3 perldelta - what's new for perl5.004
7 This document describes differences between the 5.003 release (as
8 documented in I<Programming Perl>, second edition--the Camel Book) and
11 =head1 Supported Environments
13 Perl5.004 builds out of the box on Unix, Plan9, LynxOS, VMS, OS/2,
18 Most importantly, many bugs were fixed. See the F<Changes>
19 file in the distribution for details.
21 =head2 Compilation Option: Binary Compatibility With 5.003
23 There is a new Configure question that asks if you want to maintain
24 binary compatibility with Perl 5.003. If you choose binary
25 compatibility, you do not have to recompile your extensions, but you
26 might have symbol conflicts if you embed Perl in another application,
27 just as in the 5.003 release. By default, binary compatibility
28 is preserved at the expense of symbol table pollution.
30 =head2 New Opcode Module and Revised Safe Module
32 A new Opcode module supports the creation, manipulation and
33 application of opcode masks. The revised Safe module has a new API
34 and is implemented using the new Opcode module. Please read the new
35 Opcode and Safe documentation.
37 =head2 Internal Change: FileHandle Deprecated
39 Filehandles are now stored internally as type IO::Handle.
40 Although C<use FileHandle> and C<*STDOUT{FILEHANDLE}>
41 are still supported for backwards compatibility,
42 C<use IO::Handle> (or C<IO::Seekable> or C<IO::File>) and
43 C<*STDOUT{IO}> are the way of the future.
45 =head2 Internal Change: PerlIO internal IO abstraction interface
47 It is now possible to build Perl with AT&T's sfio IO package
48 instead of stdio. See L<perlapio> for more details, and
49 the F<INSTALL> file for how to use it.
51 =head2 New and Changed Built-in Variables
57 Extended error message on some platforms. (Also known as
58 $EXTENDED_OS_ERROR if you C<use English>).
62 The current set of syntax checks enabled by C<use strict>. See the
63 documentation of C<strict> for more details. Not actually new, but
65 Because it is intended for internal use by Perl core components,
66 there is no C<use English> long name for this variable.
70 By default, running out of memory it is not trappable. However, if
71 compiled for this, Perl may use the contents of C<$^M> as an emergency
72 pool after die()ing with this message. Suppose that your Perl were
73 compiled with -DEMERGENCY_SBRK and used Perl's malloc. Then
77 would allocate a 64K buffer for use when in emergency.
78 See the F<INSTALL> file for information on how to enable this option.
79 As a disincentive to casual use of this advanced feature,
80 there is no C<use English> long name for this variable.
84 =head2 New and Changed Built-in Functions
88 =item delete on slices
90 This now works. (e.g. C<delete @ENV{'PATH', 'MANPATH'}>)
94 is now supported on more platforms, and prefers fcntl
95 to lockf when emulating.
97 =item printf and sprintf
99 now support "%i" as a synonym for "%d", and the "h" modifier.
100 So "%hi" means "short integer in decimal", and "%ho" means
101 "unsigned short integer as octal".
103 =item keys as an lvalue
105 As an lvalue, C<keys> allows you to increase the number of hash buckets
106 allocated for the given associative array. This can gain you a measure
107 of efficiency if you know the hash is going to get big. (This is
108 similar to pre-extending an array by assigning a larger number to
113 then C<%hash> will have at least 200 buckets allocated for it. These
114 buckets will be retained even if you do C<%hash = ()>; use C<undef
115 %hash> if you want to free the storage while C<%hash> is still in scope.
116 You can't shrink the number of buckets allocated for the hash using
117 C<keys> in this way (but you needn't worry about doing this by accident,
118 as trying has no effect).
120 =item my() in Control Structures
122 You can now use my() (with or without the parentheses) in the control
123 expressions of control structures such as:
125 while (my $line = <>) {
131 if ((my $answer = <STDIN>) =~ /^y(es)?$/i) {
133 } elsif ($answer =~ /^n(o)?$/i) {
137 die "`$answer' is neither `yes' nor `no'";
140 Also, you can declare a foreach loop control variable as lexical by
141 preceding it with the word "my". For example, in:
143 foreach my $i (1, 2, 3) {
147 $i is a lexical variable, and the scope of $i extends to the end of
148 the loop, but not beyond it.
150 Note that you still cannot use my() on global punctuation variables
151 such as $_ and the like.
