> >>>, C<-|>, C<|->.
2470 =item Unknown process %x sent message to prime_env_iter: %s
2472 (P) An error peculiar to VMS. Perl was reading values for %ENV before
2473 iterating over it, and someone else stuck a message in the stream of
2474 data Perl expected. Someone's very confused, or perhaps trying to
2475 subvert Perl's population of %ENV for nefarious purposes.
2477 =item Unrecognized escape \\%c passed through
2479 (W) You used a backslash-character combination which is not recognized
2480 by Perl. The character was understood literally.
2482 =item Unterminated attribute parameter in attribute list
2484 (F) The lexer saw an opening (left) parenthesis character while parsing an
2485 attribute list, but the matching closing (right) parenthesis
2486 character was not found. You may need to add (or remove) a backslash
2487 character to get your parentheses to balance. See L<attributes>.
2489 =item Unterminated attribute list
2491 (F) The lexer found something other than a simple identifier at the start
2492 of an attribute, and it wasn't a semicolon or the start of a
2493 block. Perhaps you terminated the parameter list of the previous attribute
2494 too soon. See L<attributes>.
2496 =item Unterminated attribute parameter in subroutine attribute list
2498 (F) The lexer saw an opening (left) parenthesis character while parsing a
2499 subroutine attribute list, but the matching closing (right) parenthesis
2500 character was not found. You may need to add (or remove) a backslash
2501 character to get your parentheses to balance.
2503 =item Unterminated subroutine attribute list
2505 (F) The lexer found something other than a simple identifier at the start
2506 of a subroutine attribute, and it wasn't a semicolon or the start of a
2507 block. Perhaps you terminated the parameter list of the previous attribute
2510 =item Value of CLI symbol "%s" too long
2512 (W) A warning peculiar to VMS. Perl tried to read the value of an %ENV
2513 element from a CLI symbol table, and found a resultant string longer
2514 than 1024 characters. The return value has been truncated to 1024
2517 =item Version number must be a constant number
2519 (P) The attempt to translate a C<use Module n.n LIST> statement into
2520 its equivalent C<BEGIN> block found an internal inconsistency with
2525 =head1 Obsolete Diagnostics
2529 =item Character class syntax [: :] is reserved for future extensions
2531 (W) Within regular expression character classes ([]) the syntax beginning
2532 with "[:" and ending with ":]" is reserved for future extensions.
2533 If you need to represent those character sequences inside a regular
2534 expression character class, just quote the square brackets with the
2535 backslash: "\[:" and ":\]".
2537 =item Ill-formed logical name |%s| in prime_env_iter
2539 (W) A warning peculiar to VMS. A logical name was encountered when preparing
2540 to iterate over %ENV which violates the syntactic rules governing logical
2541 names. Because it cannot be translated normally, it is skipped, and will not
2542 appear in %ENV. This may be a benign occurrence, as some software packages
2543 might directly modify logical name tables and introduce nonstandard names,
2544 or it may indicate that a logical name table has been corrupted.
2546 =item Probable precedence problem on %s
2548 (W) The compiler found a bareword where it expected a conditional,
2549 which often indicates that an || or && was parsed as part of the
2550 last argument of the previous construct, for example:
2554 =item regexp too big
2556 (F) The current implementation of regular expressions uses shorts as
2557 address offsets within a string. Unfortunately this means that if
2558 the regular expression compiles to longer than 32767, it'll blow up.
2559 Usually when you want a regular expression this big, there is a better
2560 way to do it with multiple statements. See L<perlre>.
2562 =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
2564 (D) Perl versions before 5.004 misinterpreted any type marker followed
2565 by "$" and a digit. For example, "$$0" was incorrectly taken to mean
2566 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
2568 However, the developers of Perl 5.004 could not fix this bug completely,
2569 because at least two widely-used modules depend on the old meaning of
2570 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
2571 old (broken) way inside strings; but it generates this message as a
2572 warning. And in Perl 5.005, this special treatment will cease.
2578 If you find what you think is a bug, you might check the
2579 articles recently posted to the comp.lang.perl.misc newsgroup.
2580 There may also be information at http://www.perl.com/perl/, the Perl
2583 If you believe you have an unreported bug, please run the B<perlbug>
2584 program included with your release. Be sure to trim your bug down
2585 to a tiny but sufficient test case. Your bug report, along with the
2586 output of C<perl -V>, will be sent off to perlbug@perl.com to be
2587 analysed by the Perl porting team.
2591 The F<Changes> file for exhaustive details on what changed.
2593 The F<INSTALL> file for how to build Perl.
2595 The F<README> file for general stuff.
2597 The F<Artistic> and F<Copying> files for copyright information.
2601 Written by Gurusamy Sarathy <F<gsar@activestate.com>>, with many
2602 contributions from The Perl Porters.
2604 Send omissions or corrections to <F<perlbug@perl.com>>.