3 perldelta - what's new for perl v5.6 (as of v5.005_64)
7 This is an unsupported alpha release, meant for intrepid Perl
8 developers only. The included sources may not even build correctly on
9 some platforms. Subscribing to perl5-porters is the best way to
10 monitor and contribute to the progress of development releases (see
11 http://www.hut.fi/~jhi/perl5-porters.html for info).
13 This document describes differences between the 5.005 release and this one.
15 =head1 Incompatible Changes
17 =head2 Perl Source Incompatibilities
19 Beware that any new warnings that have been added or old ones
20 that have been enhanced are B<not> considered incompatible changes.
22 Since all new warnings must be explicitly requested via the C<-w>
23 switch or the C<warnings> pragma, it is ultimately the programmer's
24 responsibility to ensure that warnings are enabled judiciously.
28 =item CHECK is a new keyword
30 In addition to C<BEGIN>, C<INIT>, C<END>, C<DESTROY> and C<AUTOLOAD>,
31 subroutines named C<CHECK> are now special. These are queued up during
32 compilation and behave similar to END blocks, except they are called at
33 the end of compilation rather than at the end of execution. They cannot
36 =item Treatment of list slices of undef has changed
38 When taking a slice of a literal list (as opposed to a slice of
39 an array or hash), Perl used to return an empty list if the
40 result happened to be composed of all undef values.
42 The new behavior is to produce an empty list if (and only if)
43 the original list was empty. Consider the following example:
45 @a = (1,undef,undef,2)[2,1,2];
47 The old behavior would have resulted in @a having no elements.
48 The new behavior ensures it has three undefined elements.
50 Note in particular that the behavior of slices of the following
51 cases remains unchanged:
55 @a = (anything_returning_empty_list())[2,1,2];
61 =head2 Perl's version numbering has changed
63 Beginning with Perl version 5.6, the version number convention has been
64 changed to a "dotted integer" scheme that is more commonly found in open
67 Maintenance versions of v5.6.0 will be released as v5.6.1, v5.6.2 etc.
68 The next development series following v5.6 will be numbered v5.7.x,
69 beginning with v5.7.0, and the next major production release following
72 The English module now sets $PERL_VERSION to $^V (a string value) rather
73 than C<$]> (a numeric value). (This is a potential incompatibility.
74 Send us a report via perlbug if you are affected by this.)
76 The v1.2.3 syntax is also now legal in Perl.
77 See L<Support for strings represented as a vector of ordinals> for more on that.
79 To cope with the new versioning system's use of at least three significant
80 digits for each version component, the method used for incrementing the
81 subversion number has also changed slightly. We assume that versions older
82 than v5.6 have been incrementing the subversion component in multiples of
83 10. Versions after v5.6.0 will increment them by 1. Thus, using the new
84 notation, 5.005_03 is the same as v5.5.30, and the first maintenance
85 version following v5.6.0 will be v5.6.1, which amounts to a floating point
88 =item Literals of the form C<1.2.3> parse differently
90 Previously, numeric literals with more than one dot in them were
91 interpreted as a floating point number concatenated with one or more
92 numbers. Such "numbers" are now parsed as strings composed of the
95 For example, C<print 97.98.99> used to output C<97.9899> in earlier
96 versions, but now prints C<abc>.
98 See L<Support for strings represented as a vector of ordinals> below.
100 =item Possibly changed pseudo-random number generator
102 In 5.005_0x and earlier, perl's rand() function used the C library
103 rand(3) function. As of 5.005_52, Configure tests for drand48(),
104 random(), and rand() (in that order) and picks the first one it finds.
105 Perl programs that depend on reproducing a specific set of pseudo-random
106 numbers will now likely produce different output. You can use
107 C<sh Configure -Drandfunc=rand> to obtain the old behavior.
109 =item Hashing function for hash keys has changed
111 Perl hashes are not order preserving. The apparently random order
112 encountered when iterating on the contents of a hash is determined
113 by the hashing algorithm used. To improve the distribution of lower
114 bits in the hashed value, the algorithm has changed slightly as of
115 5.005_52. When iterating over hashes, this may yield a random order
116 that is B<different> from that of previous versions.
118 =item C<undef> fails on read only values
120 Using the C<undef> operator on a readonly value (such as $1) has
121 the same effect as assigning C<undef> to the readonly value--it
124 =item Close-on-exec bit may be set on pipe and socket handles
126 On systems that support a close-on-exec flag on filehandles, the
127 flag will be set for any handles created by pipe(), socketpair(),
128 socket(), and accept(), if that is warranted by the value of $^F
129 that may be in effect. Earlier versions neglected to set the flag
130 for handles created with these operators. See L<perlfunc/pipe>,
131 L<perlfunc/socketpair>, L<perlfunc/socket>, L<perlfunc/accept>,
134 =item Writing C<"$$1"> to mean C<"${$}1"> is unsupported
136 Perl 5.004 deprecated the interpretation of C<$$1> and
137 similar within interpolated strings to mean C<$$ . "1">,
138 but still allowed it.
140 In Perl 5.6 and later, C<"$$1"> always means C<"${$1}">.
142 =item delete(), values() and C<\(%h)> operate on aliases to values, not copies
144 delete(), each(), values() and hashes in a list context return the actual
145 values in the hash, instead of copies (as they used to in earlier
146 versions). Typical idioms for using these constructs copy the
147 returned values, but this can make a significant difference when
148 creating references to the returned values.
150 Keys in the hash are still returned as copies when iterating on
153 =item vec(EXPR,OFFSET,BITS) enforces powers-of-two BITS
155 vec() generates a run-time error if the BITS argument is not
156 a valid power-of-two integer.
158 =item Text of some diagnostic output has changed
160 Most references to internal Perl operations in diagnostics
161 have been changed to be more descriptive. This may be an
162 issue for programs that may incorrectly rely on the exact
163 text of diagnostics for proper functioning.
165 =item C<%@> has been removed
167 The undocumented special variable C<%@> that used to accumulate
168 "background" errors (such as those that happen in DESTROY())
169 has been removed, because it could potentially result in memory
172 =item Parenthesized not() behaves like a list operator
174 The C<not> operator now falls under the "if it looks like a function,
175 it behaves like a function" rule.
177 As a result, the parenthesized form can be used with C<grep> and C<map>.
178 The following construct used to be a syntax error before, but it works
181 grep not($_), @things;
183 On the other hand, using C<not> with a literal list slice may not
184 work. The following previously allowed construct:
186 print not (1,2,3)[0];
188 needs to be written with additional parentheses now:
190 print not((1,2,3)[0]);
192 The behavior remains unaffected when C<not> is not followed by parentheses.
194 =item Semantics of bareword prototype C<(*)> have changed
196 Arguments prototyped as C<*> will now be visible within the subroutine
197 as either a simple scalar or as a reference to a typeglob. Perl 5.005
198 always coerced simple scalar arguments to a typeglob, which wasn't useful
199 in situations where the subroutine must distinguish between a simple
200 scalar and a typeglob. See L<perlsub/Prototypes>.
202 =head2 On 64-bit platforms the semantics of bit operators have changed
204 If your platform is either natively 64-bit or your Perl has been
205 configured to used 64-bit integers (say C<perl -V> and see what is
206 your ivsize: if it is 8, you are 64-bit) , be warned that the
207 semantics of all the bitwise numeric operators (& | ^ ~ << >>) have
208 been changed. They used to be forced to be 32 bits wide, but now in
209 the aforementioned platforms they are 64 bits wide. Most dramatically
210 this affects the unary ~: what used to be 32 bits wide, is now 64 bits
211 wide. If you depend on your integers being 32 bits wide, mask off the
212 excess bits with C<& 0xffffffff>.
216 =head2 C Source Incompatibilities
220 =item C<PERL_POLLUTE>
222 Release 5.005 grandfathered old global symbol names by providing preprocessor
223 macros for extension source compatibility. As of release 5.6, these
224 preprocessor definitions are not available by default. You need to explicitly
225 compile perl with C<-DPERL_POLLUTE> to get these definitions. For
226 extensions still using the old symbols, this option can be
227 specified via MakeMaker:
229 perl Makefile.PL POLLUTE=1
231 =item C<PERL_IMPLICIT_CONTEXT>
233 PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built
234 with one of -Dusethreads, -Dusemultiplicity, or both. It is not
235 intended to be enabled by users at this time.
237 This new build option provides a set of macros for all API functions
238 such that an implicit interpreter/thread context argument is passed to
239 every API function. As a result of this, something like C<sv_setsv(foo,bar)>
240 amounts to a macro invocation that actually translates to something like
241 C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
242 to not have any significant source compatibility issues, the difference
243 between a macro and a real function call will need to be considered.
245 This means that there B<is> a source compatibility issue as a result of
246 this if your extensions attempt to use pointers to any of the Perl API
249 Note that the above issue is not relevant to the default build of
250 Perl, whose interfaces continue to match those of prior versions
251 (but subject to the other options described here).
253 See L<perlguts/"The Perl API"> for detailed information on the
254 ramifications of building Perl using this option.
256 =item C<PERL_POLLUTE_MALLOC>
258 Enabling Perl's malloc in release 5.005 and earlier caused
259 the namespace of system versions of the malloc family of functions to
260 be usurped by the Perl versions, since by default they used the
263 Besides causing problems on platforms that do not allow these functions to
264 be cleanly replaced, this also meant that the system versions could not
265 be called in programs that used Perl's malloc. Previous versions of Perl
266 have allowed this behaviour to be suppressed with the HIDEMYMALLOC and
267 EMBEDMYMALLOC preprocessor definitions.
269 As of release 5.6, Perl's malloc family of functions have default names
270 distinct from the system versions. You need to explicitly compile perl with
271 C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
272 and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
275 Note that these functions do B<not> constitute Perl's memory allocation API.
276 See L<perlguts/"Memory Allocation"> for further information about that.
280 =head2 Compatible C Source API Changes
284 =item C<PATCHLEVEL> is now C<PERL_VERSION>
286 The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
287 are now available by default from perl.h, and reflect the base revision,
288 patchlevel, and subversion respectively. C<PERL_REVISION> had no
289 prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
290 previously available as C<PATCHLEVEL> and C<SUBVERSION>.
