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 Subroutine arguments created only when they're modified
32 In Perl 5.004, nonexistent array and hash elements used as subroutine
33 parameters are brought into existence only if they are actually
34 assigned to (via C<@_>).
36 Earlier versions of Perl vary in their handling of such arguments.
37 Perl versions 5.002 and 5.003 always brought them into existence.
38 Perl versions 5.000, 5.001, and 5.002 brought them into existence only
39 if they were not the first argument (which was almost certainly a
40 bug). Earlier versions of Perl never brought them into existence.
42 For example, given this code:
45 sub show { print $_[0] };
46 sub change { $_[0]++ };
50 After this code executes in Perl 5.004, $a{b} exists but $a[2] does
51 not. In Perl 5.002 and 5.003, both $a{b} and $a[2] would have existed
52 (but $a[2]'s value would have been undefined).
54 =head2 Fixed Parsing of $$<digit>, &$<digit>, etc.
56 A bug in previous versions of Perl 5.0 prevented proper parsing of
57 numeric special variables as symbolic references. That bug has been
58 fixed. As a result, the string "$$0" is no longer equivalent to
59 C<$$."0">, but rather to C<${$0}>. To get the old behavior, change
60 "$$" followed by a digit to "${$}".
62 =head2 No Resetting of $. on Implicit Close
64 The documentation for Perl 5.0 has always stated that C<$.> is I<not>
65 reset when an already-open file handle is re-opened with no intervening
66 call to C<close>. Due to a bug, perl versions 5.000 through 5.0003
67 I<did> reset C<$.> under that circumstance; Perl 5.004 does not.
69 =head2 Changes to Tainting Checks
71 A bug in previous versions may have failed to detect some insecure
72 conditions when taint checks are turned on. (Taint checks are used
73 in setuid or setgid scripts, or when explicitly turned on with the
74 C<-T> invocation option.) Although it's unlikely, this may cause a
75 previously-working script to now fail -- which should be construed
76 as a blessing, since that indicates a potentially-serious security
77 hole was just plugged.
79 =head2 New Opcode Module and Revised Safe Module
81 A new Opcode module supports the creation, manipulation and
82 application of opcode masks. The revised Safe module has a new API
83 and is implemented using the new Opcode module. Please read the new
84 Opcode and Safe documentation.
86 =head2 Embedding Improvements
88 In older versions of Perl it was not possible to create more than one
89 Perl interpreter instance inside a single process without leaking like a
90 sieve and/or crashing. The bugs that caused this behavior have all been
91 fixed. However, you still must take care when embedding Perl in a C
92 program. See the updated perlembed manpage for tips on how to manage
95 =head2 Internal Change: FileHandle Class Based on IO::* Classes
97 File handles are now stored internally as type IO::Handle. The
98 FileHandle module is still supported for backwards compatibility, but
99 it is now merely a front end to the IO::* modules -- specifically,
100 IO::Handle, IO::Seekable, and IO::File. We suggest, but do not
101 require, that you use the IO::* modules in new code.
103 In harmony with this change, C<*GLOB{FILEHANDLE}> is now a
104 backward-compatible synonym for C<*STDOUT{IO}>.
106 =head2 Internal Change: PerlIO internal IO abstraction interface
108 It is now possible to build Perl with AT&T's sfio IO package
109 instead of stdio. See L<perlapio> for more details, and
110 the F<INSTALL> file for how to use it.
112 =head2 New and Changed Built-in Variables
118 Extended error message on some platforms. (Also known as
119 $EXTENDED_OS_ERROR if you C<use English>).
123 The current set of syntax checks enabled by C<use strict>. See the
124 documentation of C<strict> for more details. Not actually new, but
126 Because it is intended for internal use by Perl core components,
127 there is no C<use English> long name for this variable.
131 By default, running out of memory it is not trappable. However, if
132 compiled for this, Perl may use the contents of C<$^M> as an emergency
133 pool after die()ing with this message. Suppose that your Perl were
134 compiled with -DEMERGENCY_SBRK and used Perl's malloc. Then
138 would allocate a 64K buffer for use when in emergency.
139 See the F<INSTALL> file for information on how to enable this option.
140 As a disincentive to casual use of this advanced feature,
141 there is no C<use English> long name for this variable.