153 =item unpack() and pack()
155 A new format 'w' represents a BER compressed integer (as defined in
156 ASN.1). Its format is a sequence of one or more bytes, each of which
157 provides seven bits of the total value, with the most significant
158 first. Bit eight of each byte is set, except for the last byte, in
159 which bit eight is clear.
163 If the first argument to C<use> is a number, it is treated as a version
164 number instead of a module name. If the version of the Perl interpreter
165 is less than VERSION, then an error message is printed and Perl exits
166 immediately. Because C<use> occurs at compile time, this check happens
167 immediately during the compilation process, unlike C<require VERSION>,
168 which waits until run-time for the check. This is often useful if you
169 need to check the current Perl version before C<use>ing library modules
170 which have changed in incompatible ways from older versions of Perl.
171 (We try not to do this more than we have to.)
173 =item use Module VERSION LIST
175 If the VERSION argument is present between Module and LIST, then the
176 C<use> will call the VERSION method in class Module with the given
177 version as an argument. The default VERSION method, inherited from
178 the Universal class, croaks if the given version is larger than the
179 value of the variable $Module::VERSION. (Note that there is not a
180 comma after VERSION!)
182 This version-checking mechanism is similar to the one currently used
183 in the Exporter module, but it is faster and can be used with modules
184 that don't use the Exporter. It is the recommended method for new
187 =item prototype(FUNCTION)
189 Returns the prototype of a function as a string (or C<undef> if the
190 function has no prototype). FUNCTION is a reference to or the name of the
191 function whose prototype you want to retrieve.
192 (Not actually new; just never documented before.)
196 Functions documented in the Camel to default to $_ now in
197 fact do, and all those that do are so documented in L<perlfunc>.
199 =item C<m//g> does not trigger a pos() reset on failure
201 The C<m//g> match iteration construct used to reset the iteration
202 when it failed to match (so that the next C<m//g> match would start at
203 the beginning of the string). You now have to explicitly do a
204 C<pos $str = 0;> to reset the "last match" position, or modify the
205 string in some way. This change makes it practical to chain C<m//g>
206 matches together in conjunction with ordinary matches using the C<\G>
207 zero-width assertion. See L<perlop> and L<perlre>.
209 =item nested C<sub{}> closures work now
211 Prior to the 5.004 release, nested anonymous functions
212 didn't work right. They do now.
214 =item formats work right on changing lexicals
216 Just like anonymous functions that contain lexical variables
217 that change (like a lexical index variable for a C<foreach> loop),
218 formats now work properly. For example, this silently failed
219 before, and is fine now:
222 foreach $i ( 1 .. 10 ) {
232 =head2 New Built-in Methods
234 The C<UNIVERSAL> package automatically contains the following methods that
235 are inherited by all other classes:
241 C<isa> returns I<true> if its object is blessed into a sub-class of C<CLASS>
243 C<isa> is also exportable and can be called as a sub with two arguments. This
244 allows the ability to check what a reference points to. Example:
246 use UNIVERSAL qw(isa);
248 if(isa($ref, 'ARRAY')) {
254 C<can> checks to see if its object has a method called C<METHOD>,
255 if it does then a reference to the sub is returned; if it does not then
256 I<undef> is returned.
258 =item VERSION( [NEED] )
260 C<VERSION> returns the version number of the class (package). If the
261 NEED argument is given then it will check that the current version (as
262 defined by the $VERSION variable in the given package) not less than
263 NEED; it will die if this is not the case. This method is normally
264 called as a class method. This method is called automatically by the
265 C<VERSION> form of C<use>.
267 use A 1.2 qw(some imported subs);
273 B<NOTE:> C<can> directly uses Perl's internal code for method lookup, and
274 C<isa> uses a very similar method and caching strategy. This may cause
275 strange effects if the Perl code dynamically changes @ISA in any package.
277 You may add other methods to the UNIVERSAL class via Perl or XS code.
278 You do not need to C<use UNIVERSAL> in order to make these methods
279 available to your program. This is necessary only if you wish to
280 have C<isa> available as a plain subroutine in the current package.
282 =head2 TIEHANDLE Now Supported
284 See L<perltie> for other kinds of tie()s.
288 =item TIEHANDLE classname, LIST
290 This is the constructor for the class. That means it is expected to
291 return an object of some sort. The reference can be used to
292 hold some internal information.