292 The new names cause less pollution of the B<cpp> namespace and reflect what
293 the numbers have come to stand for in common practice. For compatibility,
294 the old names are still supported when F<patchlevel.h> is explicitly
295 included (as required before), so there is no source incompatibility
298 =item Support for C++ exceptions
300 change#3386, also needs perlguts documentation
301 [TODO - Chip Salzenberg <chip@perlsupport.com>]
305 =head2 Binary Incompatibilities
307 In general, the default build of this release is expected to be binary
308 compatible for extensions built with the 5.005 release or its maintenance
309 versions. However, specific platforms may have broken binary compatibility
310 due to changes in the defaults used in hints files. Therefore, please be
311 sure to always check the platform-specific README files for any notes to
314 The usethreads or usemultiplicity builds are B<not> binary compatible
315 with the corresponding builds in 5.005.
317 On platforms that require an explicit list of exports (AIX, OS/2 and Windows,
318 among others), purely internal symbols such as parser functions and the
319 run time opcodes are not exported by default. Perl 5.005 used to export
320 all functions irrespective of whether they were considered part of the
323 For the full list of public API functions, see L<perlapi>.
325 =head1 Installation and Configuration Improvements
327 =head2 -Dusethreads means something different
329 WARNING: Support for threads continues to be an experimental feature.
330 Interfaces and implementation are subject to sudden and drastic changes.
332 The -Dusethreads flag now enables the experimental interpreter-based thread
333 support by default. To get the flavor of experimental threads that was in
334 5.005 instead, you need to run Configure with "-Dusethreads -Duse5005threads".
336 As of v5.5.640, interpreter-threads support is still lacking a way to
337 create new threads from Perl (i.e., C<use Thread;> will not work with
338 interpreter threads). C<use Thread;> continues to be available when you
339 ask for use5005threads, bugs and all.
341 =head2 New Configure flags
343 The following new flags may be enabled on the Configure command line
344 by running Configure with C<-Dflag>.
347 usethreads useithreads (new interpreter threads: no Perl API yet)
348 usethreads use5005threads (threads as they were in 5.005)
350 use64bitint (equal to now deprecated 'use64bits')
356 usesocks (only SOCKS v5 supported)
358 =head2 Threadedness and 64-bitness now more daring
360 The Configure options enabling the use of threads and the use of
361 64-bitness are now more daring in the sense that they no more have an
362 explicit list of operating systems of known threads/64-bit
363 capabilities. In other words: if your operating system has the
364 necessary APIs and datatypes, you should be able just to go ahead and
365 use them, for threads by Configure -Dusethreads, and for 64 bits
366 either explicitly by Configure -Duse64bitint or implicitly if your
367 system has 64 bit wide datatypes. See also L<"64-bit support">.
371 Some platforms have "long doubles", floating point numbers of even
372 larger range than ordinary "doubles". To enable using long doubles for
373 Perl's scalars, use -Duselongdouble.
377 You can enable both -Duse64bitint and -Dlongdouble by -Dusemorebits.
378 See also L<"64-bit support">.
380 =head2 -Duselargefiles
382 Some platforms support large files, files larger than two gigabytes.
383 See L<"Large file support"> for more information.
385 =head2 installusrbinperl
387 You can use "Configure -Uinstallusrbinperl" which causes installperl
388 to skip installing perl also as /usr/bin/perl. This is useful if you
389 prefer not to modify /usr/bin for some reason or another but harmful
390 because many scripts assume to find Perl in /usr/bin/perl.
394 You can use "Configure -Dusesocks" which causes Perl to probe
395 for the SOCKS (v5, not v4) proxy protocol library,
396 http://www.socks.nec.com/
400 You can "post-edit" the Configure variables using the Configure C<-A>
401 flag. The editing happens immediately after the platform specific
402 hints files have been processed but before the actual configuration
403 process starts. Run C<Configure -h> to find out the full C<-A> syntax.
405 =head2 Enhanced Installation Directories
407 The installation structure has been enriched to improve the support
408 for maintaining multiple versions of perl, to provide locations for
409 vendor-supplied modules, scripts, and manpages, and to ease maintenance
410 of locally-added modules, scripts, and manpages. See the section on
411 Installation Directories in the INSTALL file for complete details.
412 For most users building and installing from source, the defaults should
415 If you previously used C<Configure -Dsitelib> or C<-Dsitearch> to set
416 special values for library directories, you might wish to consider using
417 the new C<-Dsiteprefix> setting instead. Also, if you wish to re-use a
418 config.sh file from an earlier version of perl, you should be sure to
419 check that Configure makes sensible choices for the new directories.
420 See INSTALL for complete details.
424 =head2 Unicode and UTF-8 support
426 Perl can optionally use UTF-8 as its internal representation for character
427 strings. The C<utf8> and C<bytes> pragmas are used to control this support
428 in the current lexical scope. See L<perlunicode>, L<utf8> and L<bytes> for
431 =head2 Interpreter cloning, threads, and concurrency
433 WARNING: This is an experimental feature in a pre-alpha state. Use
436 Perl 5.005_63 introduces the beginnings of support for running multiple
437 interpreters concurrently in different threads. In conjunction with
438 the perl_clone() API call, which can be used to selectively duplicate
439 the state of any given interpreter, it is possible to compile a
440 piece of code once in an interpreter, clone that interpreter
441 one or more times, and run all the resulting interpreters in distinct
444 On Windows, this feature is used to emulate fork() at the interpreter
445 level. See L<perlfork>.
447 This feature is still in evolution. It is eventually meant to be used
448 to selectively clone a subroutine and data reachable from that
449 subroutine in a separate interpreter and run the cloned subroutine
450 in a separate thread. Since there is no shared data between the
451 interpreters, little or no locking will be needed (unless parts of
452 the symbol table are explicitly shared). This is obviously intended
453 to be an easy-to-use replacement for the existing threads support.
455 Support for cloning interpreters and interpreter concurrency can be
456 enabled using the -Dusethreads Configure option (see win32/Makefile for
457 how to enable it on Windows.) The resulting perl executable will be
458 functionally identical to one that was built with -Dmultiplicity, but
459 the perl_clone() API call will only be available in the former.
461 -Dusethreads enables, the cpp macros USE_ITHREADS by default, which enables
462 Perl source code changes that provide a clear separation between the op tree
463 and the data it operates with. The former is considered immutable, and can
464 therefore be shared between an interpreter and all of its clones, while the
465 latter is considered local to each interpreter, and is therefore copied for
468 Note that building Perl with the -Dusemultiplicity Configure option
469 is adequate if you wish to run multiple B<independent> interpreters
470 concurrently in different threads. -Dusethreads only provides the
471 additional functionality of the perl_clone() API call and other
472 support for running B<cloned> interpreters concurrently.
474 [XXX TODO - the Compiler backends may be broken when USE_ITHREADS is
477 =head2 Lexically scoped warning categories
479 You can now control the granularity of warnings emitted by perl at a finer
480 level using the C<use warnings> pragma. See L<warnings> and L<perllexwarn>
483 =head2 Lvalue subroutines
485 WARNING: This is an experimental feature.
488 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>,
489 Tuomas Lukka <lukka@iki.fi>)]
491 =head2 "our" declarations
493 An "our" declaration introduces a value that can be best understood
494 as a lexically scoped symbolic alias to a global variable in the
495 package that was current where the variable was declared. This is
496 mostly useful as an alternative to the C<vars> pragma, but also provides
497 the opportunity to introduce typing and other attributes for such
498 variables. See L<perlfunc/our>.
500 =head2 Support for strings represented as a vector of ordinals
502 Literals of the form C<v1.2.3.4> are now parsed as a string composed of
503 of characters with the specified ordinals. This is an alternative, more
504 readable way to construct (possibly unicode) strings instead of
505 interpolating characters, as in C<"\x{1}\x{2}\x{3}\x{4}">. The leading
506 C<v> may be omitted if there are more than two ordinals, so C<1.2.3> is
507 parsed the same as C<v1.2.3>.
509 Strings written in this form are also useful to represent version "numbers".
510 It is easy to compare such version "numbers" (which are really just plain
511 strings) using any of the usual string comparison operators C<eq>, C<ne>,
512 C<lt>, C<gt>, etc., or perform bitwise string operations on them using C<|>,
515 In conjunction with the new C<$^V> magic variable (which contains
516 the perl version as a string), such literals can be used as a readable way
517 to check if you're running a particular version of Perl:
519 # this will parse in older versions of Perl also
520 if ($^V and $^V gt v5.5.640) {
521 # new features supported
524 C<require> and C<use> also have some special magic to support such literals.
525 They will be interpreted as a version rather than as a module name:
527 require v5.6.0; # croak if $^V lt v5.6.0
528 use v5.6.0; # same, but croaks at compile-time
530 Alternatively, the C<v> may be omitted if there is more than one dot:
535 Also, C<sprintf> and C<printf> support the Perl-specific format flag C<%v>
536 to print ordinals of characters in arbitrary strings:
538 printf "v%vd", $^V; # prints current version, such as "v5.5.650"
539 printf "%*vX", ":", $addr; # formats IPv6 address
540 printf "%*vb", " ", $bits; # displays bitstring
542 See L<perldata/"Scalar value constructors"> for additional information.
544 =head2 Weak references
546 WARNING: This is an experimental feature.
548 In previous versions of Perl, you couldn't cache objects so as
549 to allow them to be deleted if the last reference from outside
550 the cache is deleted. The reference in the cache would hold a
551 reference count on the object and the objects would never be
554 Another familiar problem is with circular references. When an
555 object references itself, its reference count would never go
556 down to zero, and it would not get destroyed until the program
559 Weak references solve this by allowing you to "weaken" any
560 reference, that is, make it not count towards the reference count.
561 When the last non-weak reference to an object is deleted, the object
562 is destroyed and all the weak references to the object are
563 automatically undef-ed.
565 To use this feature, you need the WeakRef package from CPAN, which
566 contains additional documentation.
568 change#3385, also need perlguts documentation
569 [TODO - Tuomas Lukka <lukka@iki.fi>]
571 =head2 File globbing implemented internally
573 WARNING: This is currently an experimental feature. Interfaces and
574 implementation are likely to change.