145 =head2 New and Changed Built-in Functions
149 =item delete on slices
151 This now works. (e.g. C<delete @ENV{'PATH', 'MANPATH'}>)
155 is now supported on more platforms, prefers fcntl to lockf when
156 emulating, and always flushes before (un)locking.
158 =item printf and sprintf
160 now support "%i" as a synonym for "%d", and the "h" modifier.
161 So "%hi" means "short integer in decimal", and "%ho" means
162 "unsigned short integer as octal".
164 =item keys as an lvalue
166 As an lvalue, C<keys> allows you to increase the number of hash buckets
167 allocated for the given hash. This can gain you a measure of efficiency if
168 you know the hash is going to get big. (This is similar to pre-extending
169 an array by assigning a larger number to $#array.) If you say
173 then C<%hash> will have at least 200 buckets allocated for it. These
174 buckets will be retained even if you do C<%hash = ()>; use C<undef
175 %hash> if you want to free the storage while C<%hash> is still in scope.
176 You can't shrink the number of buckets allocated for the hash using
177 C<keys> in this way (but you needn't worry about doing this by accident,
178 as trying has no effect).
180 =item my() in Control Structures
182 You can now use my() (with or without the parentheses) in the control
183 expressions of control structures such as:
185 while (defined(my $line = <>)) {
191 if ((my $answer = <STDIN>) =~ /^y(es)?$/i) {
193 } elsif ($answer =~ /^n(o)?$/i) {
197 die "`$answer' is neither `yes' nor `no'";
200 Also, you can declare a foreach loop control variable as lexical by
201 preceding it with the word "my". For example, in:
203 foreach my $i (1, 2, 3) {
207 $i is a lexical variable, and the scope of $i extends to the end of
208 the loop, but not beyond it.
210 Note that you still cannot use my() on global punctuation variables
211 such as $_ and the like.
213 =item unpack() and pack()
215 A new format 'w' represents a BER compressed integer (as defined in
216 ASN.1). Its format is a sequence of one or more bytes, each of which
217 provides seven bits of the total value, with the most significant
218 first. Bit eight of each byte is set, except for the last byte, in
219 which bit eight is clear.
223 If the first argument to C<use> is a number, it is treated as a version
224 number instead of a module name. If the version of the Perl interpreter
225 is less than VERSION, then an error message is printed and Perl exits
226 immediately. Because C<use> occurs at compile time, this check happens
227 immediately during the compilation process, unlike C<require VERSION>,
228 which waits until run-time for the check. This is often useful if you
229 need to check the current Perl version before C<use>ing library modules
230 which have changed in incompatible ways from older versions of Perl.
231 (We try not to do this more than we have to.)
233 =item use Module VERSION LIST
235 If the VERSION argument is present between Module and LIST, then the
236 C<use> will call the VERSION method in class Module with the given
237 version as an argument. The default VERSION method, inherited from
238 the Universal class, croaks if the given version is larger than the
239 value of the variable $Module::VERSION. (Note that there is not a
240 comma after VERSION!)
242 This version-checking mechanism is similar to the one currently used
243 in the Exporter module, but it is faster and can be used with modules
244 that don't use the Exporter. It is the recommended method for new
247 =item prototype(FUNCTION)
249 Returns the prototype of a function as a string (or C<undef> if the
250 function has no prototype). FUNCTION is a reference to or the name of the
251 function whose prototype you want to retrieve.
252 (Not actually new; just never documented before.)
256 The default seed for C<srand>, which used to be C<time>, has been changed.
257 Now it's a heady mix of difficult-to-predict system-dependent values,
258 which should be sufficient for most everyday purposes.
260 Previous to version 5.004, calling C<rand> without first calling C<srand>
261 would yield the same sequence of random numbers on most or all machines.
262 Now, when perl sees that you're calling C<rand> and haven't yet called
263 C<srand>, it calls C<srand> with the default seed. You should still call
264 C<srand> manually if your code might ever be run on a pre-5.004 system,
265 of course, or if you want a seed other than the default.
269 Functions documented in the Camel to default to $_ now in
270 fact do, and all those that do are so documented in L<perlfunc>.