297 return bless \$i, shift;
300 =item PRINT this, LIST
302 This method will be triggered every time the tied handle is printed to.
303 Beyond its self reference it also expects the list that was passed to
309 return print join( $, => map {uc} @_), $\;
314 This method will be called when the handle is read from. The method
315 should return undef when there is no more data.
319 return "PRINT called $$r times\n";
324 As with the other types of ties, this method will be called when the
325 tied handle is about to be destroyed. This is useful for debugging and
326 possibly for cleaning up.
334 =head2 Efficiency Enhancements
336 All hash keys with the same string are only allocated once, so
337 even if you have 100 copies of the same hash, the immutable keys
338 never have to be re-allocated.
340 Functions that have an empty prototype and that do nothing but return
341 a fixed value are now inlined (e.g. C<sub PI () { 3.14159 }>).
343 =head2 Malloc Enhancements
345 If perl's malloc() is used, you can require memory statistics at
346 runtime by running perl thusly:
348 env PERL_DEBUG_MSTATS=2 perl your_script_here
350 The value of 2 means that the statistics is required after the
351 compilation and on exit, if value is 1 the statistics is printed on
352 exit only. (If you want the statistics at an arbitrary time, you need
353 to install an optional module Devel::Peek.)
355 Three new flags are recognized by malloc.c. (They have no effect if
356 perl is compiled with system malloc().)
360 =item B<-D>C<EMERGENCY_SBRK>
362 If defined, running out of memory may be not a fatal error if a memory
363 pool is allocated by assigning to variable $^M, see L<"$^M">.
365 =item B<-D>C<PACK_MALLOC>
367 Perl memory allocation is by bucket with sizes close to powers of
368 two. Because of these malloc overhead may be big, especially for data
369 of size exactly a power of two. If C<PACK_MALLOC> is defined, perl
370 uses a slightly different algorithm for small allocations (up to 64
371 bytes long), which makes it possible to have overhead down to 1 byte
372 for allocations which are powers of two (and appear quite often).
374 Expected memory savings (with 8-byte alignment in C<alignbytes>) are
375 circa 20% for typical Perl usage. Expected slowdown due to additional
376 malloc overhead is in fractions of percent (hard to measure, because
377 of the effect of saved memory on speed).
379 =item B<-D>C<TWO_POT_OPTIMIZE>
381 Similarly to C<PACK_MALLOC>, this improves allocations of data with
382 size close ot power of two, only it works for big allocations
383 (starting with 16K by default). Such allocations are typical for big
384 hashes and special-purpose scripts, say, image processing.
386 On young systems the fact that perl requires 2M from system for 1M
387 allocation will not affect the speed, since the tail of such a chunk
388 is not going to be touched, however, this may affect the maximal
389 number of allocations (unless the system overcommits memory).
391 Expected saving of memory is 0-100% (100% in applications which
392 require most memory in such 2**n chunks), expected slowdown
399 Four new pragmatic modules exist:
407 Looks for MakeMaker-like I<'blib'> directory structure starting in
408 I<dir> (or current directory) and working back up to five levels of
411 Intended for use on command line with B<-M> option as a way of testing
412 arbitrary scripts against an uninstalled version of a package.
416 Tells the compiler to enable (or disable) the use of POSIX locales for
419 When C<use locale> is in effect, the current LC_CTYPE locale is used
420 for regular expressions and case mapping; LC_COLLATE for string
421 ordering; and LC_NUMERIC for numeric formating in printf and sprintf
422 (but B<not> in print). LC_NUMERIC is always used in write, since
423 lexical scoping of formats is problematic at best.
425 Each C<use locale> or C<no locale> affects statements to the end of
426 the enclosing BLOCK or, if not inside a BLOCK, to the end of the
427 current file. Locales can be switched and queried with
430 See L<perllocale> for more information.
434 Disable unsafe opcodes, or any named opcodes, when compiling Perl code.
438 Enable VMS-specific language features. Currently, there are three
439 VMS-specific feature available: 'status', which makes C<$?> and
440 C<system> return genuine VMS status values instead of emulating POSIX;
441 'exit', which makes C<exit> take a genuine VMS status value instead of
442 assuming that C<exit 1> is an error; and 'time', which makes all times
443 relative to the local time zone, in the VMS tradition.