576 Perl now uses the File::Glob implementation of the glob() operator
577 automatically. This avoids using an external csh process and the
578 problems associated with it.
580 =head2 Binary numbers supported
582 Binary numbers are now supported as literals, in s?printf formats, and
586 printf "The answer is: %b\n", oct("0b101010");
588 =head2 Some arrows may be omitted in calls through references
590 Perl now allows the arrow to be omitted in many constructs
591 involving subroutine calls through references. For example,
592 C<$foo[10]-E<gt>('foo')> may now be written C<$foo[10]('foo')>.
593 This is rather similar to how the arrow may be omitted from
594 C<$foo[10]-E<gt>{'foo'}>. Note however, that the arrow is still
595 required for C<foo(10)-E<gt>('bar')>.
597 =head2 exists() is supported on subroutine names
599 The exists() builtin now works on subroutine names. A subroutine
600 is considered to exist if it has been declared (even if implicitly).
601 See L<perlfunc/exists> for examples.
603 =head2 exists() and delete() are supported on array elements
605 The exists() and delete() builtins now work on simple arrays as well.
606 The behavior is similar to that on hash elements.
608 exists() can be used to check whether an array element has been
609 initialized. This avoids autovivifying array elements that don't exist.
610 If the array is tied, the EXISTS() method in the corresponding tied
611 package will be invoked.
613 delete() may be used to remove an element from the array and return
614 it. The array element at that position returns to its unintialized
615 state, so that testing for the same element with exists() will return
616 false. If the element happens to be the one at the end, the size of
617 the array also shrinks up to the highest element that tests true for
618 exists(), or 0 if none such is found. If the array is tied, the DELETE()
619 method in the corresponding tied package will be invoked.
621 See L<perlfunc/exists> and L<perlfunc/delete> for examples.
623 =head2 syswrite() ease-of-use
625 The length argument of C<syswrite()> has become optional.
627 =head2 File and directory handles can be autovivified
629 Similar to how constructs such as C<$x-E<gt>[0]> autovivify a reference,
630 handle constructors (open(), opendir(), pipe(), socketpair(), sysopen(),
631 socket(), and accept()) now autovivify a file or directory handle
632 if the handle passed to them is an uninitialized scalar variable. This
633 allows the constructs such as C<open(my $fh, ...)> and C<open(local $fh,...)>
634 to be used to create filehandles that will conveniently be closed
635 automatically when the scope ends, provided there are no other references
636 to them. This largely eliminates the need for typeglobs when opening
637 filehandles that must be passed around, as in the following example:
641 or die "Can't open '@_': $!";
646 my $f = myopen("</etc/motd");
648 # $f implicitly closed here
651 [TODO - this idiom needs more pod penetration]
653 =head2 64-bit support
655 NOTE: The Configure flags -Duselonglong and -Duse64bits
656 have been deprecated. Use -Duse64bitint instead.
658 Any platform that has 64-bit integers either (a) natively as longs or
659 ints (b) via special compiler flags (c) using long long are able to
660 use "quads" (64-integers) as follows:
666 constants (decimal, hexadecimal, octal, binary) in the code
670 arguments to oct() and hex()
674 arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)
682 pack() and unpack() "q" and "Q" formats
686 in basic arithmetics: + - * / % (NOTE: operating close to the limits
687 of the integer values may produce surprising results)
691 in bit arithmetics: & | ^ ~ << >> (NOTE: these used to be forced
700 Note that unless you have the case (a) you will have to configure
701 and compile Perl using the -Duse64bitint Configure flag.
703 There are actually two modes of 64-bitness: the first one is achieved
704 using Configure -Duse64bitint and the second one using Configure
705 -Duse64bitall. The difference is that the first one is minimal and
706 the second one maximal. The first one does only as much as is
707 required to get 64-bit integers into Perl (this may mean, for example,
708 using "long longs") while your memory may still be limited to 2
709 gigabytes (because your pointers most likely are 32-bit); the second
710 one goes all the way by attempting to switch also longs (and pointers)
711 being 64-bit. This may create an even more binary incompatible Perl
712 than -Duse64bitint: the resulting executable may not run at all in a
713 CPU-bit box, or you may have to reboot/reconfigure/rebuild your
714 operating system to be 64-bit aware.
716 Natively 64-bit systems like Alpha and Cray need neither -Duse64bitint
719 Last but not least: note that due to Perl's habit of always using
720 floating point numbers the quads are still not true integers.
721 When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned,
722 -9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they
723 are silently promoted to floating point numbers, after which they will
724 start losing precision (their lower digits).
726 =head2 Large file support
728 If you have filesystems that support "large files" (files larger than
729 2 gigabytes), you may now also be able to create and access them from
730 Perl. You have to use Configure -Duselargefiles. Turning on the
731 large file support turns on also the 64-bit support on many platforms.
732 Beware that unless your filesystem also supports "sparse files" seeking
733 to umpteen petabytes may be unadvisable.
735 Note that in addition to requiring a proper file system to do large
736 files you may also need to adjust your per-process (or your
737 per-system, or per-process-group, or per-user-group) maximum filesize
738 limits before running Perl scripts that try to handle large files,
739 especially if you intend to write such files.
741 Finally, in addition to your process/process group maximum filesize
742 limits, you may have quota limits on your filesystems that stop you
743 (your user id or your user group id) from using large files.
745 Adjusting your process/user/group/file system/operating system limits
746 is outside the scope of Perl core language. For process limits, you
747 may try increasing the limits using your shell's limits/limit/ulimit
748 command before running Perl. The BSD::Resource extension (not
749 included with the standard Perl distribution) may also be of use, it
750 offers the getrlimit/setrlimit interface that can be used to adjust
751 process resource usage limits, including the maximum filesize limit.
755 In some systems you may be able to use long doubles to enhance the
756 range and precision of your double precision floating point numbers
757 (that is, Perl's numbers). Use Configure -Duselongdouble to enable
758 this support (if it is available).
762 You can "Configure -Dusemorebits" to turn on both the 64-bit support
763 and the long double support.
765 =head2 Enhanced support for sort() subroutines
767 Perl subroutines with a prototype of C<($$)> and XSUBs in general can
768 now be used as sort subroutines. In either case, the two elements to
769 be compared are passed as normal parameters in @_. See L<perlfunc/sort>.
771 For unprototyped sort subroutines, the historical behavior of passing
772 the elements to be compared as the global variables $a and $b remains
775 =head2 Better syntax checks on parenthesized unary operators
779 print defined(&foo,&bar,&baz);
780 print uc("foo","bar","baz");
783 used to be accidentally allowed in earlier versions, and produced
784 unpredictable behaviour. Some produced ancillary warnings
785 when used in this way; others silently did the wrong thing.
787 The parenthesized forms of most unary operators that expect a single
788 argument now ensure that they are not called with more than one
789 argument, making the cases shown above syntax errors. The usual
792 print defined &foo, &bar, &baz;
793 print uc "foo", "bar", "baz";
796 remains unchanged. See L<perlop>.
798 =head2 POSIX character class syntax [: :] supported
800 For example to match alphabetic characters use /[[:alpha:]]/.
801 See L<perlre> for details.
803 =head2 Improved C<qw//> operator
805 The C<qw//> operator is now evaluated at compile time into a true list
806 instead of being replaced with a run time call to C<split()>. This
807 removes the confusing misbehaviour of C<qw//> in scalar context, which
808 had inherited that behaviour from split().
812 $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
814 now correctly prints "3|a", instead of "2|a".
816 =head2 pack() format 'Z' supported
818 The new format type 'Z' is useful for packing and unpacking null-terminated
819 strings. See L<perlfunc/"pack">.
821 =head2 pack() format modifier '!' supported
823 The new format type modifier '!' is useful for packing and unpacking
824 native shorts, ints, and longs. See L<perlfunc/"pack">.
826 =head2 pack() and unpack() support counted strings
828 The template character '/' can be used to specify a counted string
829 type to be packed or unpacked. See L<perlfunc/"pack">.
831 =head2 Comments in pack() templates
833 The '#' character in a template introduces a comment up to
834 end of the line. This facilitates documentation of pack()
837 =head2 $^X variables may now have names longer than one character
839 Formerly, $^X was synonymous with ${"\cX"}, but $^XY was a syntax
840 error. Now variable names that begin with a control character may be
841 arbitrarily long. However, for compatibility reasons, these variables
842 I<must> be written with explicit braces, as C<${^XY}> for example.
843 C<${^XYZ}> is synonymous with ${"\cXYZ"}. Variable names with more
844 than one control character, such as C<${^XY^Z}>, are illegal.
846 The old syntax has not changed. As before, `^X' may be either a
847 literal control-X character or the two-character sequence `caret' plus
848 `X'. When braces are omitted, the variable name stops after the
849 control character. Thus C<"$^XYZ"> continues to be synonymous with
850 C<$^X . "YZ"> as before.
852 As before, lexical variables may not have names beginning with control
853 characters. As before, variables whose names begin with a control
854 character are always forced to be in package `main'. All such variables
855 are reserved for future extensions, except those that begin with
856 C<^_>, which may be used by user programs and are guaranteed not to
857 acquire special meaning in any future version of Perl.
859 =head2 C<use attrs> implicit in subroutine attributes
861 Formerly, if you wanted to mark a subroutine as being a method call or
862 as requiring an automatic lock() when it is entered, you had to declare
863 that with a C<use attrs> pragma in the body of the subroutine.
864 That can now be accomplished with declaration syntax, like this:
866 sub mymethod : locked method ;
868 sub mymethod : locked method {
872 sub othermethod :locked :method ;
874 sub othermethod :locked :method {
879 (Note how only the first C<:> is mandatory, and whitespace surrounding
880 the C<:> is optional.)
882 F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes
883 with the stubs they provide. See L<attributes>.
885 =head2 Regular expression improvements
887 change#2827,2373,2372,2365,1813,1800,4112,4158,4215,4301
888 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
890 =head2 Overloading improvements
893 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
895 =head2 open() with more than two arguments
897 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
899 =head2 Support for interpolating named characters
902 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
904 =head2 Experimental support for user-hooks in @INC
906 [TODO - Ken Fox <kfox@ford.com>]
908 =head2 C<require> and C<do> may be overridden
910 C<require> and C<do 'file'> operations may be overridden locally
911 by importing subroutines of the same name into the current package
912 (or globally by importing them into the CORE::GLOBAL:: namespace).