272 =item C<m//g> does not trigger a pos() reset on failure
274 The C<m//g> match iteration construct used to reset the iteration
275 when it failed to match (so that the next C<m//g> match would start at
276 the beginning of the string). You now have to explicitly do a
277 C<pos $str = 0;> to reset the "last match" position, or modify the
278 string in some way. This change makes it practical to chain C<m//g>
279 matches together in conjunction with ordinary matches using the C<\G>
280 zero-width assertion. See L<perlop> and L<perlre>.
282 =item C<m//x> ignores whitespace before ?*+{}
284 The C<m//x> construct has always been intended to ignore all unescaped
285 whitespace. However, before Perl 5.004, whitespace had the effect of
286 esacping repeat modifier like "*" or "?". For example, C</a *b/x> was
287 (mis)interpreted as C</a\*b/x>. This bug has been fixed in 5.004.
289 =item nested C<sub{}> closures work now
291 Prior to the 5.004 release, nested anonymous functions didn't work
294 =item formats work right on changing lexicals
296 Just like anonymous functions that contain lexical variables
297 that change (like a lexical index variable for a C<foreach> loop),
298 formats now work properly. For example, this silently failed
299 before, and is fine now:
302 foreach $i ( 1 .. 10 ) {
312 =head2 New Built-in Methods
314 The C<UNIVERSAL> package automatically contains the following methods that
315 are inherited by all other classes:
321 C<isa> returns I<true> if its object is blessed into a subclass of C<CLASS>
323 C<isa> is also exportable and can be called as a sub with two arguments. This
324 allows the ability to check what a reference points to. Example:
326 use UNIVERSAL qw(isa);
328 if(isa($ref, 'ARRAY')) {
334 C<can> checks to see if its object has a method called C<METHOD>,
335 if it does then a reference to the sub is returned; if it does not then
336 I<undef> is returned.
338 =item VERSION( [NEED] )
340 C<VERSION> returns the version number of the class (package). If the
341 NEED argument is given then it will check that the current version (as
342 defined by the $VERSION variable in the given package) not less than
343 NEED; it will die if this is not the case. This method is normally
344 called as a class method. This method is called automatically by the
345 C<VERSION> form of C<use>.
347 use A 1.2 qw(some imported subs);
353 B<NOTE:> C<can> directly uses Perl's internal code for method lookup, and
354 C<isa> uses a very similar method and caching strategy. This may cause
355 strange effects if the Perl code dynamically changes @ISA in any package.
357 You may add other methods to the UNIVERSAL class via Perl or XS code.
358 You do not need to C<use UNIVERSAL> in order to make these methods
359 available to your program. This is necessary only if you wish to
360 have C<isa> available as a plain subroutine in the current package.
362 =head2 TIEHANDLE Now Supported
364 See L<perltie> for other kinds of tie()s.
368 =item TIEHANDLE classname, LIST
370 This is the constructor for the class. That means it is expected to
371 return an object of some sort. The reference can be used to
372 hold some internal information.
377 return bless \$i, shift;
380 =item PRINT this, LIST
382 This method will be triggered every time the tied handle is printed to.
383 Beyond its self reference it also expects the list that was passed to
389 return print join( $, => map {uc} @_), $\;
394 This method will be called when the handle is read from via the C<read>
395 or C<sysread> functions.
399 my($buf,$len,$offset) = @_;
400 print "READ called, \$buf=$buf, \$len=$len, \$offset=$offset";
405 This method will be called when the handle is read from. The method
406 should return undef when there is no more data.
410 return "PRINT called $$r times\n"
415 This method will be called when the C<getc> function is called.
417 sub GETC { print "Don't GETC, Get Perl"; return "a"; }
421 As with the other types of ties, this method will be called when the
422 tied handle is about to be destroyed. This is useful for debugging and
423 possibly for cleaning up.
431 =head2 Malloc Enhancements
433 Four new compilation flags are recognized by malloc.c. (They have no
434 effect if perl is compiled with system malloc().)
438 =item -DDEBUGGING_MSTATS
440 If perl is compiled with C<DEBUGGING_MSTATS> defined, you can print
441 memory statistics at runtime by running Perl thusly:
443 env PERL_DEBUG_MSTATS=2 perl your_script_here
445 The value of 2 means to print statistics after compilation and on
446 exit; with a value of 1, the statistics ares printed only on exit.