449 =head2 Installation Directories
451 The I<installperl> script now places the Perl source files for
452 extensions in the architecture-specific library directory, which is
453 where the shared libraries for extensions have always been. This
454 change is intended to allow administrators to keep the Perl 5.004
455 library directory unchanged from a previous version, without running
456 the risk of binary incompatibility between extensions' Perl source and
461 New constants in the existing Fcntl modules are now supported,
462 provided that your operating system happens to support them:
465 O_ASYNC O_DEFER O_DSYNC O_FSYNC O_SYNC
468 These constants are intended for use with the Perl operators sysopen()
469 and fcntl() and the basic database modules like SDBM_File. For the
470 exact meaning of these and other Fcntl constants please refer to your
471 operating system's documentation for fcntl() and open().
473 In addition, the Fcntl module now provides these constants for use
474 with the Perl operator flock():
476 LOCK_SH LOCK_EX LOCK_NB LOCK_UN
478 These constants are defined in all environments (because where there is
479 no flock() system call, Perl emulates it). However, for historical
480 reasons, these constants are not exported unless they are explicitly
481 requested with the ":flock" tag (e.g. C<use Fcntl ':flock'>).
483 =head2 Module Information Summary
485 Brand new modules, arranged by topic rather than strictly
488 CPAN interface to Comprehensive Perl Archive Network
489 CPAN::FirstTime create a CPAN configuration file
490 CPAN::Nox run CPAN while avoiding compiled extensions
492 IO.pm Top-level interface to IO::* classes
493 IO/File.pm IO::File extension Perl module
494 IO/Handle.pm IO::Handle extension Perl module
495 IO/Pipe.pm IO::Pipe extension Perl module
496 IO/Seekable.pm IO::Seekable extension Perl module
497 IO/Select.pm IO::Select extension Perl module
498 IO/Socket.pm IO::Socket extension Perl module
500 Opcode.pm Disable named opcodes when compiling Perl code
502 ExtUtils/Embed.pm Utilities for embedding Perl in C programs
503 ExtUtils/testlib.pm Fixes up @INC to use just-built extension
505 FindBin.pm Find path of currently executing program
507 Class/Template.pm Structure/member template builder
508 File/stat.pm Object-oriented wrapper around CORE::stat
509 Net/hostent.pm Object-oriented wrapper around CORE::gethost*
510 Net/netent.pm Object-oriented wrapper around CORE::getnet*
511 Net/protoent.pm Object-oriented wrapper around CORE::getproto*
512 Net/servent.pm Object-oriented wrapper around CORE::getserv*
513 Time/gmtime.pm Object-oriented wrapper around CORE::gmtime
514 Time/localtime.pm Object-oriented wrapper around CORE::localtime
515 Time/tm.pm Perl implementation of "struct tm" for {gm,local}time
516 User/grent.pm Object-oriented wrapper around CORE::getgr*
517 User/pwent.pm Object-oriented wrapper around CORE::getpw*
519 Tie/RefHash.pm Base class for tied hashes with references as keys
521 UNIVERSAL.pm Base class for *ALL* classes
525 The IO module provides a simple mechanism to load all of the IO modules at one
526 go. Currently this includes:
534 For more information on any of these modules, please see its
535 respective documentation.
539 The Math::Complex module has been totally rewritten, and now supports
540 more operations. These are overloaded:
542 + - * / ** <=> neg ~ abs sqrt exp log sin cos atan2 "" (stringify)
544 And these functions are now exported:
548 tan cotan asin acos atan acotan
549 sinh cosh tanh cotanh asinh acosh atanh acotanh
554 There have been quite a few changes made to DB_File. Here are a few of
561 Fixed a handful of bugs.
565 By public demand, added support for the standard hash function exists().
569 Made it compatible with Berkeley DB 1.86.
573 Made negative subscripts work with RECNO interface.
577 Changed the default flags from O_RDWR to O_CREAT|O_RDWR and the default
578 mode from 0640 to 0666.
582 Made DB_File automatically import the open() constants (O_RDWR,
583 O_CREAT etc.) from Fcntl, if available.
587 Updated documentation.
591 Refer to the HISTORY section in DB_File.pm for a complete list of
592 changes. Everything after DB_File 1.01 has been added since 5.003.