913 Overriding C<require> will also affect C<use>, provided the override
914 is visible at compile-time.
915 See L<perlsub/"Overriding Built-in Functions">.
917 =head2 New variable $^C reflects C<-c> switch
919 C<$^C> has a boolean value that reflects whether perl is being run
920 in compile-only mode (i.e. via the C<-c> switch). Since
921 BEGIN blocks are executed under such conditions, this variable
922 enables perl code to determine whether actions that make sense
923 only during normal running are warranted. See L<perlvar>.
925 =head2 New variable $^V contains Perl version in v5.6.0 format
927 C<$^V> contains the Perl version number as a string composed of
928 characters whose ordinals match the version numbers, so that it may
929 be used in string comparisons.
931 See C<Support for strings represented as a vector of ordinals> for an
934 =head2 Optional Y2K warnings
936 If Perl is built with the cpp macro C<PERL_Y2KWARN> defined,
937 it emits optional warnings when concatenating the number 19
940 This behavior must be specifically enabled when running Configure.
941 See F<INSTALL> and F<README.Y2K>.
943 =head1 Significant bug fixes
945 =head2 E<lt>HANDLEE<gt> on empty files
947 With C<$/> set to C<undef>, "slurping" an empty file returns a string of
948 zero length (instead of C<undef>, as it used to) the first time the
949 HANDLE is read after C<$/> is set to C<undef>. Further reads yield
952 This means that the following will append "foo" to an empty file (it used
955 perl -0777 -pi -e 's/^/foo/' empty_file
959 perl -pi -e 's/^/foo/' empty_file
961 is unchanged (it continues to leave the file empty).
963 =head2 C<eval '...'> improvements
965 Line numbers (as reflected by caller() and most diagnostics) within
966 C<eval '...'> were often incorrect when here documents were involved.
967 This has been corrected.
969 Lexical lookups for variables appearing in C<eval '...'> within
970 functions that were themselves called within an C<eval '...'> were
971 searching the wrong place for lexicals. The lexical search now
972 correctly ends at the subroutine's block boundary.
974 Parsing of here documents used to be flawed when they appeared as
975 the replacement expression in C<eval 's/.../.../e'>. This has
978 =head2 All compilation errors are true errors
980 Some "errors" encountered at compile time were by neccessity
981 generated as warnings followed by eventual termination of the
982 program. This enabled more such errors to be reported in a
983 single run, rather than causing a hard stop at the first error
984 that was encountered.
986 The mechanism for reporting such errors has been reimplemented
987 to queue compile-time errors and report them at the end of the
988 compilation as true errors rather than as warnings. This fixes
989 cases where error messages leaked through in the form of warnings
990 when code was compiled at run time using C<eval STRING>, and
991 also allows such errors to be reliably trapped using __DIE__ hooks.
993 =head2 Automatic flushing of output buffers
995 fork(), exec(), system(), qx//, and pipe open()s now flush buffers
996 of all files opened for output when the operation
997 was attempted. This mostly eliminates confusing
998 buffering mishaps suffered by users unaware of how Perl internally
1001 =head2 Better diagnostics on meaningless filehandle operations
1003 Constructs such as C<open(E<lt>FHE<gt>)> and C<close(E<lt>FHE<gt>)>
1004 are compile time errors. Attempting to read from filehandles that
1005 were opened only for writing will now produce warnings (just as
1006 writing to read-only filehandles does).
1008 =head2 Where possible, buffered data discarded from duped input filehandle
1010 C<open(NEW, "E<lt>&OLD")> now attempts to discard any data that
1011 was previously read and buffered in C<OLD> before duping the handle.
1012 On platforms where doing this is allowed, the next read operation
1013 on C<NEW> will return the same data as the corresponding operation
1014 on C<OLD>. Formerly, it would have returned the data from the start
1015 of the following disk block instead.
1017 =head2 eof() has the same old magic as <>
1019 C<eof()> would return true if no attempt to read from C<E<lt>E<gt>> had
1020 yet been made. C<eof()> has been changed to have a little magic of its
1021 own, it now opens the C<E<lt>E<gt>> files.
1023 =head2 system(), backticks and pipe open now reflect exec() failure
1025 On Unix and similar platforms, system(), qx() and open(FOO, "cmd |")
1026 etc., are implemented via fork() and exec(). When the underlying
1027 exec() fails, earlier versions did not report the error properly,
1028 since the exec() happened to be in a different process.
1030 The child process now communicates with the parent about the
1031 error in launching the external command, which allows these
1032 constructs to return with their usual error value and set $!.
1034 =head2 Implicitly closed filehandles are safer
1036 Sometimes implicitly closed filehandles (as when they are localized,
1037 and Perl automatically closes them on exiting the scope) could
1038 inadvertently set $? or $!. This has been corrected.
1040 =head2 C<(\$)> prototype and C<$foo{a}>
1042 An scalar reference prototype now correctly allows a hash or
1043 array element in that slot.
1045 =head2 Pseudo-hashes work better
1047 Dereferencing some types of reference values in a pseudo-hash,
1048 such as C<$ph-E<gt>{foo}[1]>, was accidentally disallowed. This has
1051 When applied to a pseudo-hash element, exists() now reports whether
1052 the specified value exists, not merely if the key is valid.
1054 delete() now works on pseudo-hashes. When given a pseudo-hash element
1055 or slice it deletes the values corresponding to the keys (but not the keys
1056 themselves). See L<perlref/"Pseudo-hashes: Using an array as a hash">.
1058 Pseudo-hash slices with constant keys are now optimized to array lookups
1061 The C<fields> pragma now provides ways to create pseudo-hashes, via
1062 fields::new() and fields::phash(). See L<fields>.
1064 =head2 C<goto &sub> and AUTOLOAD
1066 The C<goto &sub> construct works correctly when C<&sub> happens
1069 =head2 C<-bareword> allowed under C<use integer>
1071 The autoquoting of barewords preceded by C<-> did not work
1072 in prior versions when the C<integer> pragma was enabled.
1073 This has been fixed.
1075 =head2 Boolean assignment operators are legal lvalues
1077 Constructs such as C<($a ||= 2) += 1> are now allowed.
1079 =head2 C<sort $coderef @foo> allowed
1081 sort() did not accept a subroutine reference as the comparison
1082 function in earlier versions. This is now permitted.
1084 =head2 Failures in DESTROY()
1086 When code in a destructor threw an exception, it went unnoticed
1087 in earlier versions of Perl, unless someone happened to be
1088 looking in $@ just after the point the destructor happened to
1089 run. Such failures are now visible as warnings when warnings are
1092 =head2 Locale bugs fixed
1094 printf() and sprintf() previously reset the numeric locale
1095 back to the default "C" locale. This has been fixed.
1097 Numbers formatted according to the local numeric locale
1098 (such as using a decimal comma instead of a decimal dot) caused
1099 "isn't numeric" warnings, even while the operations accessing
1100 those numbers produced correct results. The warnings are gone.
1104 The C<eval 'return sub {...}'> construct could sometimes leak
1105 memory. This has been fixed.
1107 Operations that aren't filehandle constructors used to leak memory
1108 when used on invalid filehandles. This has been fixed.
1110 Constructs that modified C<@_> could fail to deallocate values
1111 in C<@_> and thus leak memory. This has been corrected.
1113 =head2 Spurious subroutine stubs after failed subroutine calls
1115 Perl could sometimes create empty subroutine stubs when a
1116 subroutine was not found in the package. Such cases stopped
1117 later method lookups from progressing into base packages.
1118 This has been corrected.
1120 =head2 Consistent numeric conversions
1123 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1125 =head2 Taint failures under C<-U>
1127 When running in unsafe mode, taint violations could sometimes
1128 cause silent failures. This has been fixed.
1130 =head2 END blocks and the C<-c> switch
1132 Prior versions used to run BEGIN B<and> END blocks when Perl was
1133 run in compile-only mode. Since this is typically not the expected
1134 behavior, END blocks are not executed anymore when the C<-c> switch
1137 See L<CHECK blocks> for how to run things when the compile phase ends.
1139 =head2 Potential to leak DATA filehandles
1141 Using the C<__DATA__> token creates an implicit filehandle to
1142 the file that contains the token. It is the program's
1143 responsibility to close it when it is done reading from it.
1145 This caveat is now better explained in the documentation.
1148 =head2 Diagnostics follow STDERR
1150 Diagnostic output now goes to whichever file the C<STDERR> handle
1151 is pointing at, instead of always going to the underlying C runtime
1152 library's C<stderr>.
1154 =head2 Other fixes for better diagnostics
1156 Line numbers are no longer suppressed (under most likely circumstances)
1157 during the global destruction phase.
1159 Diagnostics emitted from code running in threads other than the main
1160 thread are now accompanied by the thread ID.
1162 Embedded null characters in diagnostics now actually show up. They
1163 used to truncate the message in prior versions.
1165 $foo::a and $foo::b are now exempt from "possible typo" warnings only
1166 if sort() is encountered in package foo.
1168 Unrecognized alphabetic escapes encountered when parsing quote
1169 constructs now generate a warning, since they may take on new
1170 semantics in later versions of Perl.
1172 =head1 Performance enhancements
1174 =head2 Simple sort() using { $a <=> $b } and the like are optimized
1176 Many common sort() operations using a simple inlined block are now
1177 optimized for faster performance.
1179 =head2 Optimized assignments to lexical variables
1181 Certain operations in the RHS of assignment statements have been
1182 optimized to directly set the lexical variable on the LHS,
1183 eliminating redundant copying overheads.
1185 =head2 Method lookups optimized
1187 [TODO - Chip Salzenberg <chip@perlsupport.com>]
1189 =head2 Faster mechanism to invoke XSUBs
1192 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1194 =head2 Perl_malloc() improvements
1197 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1199 =head2 Faster subroutine calls
1201 Minor changes in how subroutine calls are handled internally
1202 provide marginal improvements in performance.