447 (If you want the statistics at an arbitrary time, you'll need to
448 install the optional module Devel::Peek.)
450 =item -DEMERGENCY_SBRK
452 If this macro is defined, running out of memory need not be a fatal
453 error: a memory pool can allocated by assigning to the special
454 variable C<$^M>. See L<"$^M">.
458 Perl memory allocation is by bucket with sizes close to powers of two.
459 Because of these malloc overhead may be big, especially for data of
460 size exactly a power of two. If C<PACK_MALLOC> is defined, perl uses
461 a slightly different algorithm for small allocations (up to 64 bytes
462 long), which makes it possible to have overhead down to 1 byte for
463 allocations which are powers of two (and appear quite often).
465 Expected memory savings (with 8-byte alignment in C<alignbytes>) is
466 about 20% for typical Perl usage. Expected slowdown due to additional
467 malloc overhead is in fractions of a percent (hard to measure, because
468 of the effect of saved memory on speed).
470 =item -DTWO_POT_OPTIMIZE
472 Similarly to C<PACK_MALLOC>, this macro improves allocations of data
473 with size close to a power of two; but this works for big allocations
474 (starting with 16K by default). Such allocations are typical for big
475 hashes and special-purpose scripts, especially image processing.
477 On recent systems, the fact that perl requires 2M from system for 1M
478 allocation will not affect speed of execution, since the tail of such
479 a chunk is not going to be touched (and thus will not require real
480 memory). However, it may result in a premature out-of-memory error.
481 So if you will be manipulating very large blocks with sizes close to
482 powers of two, it would be wise to define this macro.
484 Expected saving of memory is 0-100% (100% in applications which
485 require most memory in such 2**n chunks); expected slowdown is
490 =head2 Miscellaneous Efficiency Enhancements
492 Functions that have an empty prototype and that do nothing but return
493 a fixed value are now inlined (e.g. C<sub PI () { 3.14159 }>).
495 Each unique hash key is only allocated once, no matter how many hashes
496 have an entry with that key. So even if you have 100 copies of the
497 same hash, the hash keys never have to be reallocated.
501 Four new pragmatic modules exist:
509 Looks for MakeMaker-like I<'blib'> directory structure starting in
510 I<dir> (or current directory) and working back up to five levels of
513 Intended for use on command line with B<-M> option as a way of testing
514 arbitrary scripts against an uninstalled version of a package.
518 Tells the compiler to enable (or disable) the use of POSIX locales for
521 When C<use locale> is in effect, the current LC_CTYPE locale is used
522 for regular expressions and case mapping; LC_COLLATE for string
523 ordering; and LC_NUMERIC for numeric formating in printf and sprintf
524 (but B<not> in print). LC_NUMERIC is always used in write, since
525 lexical scoping of formats is problematic at best.
527 Each C<use locale> or C<no locale> affects statements to the end of
528 the enclosing BLOCK or, if not inside a BLOCK, to the end of the
529 current file. Locales can be switched and queried with
532 See L<perllocale> for more information.
536 Disable unsafe opcodes, or any named opcodes, when compiling Perl code.
540 Enable VMS-specific language features. Currently, there are three
541 VMS-specific features available: 'status', which makes C<$?> and
542 C<system> return genuine VMS status values instead of emulating POSIX;
543 'exit', which makes C<exit> take a genuine VMS status value instead of
544 assuming that C<exit 1> is an error; and 'time', which makes all times
545 relative to the local time zone, in the VMS tradition.
551 =head2 Installation Directories
553 The I<installperl> script now places the Perl source files for
554 extensions in the architecture-specific library directory, which is
555 where the shared libraries for extensions have always been. This
556 change is intended to allow administrators to keep the Perl 5.004
557 library directory unchanged from a previous version, without running
558 the risk of binary incompatibility between extensions' Perl source and
563 New constants in the existing Fcntl modules are now supported,
564 provided that your operating system happens to support them:
567 O_ASYNC O_DEFER O_DSYNC O_FSYNC O_SYNC
570 These constants are intended for use with the Perl operators sysopen()
571 and fcntl() and the basic database modules like SDBM_File. For the
572 exact meaning of these and other Fcntl constants please refer to your
573 operating system's documentation for fcntl() and open().