596 Major rewrite - support added for both udp echo and real icmp pings.
598 =head2 Overridden Built-ins
600 Many of the Perl built-ins returning lists now have
601 object-oriented overrides. These are:
613 For example, you can now say
617 $his = (stat($filename)->st_uid == pwent($whoever)->pw_uid);
619 =head1 Utility Changes
625 =item C<void> XSUBs now default to returning nothing
627 Due to a documentation/implementation bug in previous versions of
628 Perl, XSUBs with a return type of C<void> have actually been
629 returning one value. Usually that value was the GV for the XSUB,
630 but sometimes it was some already freed or reused value, which would
631 sometimes lead to program failure.
633 In Perl 5.004, if an XSUB is declared as returning C<void>, it
634 actually returns no value, i.e. an empty list (though there is a
635 backward-compatibility exception; see below). If your XSUB really
636 does return an SV, you should give it a return type of C<SV *>.
638 For backward compatibility, I<xsubpp> tries to guess whether a
639 C<void> XSUB is really C<void> or if it wants to return an C<SV *>.
640 It does so by examining the text of the XSUB: if I<xsubpp> finds
641 what looks like an assignment to C<ST(0)>, it assumes that the
642 XSUB's return type is really C<SV *>.
646 =head1 C Language API Changes
650 =item C<gv_fetchmethod> and C<perl_call_sv>
652 The C<gv_fetchmethod> function finds a method for an object, just like
653 in Perl 5.003. The GV it returns may be a method cache entry.
654 However, in Perl 5.004, method cache entries are not visible to users;
655 therefore, they can no longer be passed directly to C<perl_call_sv>.
656 Instead, you should use the C<GvCV> macro on the GV to extract its CV,
657 and pass the CV to C<perl_call_sv>.
659 The most likely symptom of passing the result of C<gv_fetchmethod> to
660 C<perl_call_sv> is Perl's producing an "Undefined subroutine called"
661 error on the I<second> call to a given method (since there is no cache
666 =head1 Documentation Changes
668 Many of the base and library pods were updated. These
669 new pods are included in section 1:
679 Locale support (internationalization and localization).
683 Tutorial on Perl OO programming.
687 Perl internal IO abstraction interface.
691 Although not new, this has been massively updated.
695 Although not new, this has been massively updated.
699 =head1 New Diagnostics
701 Several new conditions will trigger warnings that were
702 silent before. Some only affect certain platforms.
703 The following new warnings and errors outline these.
704 These messages are classified as follows (listed in
705 increasing order of desperation):
707 (W) A warning (optional).
708 (D) A deprecation (optional).
709 (S) A severe warning (mandatory).
710 (F) A fatal error (trappable).
711 (P) An internal error you should never see (trappable).
712 (X) A very fatal error (non-trappable).
713 (A) An alien error message (not generated by Perl).
717 =item "my" variable %s masks earlier declaration in same scope
719 (S) A lexical variable has been redeclared in the same scope, effectively
720 eliminating all access to the previous instance. This is almost always
721 a typographical error. Note that the earlier variable will still exist
722 until the end of the scope or until all closure referents to it are
725 =item %s argument is not a HASH element or slice
727 (F) The argument to delete() must be either a hash element, such as
730 $ref->[12]->{"susie"}
732 or a hash slice, such as
734 @foo{$bar, $baz, $xyzzy}
735 @{$ref->[12]}{"susie", "queue"}
737 =item Allocation too large: %lx
739 (X) You can't allocate more than 64K on an MSDOS machine.
741 =item Allocation too large
743 (F) You can't allocate more than 2^31+"small amount" bytes.
745 =item Attempt to free non-existent shared string
747 (P) Perl maintains a reference counted internal table of strings to
748 optimize the storage and access of hash keys and other strings. This
749 indicates someone tried to decrement the reference count of a string
750 that can no longer be found in the table.
752 =item Attempt to use reference as lvalue in substr
754 (W) You supplied a reference as the first argument to substr() used
755 as an lvalue, which is pretty strange. Perhaps you forgot to
756 dereference it first. See L<perlfunc/substr>.
758 =item Unsupported function fork
760 (F) Your version of executable does not support forking.
762 Note that under some systems, like OS/2, there may be different flavors of
763 Perl executables, some of which may support fork, some not. Try changing
764 the name you call Perl by to C<perl_>, C<perl__>, and so on.