1204 =head1 Platform specific changes
1206 =head2 Additional supported platforms
1212 VM/ESA is now supported.
1216 Siemens BS2000 is now supported under the POSIX Shell.
1220 The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
1225 GNU/Hurd is now supported.
1229 Rhapsody is now supported.
1233 EPOC is is now supported (on Psion 5).
1243 Perl now works with djgpp 2.02 (and 2.03 alpha).
1247 Environment variable names are not converted to uppercase any more.
1251 Wrong exit code from backticks now fixed.
1255 This port is still using its own builtin globbing.
1261 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1265 Numerous revisions and extensions to configuration, build, testing, and
1266 installation process to accomodate core changes and VMS-specific options
1268 Expand %ENV-handling code to allow runtime mapping to logical names,
1269 CLI symbols, and CRTL environ array
1271 Extension of subprocess invocation code to accept filespecs as command "verbs"
1273 Add to Perl command line processing the ability to use default file types and
1274 to recognize Unix-style C<2E<gt>&1>.
1276 Expansion of File::Spec::VMS routines, and integration into ExtUtils::MM_VMS
1278 Extension of ExtUtils::MM_VMS to handle complex extensions more flexibly
1280 Barewords at start of Unix-syntax paths may be treated as text rather than
1281 only as logical names
1283 Optional secure translation of several logical names used internally by Perl
1285 Miscellaneous bugfixing and porting of new core code to VMS
1287 Thanks are gladly extended to the many people who have contributed VMS
1288 patches, testing, and ideas.
1292 Site library searches failed to look for ".../site/5.XXX/lib"
1293 if ".../site/5.XXXYY/lib" wasn't found. This has been corrected.
1295 When given a pathname that consists only of a drivename, such
1296 as C<A:>, opendir() and stat() now use the current working
1297 directory for the drive rather than the drive root.
1299 The builtin XSUB functions in the Win32:: namespace are
1300 documented. See L<Win32>.
1302 $^X now contains the full path name of the running executable.
1304 A Win32::GetLongPathName() function is provided to complement
1305 Win32::GetFullPathName() and Win32::GetShortPathName(). See L<Win32>.
1307 POSIX::uname() is supported.
1309 system(1,...) now returns true process IDs rather than process
1310 handles. kill() accepts any real process id, rather than strictly
1311 return values from system(1,...).
1313 For better compatibility with Unix, C<kill(0, $pid)> can now be used to
1314 test whether a process exists.
1316 The C<Shell> module is supported.
1318 Rudimentary support for building under command.com in Windows 95
1321 Scripts are read in binary mode by default to allow ByteLoader (and
1322 the filter mechanism in general) to work properly. For compatibility,
1323 the DATA filehandle will be set to text mode if a carriage return is
1324 detected at the end of the line containing the __END__ or __DATA__
1325 token; if not, the DATA filehandle will be left open in binary mode.
1326 Earlier versions always opened the DATA filehandle in text mode.
1328 The glob() operator is implemented via the C<File::Glob> extension,
1329 which supports glob syntax of the C shell. This increases the flexibility
1330 of the glob() operator, but there may be compatibility issues for
1331 programs that relied on the older globbing syntax. If you want to
1332 preserve compatibility with the older syntax, you might want to put
1333 a C<use File::DosGlob;> in your program. For details and compatibility
1334 information, see L<File::Glob>.
1344 Compatibility tests for C<sub : attrs> vs the older C<use attrs>.
1348 Tests for new environment scalar capability (e.g., C<use Env qw($BAR);>).
1352 Tests for new environment array capability (e.g., C<use Env qw(@PATH);>).
1356 IO constants (SEEK_*, _IO*).
1360 Directory-related IO methods (new, read, close, rewind, tied delete).
1362 =item lib/io_multihomed
1364 INET sockets with multi-homed hosts.
1376 Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>.
1380 File test operators.
1384 Verify operations that access pad objects (lexicals and temporaries).
1388 Verify C<exists &sub> operations.
1392 =head1 Modules and Pragmata
1400 While used internally by Perl as a pragma, this module also
1401 provides a way to fetch subroutine and variable attributes.
1406 The Perl Compiler suite has been extensively reworked for this
1409 [TODO - Vishal Bhatia <vishal@gol.com>,
1410 Nick Ing-Simmons <nick@ni-s.u-net.com>]
1414 The ByteLoader is a dedicated extension to generate and run
1415 Perl bytecode. See L<ByteLoader>.
1419 References can now be used.
1421 The new version also allows a leading underscore in constant names, but
1422 disallows a double leading underscore (as in "__LINE__"). Some other names
1423 are disallowed or warned against, including BEGIN, END, etc. Some names
1424 which were forced into main:: used to fail silently in some cases; now they're
1425 fatal (outside of main::) and an optional warning (inside of main::).
1426 The ability to detect whether a constant had been set with a given name has
1434 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1438 A C<Maxdepth> setting can be specified to avoid venturing
1439 too deeply into deep data structures. See L<Data::Dumper>.
1441 Dumping C<qr//> objects works correctly.
1445 C<DB> is an experimental module that exposes a clean abstraction
1446 to Perl's debugging API.
1450 DB_File can now be built with Berkeley DB versions 1, 2 or 3.
1451 See C<ext/DB_File/Changes>.
1455 Devel::DProf, a Perl source code profiler has been added. See
1456 L<Devel::DProf> and L<dprofpp>.
1460 The Dumpvalue module provides screen dumps of Perl data.
1464 Overall, Benchmark results exhibit lower average error and better timing
1467 You can now run tests for I<n> seconds instead of guessing the right
1468 number of tests to run: e.g. timethese(-5, ...) will run each
1469 code for at least 5 CPU seconds. Zero as the "number of repetitions"
1470 means "for at least 3 CPU seconds". The output format has also
1471 changed. For example:
1473 use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
1475 will now output something like this:
1477 Benchmark: running a, b, each for at least 5 CPU seconds...
1478 a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
1479 b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
1481 New features: "each for at least N CPU seconds...", "wallclock secs",
1482 and the "@ operations/CPU second (n=operations)".
1484 timethese() now returns a reference to a hash of Benchmark objects containing
1485 the test results, keyed on the names of the tests.
1487 timethis() now returns the iterations field in the Benchmark result object
1490 timethese(), timethis(), and the new cmpthese() (see below) can also take
1491 a format specifier of 'none' to suppress output.
1493 A new function countit() is just like timeit() except that it takes a
1494 TIME instead of a COUNT.
1496 A new function cmpthese() prints a chart comparing the results of each test
1497 returned from a timethese() call. For each possible pair of tests, the
1498 percentage speed difference (iters/sec or seconds/iter) is shown.
1500 For other details, see L<Benchmark>.
1504 The Devel::Peek module provides access to the internal representation
1505 of Perl variables and data. It is a data debugging tool for the XS programmer.
1509 $PERL_VERSION now stands for C<$^V> (a string value) rather than for C<$]>
1514 Env now supports accessing environment variables like PATH as array
1517 =item ExtUtils::MakeMaker
1519 change#4135, also needs docs in module pod
1520 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1524 More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
1525 large file (more than 4GB) access Note that the O_LARGEFILE is
1526 automatically/transparently added to sysopen() flags if large file
1527 support has been configured), Free/Net/OpenBSD locking behaviour flags
1528 F_FLOCK, F_POSIX, Linux F_SHLCK, and O_ACCMODE: the combined mask of
1529 O_RDONLY, O_WRONLY, and O_RDWR. The seek()/sysseek() constants
1530 SEEK_SET, SEEK_CUR, and SEEK_END are available via the C<:seek> tag.
1531 The chmod()/stat() S_IF* constants and S_IS* functions are available
1532 via the C<:mode> tag.
1537 A compare_text() function has been added, which allows custom
1538 comparison functions. See L<File::Compare>.
1542 File::Find now works correctly when the wanted() function is either
1543 autoloaded or is a symbolic reference.
1545 A bug that caused File::Find to lose track of the working directory
1546 when pruning top-level directories has been fixed.
1548 File::Find now also supports several other options to control its
1549 behavior. It can follow symbolic links if the C<follow> option is
1550 specified. Enabling the C<no_chdir> option will make File::Find skip
1551 changing the current directory when walking directories. The C<untaint>
1552 flag can be useful when running with taint checks enabled.
1558 This extension implements BSD-style file globbing. By default,
1559 it will also be used for the internal implementation of the glob()
1560 operator. See L<File::Glob>.
1564 New methods have been added to the File::Spec module: devnull() returns
1565 the name of the null device (/dev/null on Unix) and tmpdir() the name of
1566 the temp directory (normally /tmp on Unix). There are now also methods
1567 to convert between absolute and relative filenames: abs2rel() and
1568 rel2abs(). For compatibility with operating systems that specify volume
1569 names in file paths, the splitpath(), splitdir(), and catdir() methods
1572 =item File::Spec::Functions
1574 The new File::Spec::Functions modules provides a function interface
1575 to the File::Spec module. Allows shorthand
1577 $fullname = catfile($dir1, $dir2, $file);
1581 $fullname = File::Spec->catfile($dir1, $dir2, $file);
1585 Getopt::Long licensing has changed to allow the Perl Artistic License
1586 as well as the GPL. It used to be GPL only, which got in the way of
1587 non-GPL applications that wanted to use Getopt::Long.
1589 Getopt::Long encourages the use of Pod::Usage to produce help
1590 messages. For example:
1596 GetOptions('help|?' => \$help, man => \$man) or pod2usage(2);
1597 pod2usage(1) if $help;
1598 pod2usage(-exitstatus => 0, -verbose => 2) if $man;
1604 sample - Using GetOpt::Long and Pod::Usage
1608 sample [options] [file ...]
1611 -help brief help message
1612 -man full documentation
1620 Print a brief help message and exits.
1624 Prints the manual page and exits.
1630 B<This program> will read the given input file(s) and do someting
1631 useful with the contents thereof.
1635 See L<Pod::Usage> for details.
1637 A bug that prevented the non-option call-back E<lt>E<gt> from being
1638 specified as the first argument has been fixed.
1640 To specify the characters E<lt> and E<gt> as option starters, use
1641 E<gt>E<lt>. Note, however, that changing option starters is strongly
1646 write() and syswrite() will now accept a single-argument
1647 form of the call, for consistency with Perl's syswrite().
1649 You can now create a TCP-based IO::Socket::INET without forcing
1650 a connect attempt. This allows you to configure its options
1651 (like making it non-blocking) and then call connect() manually.