575 In addition, the Fcntl module now provides these constants for use
576 with the Perl operator flock():
578 LOCK_SH LOCK_EX LOCK_NB LOCK_UN
580 These constants are defined in all environments (because where there is
581 no flock() system call, Perl emulates it). However, for historical
582 reasons, these constants are not exported unless they are explicitly
583 requested with the ":flock" tag (e.g. C<use Fcntl ':flock'>).
585 =head2 Module Information Summary
587 Brand new modules, arranged by topic rather than strictly
590 CPAN interface to Comprehensive Perl Archive Network
591 CPAN::FirstTime create a CPAN configuration file
592 CPAN::Nox run CPAN while avoiding compiled extensions
594 IO.pm Top-level interface to IO::* classes
595 IO/File.pm IO::File extension Perl module
596 IO/Handle.pm IO::Handle extension Perl module
597 IO/Pipe.pm IO::Pipe extension Perl module
598 IO/Seekable.pm IO::Seekable extension Perl module
599 IO/Select.pm IO::Select extension Perl module
600 IO/Socket.pm IO::Socket extension Perl module
602 Opcode.pm Disable named opcodes when compiling Perl code
604 ExtUtils/Embed.pm Utilities for embedding Perl in C programs
605 ExtUtils/testlib.pm Fixes up @INC to use just-built extension
607 FindBin.pm Find path of currently executing program
609 Class/Template.pm Structure/member template builder
610 File/stat.pm Object-oriented wrapper around CORE::stat
611 Net/hostent.pm Object-oriented wrapper around CORE::gethost*
612 Net/netent.pm Object-oriented wrapper around CORE::getnet*
613 Net/protoent.pm Object-oriented wrapper around CORE::getproto*
614 Net/servent.pm Object-oriented wrapper around CORE::getserv*
615 Time/gmtime.pm Object-oriented wrapper around CORE::gmtime
616 Time/localtime.pm Object-oriented wrapper around CORE::localtime
617 Time/tm.pm Perl implementation of "struct tm" for {gm,local}time
618 User/grent.pm Object-oriented wrapper around CORE::getgr*
619 User/pwent.pm Object-oriented wrapper around CORE::getpw*
621 Tie/RefHash.pm Base class for tied hashes with references as keys
623 UNIVERSAL.pm Base class for *ALL* classes
627 The IO module provides a simple mechanism to load all of the IO modules at one
628 go. Currently this includes:
636 For more information on any of these modules, please see its
637 respective documentation.
641 The Math::Complex module has been totally rewritten, and now supports
642 more operations. These are overloaded:
644 + - * / ** <=> neg ~ abs sqrt exp log sin cos atan2 "" (stringify)
646 And these functions are now exported:
650 tan cotan asin acos atan acotan
651 sinh cosh tanh cotanh asinh acosh atanh acotanh
656 There have been quite a few changes made to DB_File. Here are a few of
663 Fixed a handful of bugs.
667 By public demand, added support for the standard hash function exists().
671 Made it compatible with Berkeley DB 1.86.
675 Made negative subscripts work with RECNO interface.
679 Changed the default flags from O_RDWR to O_CREAT|O_RDWR and the default
680 mode from 0640 to 0666.
684 Made DB_File automatically import the open() constants (O_RDWR,
685 O_CREAT etc.) from Fcntl, if available.
689 Updated documentation.
693 Refer to the HISTORY section in DB_File.pm for a complete list of
694 changes. Everything after DB_File 1.01 has been added since 5.003.
698 Major rewrite - support added for both udp echo and real icmp pings.
700 =head2 Overridden Built-ins
702 Many of the Perl built-ins returning lists now have
703 object-oriented overrides. These are:
715 For example, you can now say
719 $his = (stat($filename)->st_uid == pwent($whoever)->pw_uid);
721 =head1 Utility Changes
727 =item C<void> XSUBs now default to returning nothing
729 Due to a documentation/implementation bug in previous versions of
730 Perl, XSUBs with a return type of C<void> have actually been
731 returning one value. Usually that value was the GV for the XSUB,
732 but sometimes it was some already freed or reused value, which would
733 sometimes lead to program failure.