766 =item Ill-formed logical name |%s| in prime_env_iter
768 (W) A warning peculiar to VMS. A logical name was encountered when preparing
769 to iterate over %ENV which violates the syntactic rules governing logical
770 names. Since it cannot be translated normally, it is skipped, and will not
771 appear in %ENV. This may be a benign occurrence, as some software packages
772 might directly modify logical name tables and introduce non-standard names,
773 or it may indicate that a logical name table has been corrupted.
775 =item Can't use bareword ("%s") as %s ref while "strict refs" in use
777 (F) Only hard references are allowed by "strict refs". Symbolic references
778 are disallowed. See L<perlref>.
780 =item Constant subroutine %s redefined
782 (S) You redefined a subroutine which had previously been eligible for
783 inlining. See L<perlsub/"Constant Functions"> for commentary and
788 (F) You passed die() an empty string (the equivalent of C<die "">) or
789 you called it with no args and both C<$@> and C<$_> were empty.
791 =item Integer overflow in hex number
793 (S) The literal hex number you have specified is too big for your
794 architecture. On a 32-bit architecture the largest hex literal is
797 =item Integer overflow in octal number
799 (S) The literal octal number you have specified is too big for your
800 architecture. On a 32-bit architecture the largest octal literal is
803 =item Name "%s::%s" used only once: possible typo
805 (W) Typographical errors often show up as unique variable names.
806 If you had a good reason for having a unique name, then just mention
807 it again somehow to suppress the message (the C<use vars> pragma is
808 provided for just this purpose).
810 =item Null picture in formline
812 (F) The first argument to formline must be a valid format picture
813 specification. It was found to be empty, which probably means you
814 supplied it an uninitialized value. See L<perlform>.
816 =item Offset outside string
818 (F) You tried to do a read/write/send/recv operation with an offset
819 pointing outside the buffer. This is difficult to imagine.
820 The sole exception to this is that C<sysread()>ing past the buffer
821 will extend the buffer and zero pad the new area.
823 =item Stub found while resolving method `%s' overloading `%s' in package `%s'
825 (P) Overloading resolution over @ISA tree may be broken by importing stubs.
826 Stubs should never be implicitely created, but explicit calls to C<can>
829 =item Cannot resolve method `%s' overloading `%s' in package `s'
831 (P) Internal error trying to resolve overloading specified by a method
832 name (as opposed to a subroutine reference).
836 (X|F) The malloc() function returned 0, indicating there was insufficient
837 remaining memory (or virtual memory) to satisfy the request.
839 The request was judged to be small, so the possibility to trap it
840 depends on the way Perl was compiled. By default it is not trappable.
841 However, if compiled for this, Perl may use the contents of C<$^M> as
842 an emergency pool after die()ing with this message. In this case the
843 error is trappable I<once>.
845 =item Out of memory during request for %s
847 (F) The malloc() function returned 0, indicating there was insufficient
848 remaining memory (or virtual memory) to satisfy the request. However,
849 the request was judged large enough (compile-time default is 64K), so
850 a possibility to shut down by trapping this error is granted.
852 =item Possible attempt to put comments in qw() list
854 (W) qw() lists contain items separated by whitespace; as with literal
855 strings, comment characters are not ignored, but are instead treated
856 as literal data. (You may have used different delimiters than the
857 exclamation marks parentheses shown here; braces are also frequently
860 You probably wrote something like this:
867 when you should have written this:
874 If you really want comments, build your list the
875 old-fashioned way, with quotes and commas:
879 'b', # another comment
882 =item Possible attempt to separate words with commas
884 (W) qw() lists contain items separated by whitespace; therefore commas
885 aren't needed to separate the items. (You may have used different
886 delimiters than the parentheses shown here; braces are also frequently
889 You probably wrote something like this:
893 which puts literal commas into some of the list items. Write it without
894 commas if you don't want them to appear in your data:
898 =item Scalar value @%s{%s} better written as $%s{%s}
900 (W) You've used a hash slice (indicated by @) to select a single element of
901 a hash. Generally it's better to ask for a scalar value (indicated by $).
902 The difference is that C<$foo{&bar}> always behaves like a scalar, both when
903 assigning to it and when evaluating its argument, while C<@foo{&bar}> behaves
904 like a list when you assign to it, and provides a list context to its
905 subscript, which can do weird things if you're expecting only one subscript.
907 =item untie attempted while %d inner references still exist
909 (W) A copy of the object returned from C<tie> (or C<tied>) was still
910 valid when C<untie> was called.