1653 A bug that prevented the IO::Socket::protocol() accessor
1654 from ever returning the correct value has been corrected.
1656 IO::Socket::connect now uses non-blocking IO instead of alarm()
1657 to do connect timeouts.
1659 IO::Socket::accept now uses select() instead of alarm() for doing
1662 IO::Socket::INET->new now sets $! correctly on failure. $@ is
1663 still set for backwards compatability.
1667 Java Perl Lingo is now distributed with Perl. See jpl/README
1668 for more information.
1672 C<use lib> now weeds out any trailing duplicate entries.
1673 C<no lib> removes all named entries.
1677 The bitwise operations C<E<lt>E<lt>>, C<E<gt>E<gt>>, C<&>, C<|>,
1678 and C<~> are now supported on bigints.
1682 The accessor methods Re, Im, arg, abs, rho, and theta can now also
1683 act as mutators (accessor $z->Re(), mutator $z->Re(3)).
1687 A little bit of radial trigonometry (cylindrical and spherical),
1688 radial coordinate conversions, and the great circle distance were added.
1690 =item Pod::Parser, Pod::InputObjects
1692 Pod::Parser is a base class for parsing and selecting sections of
1693 pod documentation from an input stream. This module takes care of
1694 identifying pod paragraphs and commands in the input and hands off the
1695 parsed paragraphs and commands to user-defined methods which are free
1696 to interpret or translate them as they see fit.
1698 Pod::InputObjects defines some input objects needed by Pod::Parser, and
1699 for advanced users of Pod::Parser that need more about a command besides
1702 As of release 5.6 of Perl, Pod::Parser is now the officially sanctioned
1703 "base parser code" recommended for use by all pod2xxx translators.
1704 Pod::Text (pod2text) and Pod::Man (pod2man) have already been converted
1705 to use Pod::Parser and efforts to convert Pod::HTML (pod2html) are already
1706 underway. For any questions or comments about pod parsing and translating
1707 issues and utilities, please use the pod-people@perl.org mailing list.
1709 For further information, please see L<Pod::Parser> and L<Pod::InputObjects>.
1711 =item Pod::Checker, podchecker
1713 This utility checks pod files for correct syntax, according to
1714 L<perlpod>. Obvious errors are flagged as such, while warnings are
1715 printed for mistakes that can be handled gracefully. The checklist is
1716 not complete yet. See L<Pod::Checker>.
1718 =item Pod::ParseUtils, Pod::Find
1720 These modules provide a set of gizmos that are useful mainly for pod
1721 translators. L<Pod::Find|Pod::Find> traverses directory structures and
1722 returns found pod files, along with their canonical names (like
1723 C<File::Spec::Unix>). L<Pod::ParseUtils|Pod::ParseUtils> contains
1724 B<Pod::List> (useful for storing pod list information), B<Pod::Hyperlink>
1725 (for parsing the contents of C<LE<gt>E<lt>> sequences) and B<Pod::Cache>
1726 (for caching information about pod files, e.g. link nodes).
1728 =item Pod::Select, podselect
1730 Pod::Select is a subclass of Pod::Parser which provides a function
1731 named "podselect()" to filter out user-specified sections of raw pod
1732 documentation from an input stream. podselect is a script that provides
1733 access to Pod::Select from other scripts to be used as a filter.
1736 =item Pod::Usage, pod2usage
1738 Pod::Usage provides the function "pod2usage()" to print usage messages for
1739 a Perl script based on its embedded pod documentation. The pod2usage()
1740 function is generally useful to all script authors since it lets them
1741 write and maintain a single source (the pods) for documentation, thus
1742 removing the need to create and maintain redundant usage message text
1743 consisting of information already in the pods.
1745 There is also a pod2usage script which can be used from other kinds of
1746 scripts to print usage messages from pods (even for non-Perl scripts
1747 with pods embedded in comments).
1749 For details and examples, please see L<Pod::Usage>.
1751 =item Pod::Text and Pod::Man
1753 [TODO - Russ Allbery <rra@stanford.edu>]
1757 An EXISTS method has been added to this module (and sdbm_exists() has
1758 been added to the underlying sdbm library), so one can now call exists
1759 on an SDBM_File tied hash and get the correct result, rather than a
1762 A bug that may have caused data loss when more than one disk block
1763 happens to be read from the database in a single FETCH() has been
1768 Sys::Syslog now uses XSUBs to access facilities from syslog.h so it
1769 no longer requires syslog.ph to exist.
1773 Sys::Hostname now uses XSUBs to call the C library's gethostname() or
1774 uname() if they exist.
1778 The timelocal() and timegm() functions used to silently return bogus
1779 results when the date fell outside the machine's integer range. They
1780 now consistently croak() if the date falls in an unsupported range.
1784 The error return value in list context has been changed for all functions
1785 that return a list of values. Previously these functions returned a list
1786 with a single element C<undef> if an error occurred. Now these functions
1787 return the empty list in these situations. This applies to the following
1793 The remaining functions are unchanged and continue to return C<undef> on
1794 error even in list context.
1796 The Win32::SetLastError(ERROR) function has been added as a complement
1797 to the Win32::GetLastError() function.
1799 The new Win32::GetFullPathName(FILENAME) returns the full absolute
1800 pathname for FILENAME in scalar context. In list context it returns
1801 a two-element list containing the fully qualified directory name and
1802 the filename. See L<Win32>.
1806 A new feature called "DBM Filters" has been added to all the
1807 DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
1808 DBM Filters add four new methods to each DBM module:
1815 These can be used to filter key-value pairs before the pairs are
1816 written to the database or just after they are read from the database.
1817 See L<perldbmfilter> for further information.
1823 C<use attrs> is now obsolete, and is only provided for
1824 backward-compatibility. It's been replaced by the C<sub : attributes>
1825 syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>.
1827 C<use utf8> to enable UTF-8 and Unicode support.
1829 Lexical warnings pragma, C<use warnings;>, to control optional warnings.
1832 C<use filetest> to control the behaviour of filetests (C<-r> C<-w>
1833 ...). Currently only one subpragma implemented, "use filetest
1834 'access';", that uses access(2) or equivalent to check permissions
1835 instead of using stat(2) as usual. This matters in filesystems
1836 where there are ACLs (access control lists): the stat(2) might lie,
1837 but access(2) knows better.
1839 =head1 Utility Changes
1843 [TODO - Kurt Starsinic <kstar@chapin.edu>]
1847 C<perlcc> now supports the C and Bytecode backends. By default,
1848 it generates output from the simple C backend rather than the
1849 optimized C backend.
1851 Support for non-Unix platforms has been improved.
1856 [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>]
1858 =head1 Documentation Changes
1864 The official list of public Perl API functions.
1866 =item perlcompile.pod
1868 An introduction to using the Perl Compiler suite.
1870 =item perlfilter.pod
1872 An introduction to writing Perl source filters.
1876 Some guidelines for hacking the Perl source code.
1878 =item perlintern.pod
1880 A list of internal functions in the Perl source code.
1881 (List is currently empty.)
1883 =item perlopentut.pod
1885 A tutorial on using open() effectively.
1887 =item perlreftut.pod
1889 A tutorial that introduces the essentials of references.
1893 A tutorial for beginners on object-oriented Perl.
1897 A tutorial on managing class data for object modules.
1899 =item perlunicode.pod
1901 An introduction to Unicode support features in Perl.
1905 =head1 New or Changed Diagnostics
1909 =item "%s" variable %s masks earlier declaration in same %s
1911 (W) A "my" or "our" variable has been redeclared in the current scope or statement,
1912 effectively eliminating all access to the previous instance. This is almost
1913 always a typographical error. Note that the earlier variable will still exist
1914 until the end of the scope or until all closure referents to it are
1917 =item "my sub" not yet implemented
1919 (F) Lexically scoped subroutines are not yet implemented. Don't try that
1922 =item "our" variable %s redeclared
1924 (W) You seem to have already declared the same global once before in the
1925 current lexical scope.
1927 =item '!' allowed only after types %s
1929 (F) The '!' is allowed in pack() and unpack() only after certain types.
1930 See L<perlfunc/pack>.
1932 =item / cannot take a count
1934 (F) You had an unpack template indicating a counted-length string,
1935 but you have also specified an explicit size for the string.
1936 See L<perlfunc/pack>.
1938 =item / must be followed by a, A or Z
1940 (F) You had an unpack template indicating a counted-length string,
1941 which must be followed by one of the letters a, A or Z
1942 to indicate what sort of string is to be unpacked.
1943 See L<perlfunc/pack>.
1945 =item / must be followed by a*, A* or Z*
1947 (F) You had a pack template indicating a counted-length string,
1948 Currently the only things that can have their length counted are a*, A* or Z*.
1949 See L<perlfunc/pack>.
1951 =item / must follow a numeric type
1953 (F) You had an unpack template that contained a '#',
1954 but this did not follow some numeric unpack specification.
1955 See L<perlfunc/pack>.
1957 =item /%s/: Unrecognized escape \\%c passed through
1959 (W) You used a backslash-character combination which is not recognized
1960 by Perl. This combination appears in an interpolated variable or a
1961 C<'>-delimited regular expression. The character was understood literally.
1963 =item /%s/: Unrecognized escape \\%c in character class passed through
1965 (W) You used a backslash-character combination which is not recognized
1966 by Perl inside character classes. The character was understood literally.
1968 =item /%s/ should probably be written as "%s"
1970 (W) You have used a pattern where Perl expected to find a string,
1971 as in the first argument to C<join>. Perl will treat the true
1972 or false result of matching the pattern against $_ as the string,
1973 which is probably not what you had in mind.
1975 =item %s() called too early to check prototype
1977 (W) You've called a function that has a prototype before the parser saw a
1978 definition or declaration for it, and Perl could not check that the call
1979 conforms to the prototype. You need to either add an early prototype
1980 declaration for the subroutine in question, or move the subroutine
1981 definition ahead of the call to get proper prototype checking. Alternatively,
1982 if you are certain that you're calling the function correctly, you may put
1983 an ampersand before the name to avoid the warning. See L<perlsub>.