735 In Perl 5.004, if an XSUB is declared as returning C<void>, it
736 actually returns no value, i.e. an empty list (though there is a
737 backward-compatibility exception; see below). If your XSUB really
738 does return an SV, you should give it a return type of C<SV *>.
740 For backward compatibility, I<xsubpp> tries to guess whether a
741 C<void> XSUB is really C<void> or if it wants to return an C<SV *>.
742 It does so by examining the text of the XSUB: if I<xsubpp> finds
743 what looks like an assignment to C<ST(0)>, it assumes that the
744 XSUB's return type is really C<SV *>.
748 =head1 C Language API Changes
752 =item C<gv_fetchmethod> and C<perl_call_sv>
754 The C<gv_fetchmethod> function finds a method for an object, just like
755 in Perl 5.003. The GV it returns may be a method cache entry.
756 However, in Perl 5.004, method cache entries are not visible to users;
757 therefore, they can no longer be passed directly to C<perl_call_sv>.
758 Instead, you should use the C<GvCV> macro on the GV to extract its CV,
759 and pass the CV to C<perl_call_sv>.
761 The most likely symptom of passing the result of C<gv_fetchmethod> to
762 C<perl_call_sv> is Perl's producing an "Undefined subroutine called"
763 error on the I<second> call to a given method (since there is no cache
766 =item Extended API for manipulating hashes
768 Internal handling of hash keys has changed. The old hashtable API is
769 still fully supported, and will likely remain so. The additions to the
770 API allow passing keys as C<SV*>s, so that C<tied> hashes can be given
771 real scalars as keys rather than plain strings (non-tied hashes still
772 can only use strings as keys). New extensions must use the new hash
773 access functions and macros if they wish to use C<SV*> keys. These
774 additions also make it feasible to manipulate C<HE*>s (hash entries),
775 which can be more efficient. See L<perlguts> for details.
779 =head1 Documentation Changes
781 Many of the base and library pods were updated. These
782 new pods are included in section 1:
792 Locale support (internationalization and localization).
796 Tutorial on Perl OO programming.
800 Perl internal IO abstraction interface.
804 Although not new, this has been massively updated.
808 Although not new, this has been massively updated.
812 =head1 New Diagnostics
814 Several new conditions will trigger warnings that were
815 silent before. Some only affect certain platforms.
816 The following new warnings and errors outline these.
817 These messages are classified as follows (listed in
818 increasing order of desperation):
820 (W) A warning (optional).
821 (D) A deprecation (optional).
822 (S) A severe warning (mandatory).
823 (F) A fatal error (trappable).
824 (P) An internal error you should never see (trappable).
825 (X) A very fatal error (non-trappable).
826 (A) An alien error message (not generated by Perl).
830 =item "my" variable %s masks earlier declaration in same scope
832 (S) A lexical variable has been redeclared in the same scope, effectively
833 eliminating all access to the previous instance. This is almost always
834 a typographical error. Note that the earlier variable will still exist
835 until the end of the scope or until all closure referents to it are
838 =item %s argument is not a HASH element or slice
840 (F) The argument to delete() must be either a hash element, such as
843 $ref->[12]->{"susie"}
845 or a hash slice, such as
847 @foo{$bar, $baz, $xyzzy}
848 @{$ref->[12]}{"susie", "queue"}
850 =item Allocation too large: %lx
852 (X) You can't allocate more than 64K on an MSDOS machine.
854 =item Allocation too large
856 (F) You can't allocate more than 2^31+"small amount" bytes.
858 =item Attempt to free non-existent shared string
860 (P) Perl maintains a reference counted internal table of strings to
861 optimize the storage and access of hash keys and other strings. This
862 indicates someone tried to decrement the reference count of a string
863 that can no longer be found in the table.
865 =item Attempt to use reference as lvalue in substr
867 (W) You supplied a reference as the first argument to substr() used
868 as an lvalue, which is pretty strange. Perhaps you forgot to
869 dereference it first. See L<perlfunc/substr>.
871 =item Unsupported function fork
873 (F) Your version of executable does not support forking.
875 Note that under some systems, like OS/2, there may be different flavors of
876 Perl executables, some of which may support fork, some not. Try changing
877 the name you call Perl by to C<perl_>, C<perl__>, and so on.