912 =item Value of %s construct can be "0"; test with defined()
914 (W) In a conditional expression, you used <HANDLE>, <*> (glob), or
915 C<readdir> as a boolean value. Each of these constructs can return a
916 value of "0"; that would make the conditional expression false, which
917 is probably not what you intended. When using these constructs in
918 conditional expressions, test their values with the C<defined> operator.
920 =item Variable "%s" may be unavailable
922 (W) An inner (nested) I<anonymous> subroutine is inside a I<named>
923 subroutine, and outside that is another subroutine; and the anonymous
924 (innermost) subroutine is referencing a lexical variable defined in
925 the outermost subroutine. For example:
927 sub outermost { my $a; sub middle { sub { $a } } }
929 If the anonymous subroutine is called or referenced (directly or
930 indirectly) from the outermost subroutine, it will share the variable
931 as you would expect. But if the anonymous subroutine is called or
932 referenced when the outermost subroutine is not active, it will see
933 the value of the shared variable as it was before and during the
934 *first* call to the outermost subroutine, which is probably not what
937 In these circumstances, it is usually best to make the middle
938 subroutine anonymous, using the C<sub {}> syntax. Perl has specific
939 support for shared variables in nested anonymous subroutines; a named
940 subroutine in between interferes with this feature.
942 =item Variable "%s" will not stay shared
944 (W) An inner (nested) I<named> subroutine is referencing a lexical
945 variable defined in an outer subroutine.
947 When the inner subroutine is called, it will probably see the value of
948 the outer subroutine's variable as it was before and during the
949 *first* call to the outer subroutine; in this case, after the first
950 call to the outer subroutine is complete, the inner and outer
951 subroutines will no longer share a common value for the variable. In
952 other words, the variable will no longer be shared.
954 Furthermore, if the outer subroutine is anonymous and references a
955 lexical variable outside itself, then the outer and inner subroutines
956 will I<never> share the given variable.
958 This problem can usually be solved by making the inner subroutine
959 anonymous, using the C<sub {}> syntax. When inner anonymous subs that
960 reference variables in outer subroutines are called or referenced,
961 they are automatically re-bound to the current values of such
964 =item Warning: something's wrong
966 (W) You passed warn() an empty string (the equivalent of C<warn "">) or
967 you called it with no args and C<$_> was empty.
969 =item Got an error from DosAllocMem
971 (P) An error peculiar to OS/2. Most probably you're using an obsolete
972 version of Perl, and this should not happen anyway.
974 =item Malformed PERLLIB_PREFIX
976 (F) An error peculiar to OS/2. PERLLIB_PREFIX should be of the form
984 with non-empty prefix1 and prefix2. If C<prefix1> is indeed a prefix of
985 a builtin library search path, prefix2 is substituted. The error may appear
986 if components are not found, or are too long. See L<perlos2/"PERLLIB_PREFIX">.
988 =item PERL_SH_DIR too long
990 (F) An error peculiar to OS/2. PERL_SH_DIR is the directory to find the
991 C<sh>-shell in. See L<perlos2/"PERL_SH_DIR">.
993 =item Process terminated by SIG%s
995 (W) This is a standard message issued by OS/2 applications, while *nix
996 applications die in silence. It is considered a feature of the OS/2
997 port. One can easily disable this by appropriate sighandlers, see
998 L<perlipc/"Signals">. See L<perlos2/"Process terminated by SIGTERM/SIGINT">.
1004 If you find what you think is a bug, you might check the headers of
1005 recently posted articles in the comp.lang.perl.misc newsgroup.
1006 There may also be information at http://www.perl.com/perl/, the Perl
1009 If you believe you have an unreported bug, please run the B<perlbug>
1010 program included with your release. Make sure you trim your bug
1011 down to a tiny but sufficient test case. Your bug report, along
1012 with the output of C<perl -V>, will be sent off to perlbug@perl.com
1013 to be analysed by the Perl porting team.
1017 The F<Changes> file for exhaustive details on what changed.
1019 The F<INSTALL> file for how to build Perl. This file has been
1020 significantly updated for 5.004, so even veteran users should
1023 The F<README> file for general stuff.
1025 The F<Copying> file for copyright information.
1029 Constructed by Tom Christiansen, grabbing material with permission
1030 from innumerable contributors, with kibitzing by more than a few Perl
1033 Last update: Tue Jan 14 14:03:02 EST 1997