1985 =item %s argument is not a HASH or ARRAY element
1987 (F) The argument to exists() must be a hash or array element, such as:
1990 $ref->[12]->["susie"]
1992 =item %s argument is not a HASH or ARRAY element or slice
1994 (F) The argument to delete() must be either a hash or array element, such as:
1997 $ref->[12]->["susie"]
1999 or a hash or array slice, such as:
2001 @foo[$bar, $baz, $xyzzy]
2002 @{$ref->[12]}{"susie", "queue"}
2004 =item %s argument is not a subroutine name
2006 (F) The argument to exists() for C<exists &sub> must be a subroutine
2007 name, and not a subroutine call. C<exists &sub()> will generate this error.
2009 =item %s package attribute may clash with future reserved word: %s
2011 (W) A lowercase attribute name was used that had a package-specific handler.
2012 That name might have a meaning to Perl itself some day, even though it
2013 doesn't yet. Perhaps you should use a mixed-case attribute name, instead.
2016 =item (in cleanup) %s
2018 (W) This prefix usually indicates that a DESTROY() method raised
2019 the indicated exception. Since destructors are usually called by
2020 the system at arbitrary points during execution, and often a vast
2021 number of times, the warning is issued only once for any number
2022 of failures that would otherwise result in the same message being
2025 Failure of user callbacks dispatched using the C<G_KEEPERR> flag
2026 could also result in this warning. See L<perlcall/G_KEEPERR>.
2028 =item <> should be quotes
2030 (F) You wrote C<require E<lt>fileE<gt>> when you should have written
2033 =item Attempt to join self
2035 (F) You tried to join a thread from within itself, which is an
2036 impossible task. You may be joining the wrong thread, or you may
2037 need to move the join() to some other thread.
2039 =item Bad evalled substitution pattern
2041 (F) You've used the /e switch to evaluate the replacement for a
2042 substitution, but perl found a syntax error in the code to evaluate,
2043 most likely an unexpected right brace '}'.
2045 =item Bad realloc() ignored
2047 (S) An internal routine called realloc() on something that had never been
2048 malloc()ed in the first place. Mandatory, but can be disabled by
2049 setting environment variable C<PERL_BADFREE> to 1.
2051 =item Bareword found in conditional
2053 (W) The compiler found a bareword where it expected a conditional,
2054 which often indicates that an || or && was parsed as part of the
2055 last argument of the previous construct, for example:
2059 It may also indicate a misspelled constant that has been interpreted
2062 use constant TYPO => 1;
2063 if (TYOP) { print "foo" }
2065 The C<strict> pragma is useful in avoiding such errors.
2067 =item Binary number > 0b11111111111111111111111111111111 non-portable
2069 (W) The binary number you specified is larger than 2**32-1
2070 (4294967295) and therefore non-portable between systems. See
2071 L<perlport> for more on portability concerns.
2073 =item Bit vector size > 32 non-portable
2075 (W) Using bit vector sizes larger than 32 is non-portable.
2077 =item Buffer overflow in prime_env_iter: %s
2079 (W) A warning peculiar to VMS. While Perl was preparing to iterate over
2080 %ENV, it encountered a logical name or symbol definition which was too long,
2081 so it was truncated to the string shown.
2083 =item Can't check filesystem of script "%s"
2085 (P) For some reason you can't check the filesystem of the script for nosuid.
2087 =item Can't declare class for non-scalar %s in "%s"
2089 (S) Currently, only scalar variables can declared with a specific class
2090 qualifier in a "my" or "our" declaration. The semantics may be extended
2091 for other types of variables in future.
2093 =item Can't declare %s in "%s"
2095 (F) Only scalar, array, and hash variables may be declared as "my" or
2096 "our" variables. They must have ordinary identifiers as names.
2098 =item Can't ignore signal CHLD, forcing to default
2100 (W) Perl has detected that it is being run with the SIGCHLD signal
2101 (sometimes known as SIGCLD) disabled. Since disabling this signal
2102 will interfere with proper determination of exit status of child
2103 processes, Perl has reset the signal to its default value.
2104 This situation typically indicates that the parent program under
2105 which Perl may be running (e.g. cron) is being very careless.
2107 =item Can't modify non-lvalue subroutine call
2109 (F) Subroutines meant to be used in lvalue context should be declared as
2110 such, see L<perlsub/"Lvalue subroutines">.
2112 =item Can't read CRTL environ
2114 (S) A warning peculiar to VMS. Perl tried to read an element of %ENV
2115 from the CRTL's internal environment array and discovered the array was
2116 missing. You need to figure out where your CRTL misplaced its environ
2117 or define F<PERL_ENV_TABLES> (see L<perlvms>) so that environ is not searched.
2119 =item Can't remove %s: %s, skipping file
2121 (S) You requested an inplace edit without creating a backup file. Perl
2122 was unable to remove the original file to replace it with the modified
2123 file. The file was left unmodified.
2125 =item Can't return %s from lvalue subroutine
2127 (F) Perl detected an attempt to return illegal lvalues (such
2128 as temporary or readonly values) from a subroutine used as an lvalue.
2129 This is not allowed.
2131 =item Can't weaken a nonreference
2133 (F) You attempted to weaken something that was not a reference. Only
2134 references can be weakened.
2136 =item Character class [:%s:] unknown
2138 (F) The class in the character class [: :] syntax is unknown.
2141 =item Character class syntax [%s] belongs inside character classes
2143 (W) The character class constructs [: :], [= =], and [. .] go
2144 I<inside> character classes, the [] are part of the construct,
2145 for example: /[012[:alpha:]345]/. Note that [= =] and [. .]
2146 are not currently implemented; they are simply placeholders for
2149 =item Constant is not %s reference
2151 (F) A constant value (perhaps declared using the C<use constant> pragma)
2152 is being dereferenced, but it amounts to the wrong type of reference. The
2153 message indicates the type of reference that was expected. This usually
2154 indicates a syntax error in dereferencing the constant value.
2155 See L<perlsub/"Constant Functions"> and L<constant>.
2157 =item constant(%s): %%^H is not localized
2159 (F) When setting compile-time-lexicalized hash %^H one should set the
2160 corresponding bit of $^H as well.
2162 =item constant(%s): %s
2164 (F) Compile-time-substitutions (such as overloaded constants and
2165 character names) were not correctly set up.
2167 =item defined(@array) is deprecated
2169 (D) defined() is not usually useful on arrays because it checks for an
2170 undefined I<scalar> value. If you want to see if the array is empty,
2171 just use C<if (@array) { # not empty }> for example.
2173 =item defined(%hash) is deprecated
2175 (D) defined() is not usually useful on hashes because it checks for an
2176 undefined I<scalar> value. If you want to see if the hash is empty,
2177 just use C<if (%hash) { # not empty }> for example.
2179 =item Did not produce a valid header
2183 =item Did you mean "local" instead of "our"?
2185 (W) Remember that "our" does not localize the declared global variable.
2186 You have declared it again in the same lexical scope, which seems superfluous.
2188 =item Document contains no data
2192 =item entering effective %s failed
2194 (F) While under the C<use filetest> pragma, switching the real and
2195 effective uids or gids failed.
2197 =item false [] range "%s" in regexp
2199 (W) A character class range must start and end at a literal character, not
2200 another character class like C<\d> or C<[:alpha:]>. The "-" in your false
2201 range is interpreted as a literal "-". Consider quoting the "-", "\-".
2204 =item Filehandle %s opened only for output
2206 (W) You tried to read from a filehandle opened only for writing. If you
2207 intended it to be a read/write filehandle, you needed to open it with
2208 "+E<lt>" or "+E<gt>" or "+E<gt>E<gt>" instead of with "E<lt>" or nothing. If
2209 you intended only to read from the file, use "E<lt>". See
2212 =item flock() on closed filehandle %s
2214 (W) The filehandle you're attempting to flock() got itself closed some
2215 time before now. Check your logic flow. flock() operates on filehandles.
2216 Are you attempting to call flock() on a dirhandle by the same name?
2218 =item Global symbol "%s" requires explicit package name
2220 (F) You've said "use strict vars", which indicates that all variables
2221 must either be lexically scoped (using "my"), declared beforehand using
2222 "our", or explicitly qualified to say which package the global variable
2225 =item Hexadecimal number > 0xffffffff non-portable
2227 (W) The hexadecimal number you specified is larger than 2**32-1
2228 (4294967295) and therefore non-portable between systems. See
2229 L<perlport> for more on portability concerns.
2231 =item Ill-formed CRTL environ value "%s"
2233 (W) A warning peculiar to VMS. Perl tried to read the CRTL's internal
2234 environ array, and encountered an element without the C<=> delimiter
2235 used to spearate keys from values. The element is ignored.
2237 =item Ill-formed message in prime_env_iter: |%s|
2239 (W) A warning peculiar to VMS. Perl tried to read a logical name
2240 or CLI symbol definition when preparing to iterate over %ENV, and
2241 didn't see the expected delimiter between key and value, so the
2244 =item Illegal binary digit %s
2246 (F) You used a digit other than 0 or 1 in a binary number.
2248 =item Illegal binary digit %s ignored
2250 (W) You may have tried to use a digit other than 0 or 1 in a binary number.
2251 Interpretation of the binary number stopped before the offending digit.
2253 =item Illegal number of bits in vec
2255 (F) The number of bits in vec() (the third argument) must be a power of
2256 two from 1 to 32 (or 64, if your platform supports that).
2258 =item Integer overflow in %s number
2260 (W) The hexadecimal, octal or binary number you have specified either
2261 as a literal or as an argument to hex() or oct() is too big for your
2262 architecture, and has been converted to a floating point number. On a
2263 32-bit architecture the largest hexadecimal, octal or binary number
2264 representable without overflow is 0xFFFFFFFF, 037777777777, or
2265 0b11111111111111111111111111111111 respectively. Note that Perl
2266 transparently promotes all numbers to a floating point representation
2267 internally--subject to loss of precision errors in subsequent
2270 =item Invalid %s attribute: %s
2272 The indicated attribute for a subroutine or variable was not recognized
2273 by Perl or by a user-supplied handler. See L<attributes>.
2275 =item Invalid %s attributes: %s
2277 The indicated attributes for a subroutine or variable were not recognized
2278 by Perl or by a user-supplied handler. See L<attributes>.