879 =item Ill-formed logical name |%s| in prime_env_iter
881 (W) A warning peculiar to VMS. A logical name was encountered when preparing
882 to iterate over %ENV which violates the syntactic rules governing logical
883 names. Since it cannot be translated normally, it is skipped, and will not
884 appear in %ENV. This may be a benign occurrence, as some software packages
885 might directly modify logical name tables and introduce non-standard names,
886 or it may indicate that a logical name table has been corrupted.
888 =item Can't use bareword ("%s") as %s ref while "strict refs" in use
890 (F) Only hard references are allowed by "strict refs". Symbolic references
891 are disallowed. See L<perlref>.
893 =item Constant subroutine %s redefined
895 (S) You redefined a subroutine which had previously been eligible for
896 inlining. See L<perlsub/"Constant Functions"> for commentary and
901 (F) You passed die() an empty string (the equivalent of C<die "">) or
902 you called it with no args and both C<$@> and C<$_> were empty.
904 =item Integer overflow in hex number
906 (S) The literal hex number you have specified is too big for your
907 architecture. On a 32-bit architecture the largest hex literal is
910 =item Integer overflow in octal number
912 (S) The literal octal number you have specified is too big for your
913 architecture. On a 32-bit architecture the largest octal literal is
916 =item Name "%s::%s" used only once: possible typo
918 (W) Typographical errors often show up as unique variable names.
919 If you had a good reason for having a unique name, then just mention
920 it again somehow to suppress the message (the C<use vars> pragma is
921 provided for just this purpose).
923 =item Null picture in formline
925 (F) The first argument to formline must be a valid format picture
926 specification. It was found to be empty, which probably means you
927 supplied it an uninitialized value. See L<perlform>.
929 =item Offset outside string
931 (F) You tried to do a read/write/send/recv operation with an offset
932 pointing outside the buffer. This is difficult to imagine.
933 The sole exception to this is that C<sysread()>ing past the buffer
934 will extend the buffer and zero pad the new area.
936 =item Stub found while resolving method `%s' overloading `%s' in package `%s'
938 (P) Overloading resolution over @ISA tree may be broken by importing stubs.
939 Stubs should never be implicitely created, but explicit calls to C<can>
942 =item Cannot resolve method `%s' overloading `%s' in package `s'
944 (P) Internal error trying to resolve overloading specified by a method
945 name (as opposed to a subroutine reference).
949 (X|F) The malloc() function returned 0, indicating there was insufficient
950 remaining memory (or virtual memory) to satisfy the request.
952 The request was judged to be small, so the possibility to trap it
953 depends on the way Perl was compiled. By default it is not trappable.
954 However, if compiled for this, Perl may use the contents of C<$^M> as
955 an emergency pool after die()ing with this message. In this case the
956 error is trappable I<once>.
958 =item Out of memory during request for %s
960 (F) The malloc() function returned 0, indicating there was insufficient
961 remaining memory (or virtual memory) to satisfy the request. However,
962 the request was judged large enough (compile-time default is 64K), so
963 a possibility to shut down by trapping this error is granted.
965 =item Possible attempt to put comments in qw() list
967 (W) qw() lists contain items separated by whitespace; as with literal
968 strings, comment characters are not ignored, but are instead treated
969 as literal data. (You may have used different delimiters than the
970 exclamation marks parentheses shown here; braces are also frequently
973 You probably wrote something like this:
980 when you should have written this:
987 If you really want comments, build your list the
988 old-fashioned way, with quotes and commas:
992 'b', # another comment
995 =item Possible attempt to separate words with commas
997 (W) qw() lists contain items separated by whitespace; therefore commas
998 aren't needed to separate the items. (You may have used different
999 delimiters than the parentheses shown here; braces are also frequently
1002 You probably wrote something like this:
1006 which puts literal commas into some of the list items. Write it without
1007 commas if you don't want them to appear in your data:
1011 =item Scalar value @%s{%s} better written as $%s{%s}
1013 (W) You've used a hash slice (indicated by @) to select a single element of
1014 a hash. Generally it's better to ask for a scalar value (indicated by $).
1015 The difference is that C<$foo{&bar}> always behaves like a scalar, both when
1016 assigning to it and when evaluating its argument, while C<@foo{&bar}> behaves
1017 like a list when you assign to it, and provides a list context to its
1018 subscript, which can do weird things if you're expecting only one subscript.