2280 =item invalid [] range "%s" in regexp
2282 The offending range is now explicitly displayed.
2284 =item Invalid separator character %s in attribute list
2286 (F) Something other than a colon or whitespace was seen between the
2287 elements of an attribute list. If the previous attribute
2288 had a parenthesised parameter list, perhaps that list was terminated
2289 too soon. See L<attributes>.
2291 =item Invalid separator character %s in subroutine attribute list
2293 (F) Something other than a colon or whitespace was seen between the
2294 elements of a subroutine attribute list. If the previous attribute
2295 had a parenthesised parameter list, perhaps that list was terminated
2298 =item leaving effective %s failed
2300 (F) While under the C<use filetest> pragma, switching the real and
2301 effective uids or gids failed.
2303 =item Lvalue subs returning %s not implemented yet
2305 (F) Due to limitations in the current implementation, array and hash
2306 values cannot be returned in subroutines used in lvalue context.
2307 See L<perlsub/"Lvalue subroutines">.
2309 =item Method %s not permitted
2313 =item Missing %sbrace%s on \N{}
2315 (F) Wrong syntax of character name literal C<\N{charname}> within
2316 double-quotish context.
2318 =item Missing command in piped open
2320 (W) You used the C<open(FH, "| command")> or C<open(FH, "command |")>
2321 construction, but the command was missing or blank.
2323 =item Missing name in "my sub"
2325 (F) The reserved syntax for lexically scoped subroutines requires that they
2326 have a name with which they can be found.
2328 =item No %s specified for -%c
2330 (F) The indicated command line switch needs a mandatory argument, but
2331 you haven't specified one.
2333 =item No package name allowed for variable %s in "our"
2335 (F) Fully qualified variable names are not allowed in "our" declarations,
2336 because that doesn't make much sense under existing semantics. Such
2337 syntax is reserved for future extensions.
2339 =item No space allowed after -%c
2341 (F) The argument to the indicated command line switch must follow immediately
2342 after the switch, without intervening spaces.
2344 =item no UTC offset information; assuming local time is UTC
2346 (S) A warning peculiar to VMS. Perl was unable to find the local
2347 timezone offset, so it's assuming that local system time is equivalent
2348 to UTC. If it's not, define the logical name F<SYS$TIMEZONE_DIFFERENTIAL>
2349 to translate to the number of seconds which need to be added to UTC to
2352 =item Octal number > 037777777777 non-portable
2354 (W) The octal number you specified is larger than 2**32-1 (4294967295)
2355 and therefore non-portable between systems. See L<perlport> for more
2356 on portability concerns.
2358 See also L<perlport> for writing portable code.
2360 =item panic: del_backref
2362 (P) Failed an internal consistency check while trying to reset a weak
2365 =item panic: kid popen errno read
2367 (F) forked child returned an incomprehensible message about its errno.
2369 =item panic: magic_killbackrefs
2371 (P) Failed an internal consistency check while trying to reset all weak
2372 references to an object.
2374 =item Parentheses missing around "%s" list
2376 (W) You said something like
2382 my ($foo, $bar) = @_;
2384 Remember that "my", "our", and "local" bind tighter than comma.
2386 =item Possible Y2K bug: %s
2388 (W) You are concatenating the number 19 with another number, which
2389 could be a potential Year 2000 problem.
2391 =item Premature end of script headers
2395 =item Repeat count in pack overflows
2397 (F) You can't specify a repeat count so large that it overflows
2398 your signed integers. See L<perlfunc/pack>.
2400 =item Repeat count in unpack overflows
2402 (F) You can't specify a repeat count so large that it overflows
2403 your signed integers. See L<perlfunc/unpack>.
2405 =item realloc() of freed memory ignored
2407 (S) An internal routine called realloc() on something that had already
2410 =item Reference is already weak
2412 (W) You have attempted to weaken a reference that is already weak.
2413 Doing so has no effect.
2415 =item setpgrp can't take arguments
2417 (F) Your system has the setpgrp() from BSD 4.2, which takes no arguments,
2418 unlike POSIX setpgid(), which takes a process ID and process group ID.
2420 =item Strange *+?{} on zero-length expression
2422 (W) You applied a regular expression quantifier in a place where it
2423 makes no sense, such as on a zero-width assertion.
2424 Try putting the quantifier inside the assertion instead. For example,
2425 the way to match "abc" provided that it is followed by three
2426 repetitions of "xyz" is C</abc(?=(?:xyz){3})/>, not C</abc(?=xyz){3}/>.
2428 =item switching effective %s is not implemented
2430 (F) While under the C<use filetest> pragma, we cannot switch the
2431 real and effective uids or gids.
2433 =item This Perl can't reset CRTL environ elements (%s)
2435 =item This Perl can't set CRTL environ elements (%s=%s)
2437 (W) Warnings peculiar to VMS. You tried to change or delete an element
2438 of the CRTL's internal environ array, but your copy of Perl wasn't
2439 built with a CRTL that contained the setenv() function. You'll need to
2440 rebuild Perl with a CRTL that does, or redefine F<PERL_ENV_TABLES> (see
2441 L<perlvms>) so that the environ array isn't the target of the change to
2442 %ENV which produced the warning.
2444 =item Unknown open() mode '%s'
2446 (F) The second argument of 3-argument open() is not among the list
2447 of valid modes: C<E<lt>>, C<E<gt>>, C<E<gt>E<gt>>, C<+E<lt>>,
2448 C<+E<gt>>, C<+E<gt>E<gt>>, C<-|>, C<|E<45>>.
2450 =item Unknown process %x sent message to prime_env_iter: %s
2452 (P) An error peculiar to VMS. Perl was reading values for %ENV before
2453 iterating over it, and someone else stuck a message in the stream of
2454 data Perl expected. Someone's very confused, or perhaps trying to
2455 subvert Perl's population of %ENV for nefarious purposes.
2457 =item Unrecognized escape \\%c passed through
2459 (W) You used a backslash-character combination which is not recognized
2460 by Perl. The character was understood literally.
2462 =item Unterminated attribute parameter in attribute list
2464 (F) The lexer saw an opening (left) parenthesis character while parsing an
2465 attribute list, but the matching closing (right) parenthesis
2466 character was not found. You may need to add (or remove) a backslash
2467 character to get your parentheses to balance. See L<attributes>.
2469 =item Unterminated attribute list
2471 (F) The lexer found something other than a simple identifier at the start
2472 of an attribute, and it wasn't a semicolon or the start of a
2473 block. Perhaps you terminated the parameter list of the previous attribute
2474 too soon. See L<attributes>.
2476 =item Unterminated attribute parameter in subroutine attribute list
2478 (F) The lexer saw an opening (left) parenthesis character while parsing a
2479 subroutine attribute list, but the matching closing (right) parenthesis
2480 character was not found. You may need to add (or remove) a backslash
2481 character to get your parentheses to balance.
2483 =item Unterminated subroutine attribute list
2485 (F) The lexer found something other than a simple identifier at the start
2486 of a subroutine attribute, and it wasn't a semicolon or the start of a
2487 block. Perhaps you terminated the parameter list of the previous attribute
2490 =item Value of CLI symbol "%s" too long
2492 (W) A warning peculiar to VMS. Perl tried to read the value of an %ENV
2493 element from a CLI symbol table, and found a resultant string longer
2494 than 1024 characters. The return value has been truncated to 1024
2497 =item Version number must be a constant number
2499 (P) The attempt to translate a C<use Module n.n LIST> statement into
2500 its equivalent C<BEGIN> block found an internal inconsistency with
2505 =head1 Obsolete Diagnostics
2509 =item Character class syntax [: :] is reserved for future extensions
2511 (W) Within regular expression character classes ([]) the syntax beginning
2512 with "[:" and ending with ":]" is reserved for future extensions.
2513 If you need to represent those character sequences inside a regular
2514 expression character class, just quote the square brackets with the
2515 backslash: "\[:" and ":\]".
2517 =item Ill-formed logical name |%s| in prime_env_iter
2519 (W) A warning peculiar to VMS. A logical name was encountered when preparing
2520 to iterate over %ENV which violates the syntactic rules governing logical
2521 names. Because it cannot be translated normally, it is skipped, and will not
2522 appear in %ENV. This may be a benign occurrence, as some software packages
2523 might directly modify logical name tables and introduce nonstandard names,
2524 or it may indicate that a logical name table has been corrupted.
2526 =item Probable precedence problem on %s
2528 (W) The compiler found a bareword where it expected a conditional,
2529 which often indicates that an || or && was parsed as part of the
2530 last argument of the previous construct, for example:
2534 =item regexp too big
2536 (F) The current implementation of regular expressions uses shorts as
2537 address offsets within a string. Unfortunately this means that if
2538 the regular expression compiles to longer than 32767, it'll blow up.
2539 Usually when you want a regular expression this big, there is a better
2540 way to do it with multiple statements. See L<perlre>.
2542 =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
2544 (D) Perl versions before 5.004 misinterpreted any type marker followed
2545 by "$" and a digit. For example, "$$0" was incorrectly taken to mean
2546 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
2548 However, the developers of Perl 5.004 could not fix this bug completely,
2549 because at least two widely-used modules depend on the old meaning of
2550 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
2551 old (broken) way inside strings; but it generates this message as a
2552 warning. And in Perl 5.005, this special treatment will cease.
2558 If you find what you think is a bug, you might check the
2559 articles recently posted to the comp.lang.perl.misc newsgroup.
2560 There may also be information at http://www.perl.com/perl/, the Perl
2563 If you believe you have an unreported bug, please run the B<perlbug>
2564 program included with your release. Make sure to trim your bug down
2565 to a tiny but sufficient test case. Your bug report, along with the
2566 output of C<perl -V>, will be sent off to perlbug@perl.com to be
2567 analysed by the Perl porting team.
2571 The F<Changes> file for exhaustive details on what changed.
2573 The F<INSTALL> file for how to build Perl.
2575 The F<README> file for general stuff.
2577 The F<Artistic> and F<Copying> files for copyright information.
2581 Written by Gurusamy Sarathy <F<gsar@activestate.com>>, with many
2582 contributions from The Perl Porters.
2584 Send omissions or corrections to <F<perlbug@perl.com>>.