1020 =item untie attempted while %d inner references still exist
1022 (W) A copy of the object returned from C<tie> (or C<tied>) was still
1023 valid when C<untie> was called.
1025 =item Value of %s construct can be "0"; test with defined()
1027 (W) In a conditional expression, you used <HANDLE>, <*> (glob), or
1028 C<readdir> as a boolean value. Each of these constructs can return a
1029 value of "0"; that would make the conditional expression false, which
1030 is probably not what you intended. When using these constructs in
1031 conditional expressions, test their values with the C<defined> operator.
1033 =item Variable "%s" may be unavailable
1035 (W) An inner (nested) I<anonymous> subroutine is inside a I<named>
1036 subroutine, and outside that is another subroutine; and the anonymous
1037 (innermost) subroutine is referencing a lexical variable defined in
1038 the outermost subroutine. For example:
1040 sub outermost { my $a; sub middle { sub { $a } } }
1042 If the anonymous subroutine is called or referenced (directly or
1043 indirectly) from the outermost subroutine, it will share the variable
1044 as you would expect. But if the anonymous subroutine is called or
1045 referenced when the outermost subroutine is not active, it will see
1046 the value of the shared variable as it was before and during the
1047 *first* call to the outermost subroutine, which is probably not what
1050 In these circumstances, it is usually best to make the middle
1051 subroutine anonymous, using the C<sub {}> syntax. Perl has specific
1052 support for shared variables in nested anonymous subroutines; a named
1053 subroutine in between interferes with this feature.
1055 =item Variable "%s" will not stay shared
1057 (W) An inner (nested) I<named> subroutine is referencing a lexical
1058 variable defined in an outer subroutine.
1060 When the inner subroutine is called, it will probably see the value of
1061 the outer subroutine's variable as it was before and during the
1062 *first* call to the outer subroutine; in this case, after the first
1063 call to the outer subroutine is complete, the inner and outer
1064 subroutines will no longer share a common value for the variable. In
1065 other words, the variable will no longer be shared.
1067 Furthermore, if the outer subroutine is anonymous and references a
1068 lexical variable outside itself, then the outer and inner subroutines
1069 will I<never> share the given variable.
1071 This problem can usually be solved by making the inner subroutine
1072 anonymous, using the C<sub {}> syntax. When inner anonymous subs that
1073 reference variables in outer subroutines are called or referenced,
1074 they are automatically re-bound to the current values of such
1077 =item Warning: something's wrong
1079 (W) You passed warn() an empty string (the equivalent of C<warn "">) or
1080 you called it with no args and C<$_> was empty.
1082 =item Got an error from DosAllocMem
1084 (P) An error peculiar to OS/2. Most probably you're using an obsolete
1085 version of Perl, and this should not happen anyway.
1087 =item Malformed PERLLIB_PREFIX
1089 (F) An error peculiar to OS/2. PERLLIB_PREFIX should be of the form
1097 with non-empty prefix1 and prefix2. If C<prefix1> is indeed a prefix of
1098 a builtin library search path, prefix2 is substituted. The error may appear
1099 if components are not found, or are too long. See L<perlos2/"PERLLIB_PREFIX">.
1101 =item PERL_SH_DIR too long
1103 (F) An error peculiar to OS/2. PERL_SH_DIR is the directory to find the
1104 C<sh>-shell in. See L<perlos2/"PERL_SH_DIR">.
1106 =item Process terminated by SIG%s
1108 (W) This is a standard message issued by OS/2 applications, while *nix
1109 applications die in silence. It is considered a feature of the OS/2
1110 port. One can easily disable this by appropriate sighandlers, see
1111 L<perlipc/"Signals">. See L<perlos2/"Process terminated by SIGTERM/SIGINT">.
1117 If you find what you think is a bug, you might check the headers of
1118 recently posted articles in the comp.lang.perl.misc newsgroup.
1119 There may also be information at http://www.perl.com/perl/, the Perl
1122 If you believe you have an unreported bug, please run the B<perlbug>
1123 program included with your release. Make sure you trim your bug down
1124 to a tiny but sufficient test case. Your bug report, along with the
1125 output of C<perl -V>, will be sent off to <F<perlbug@perl.com>> to be
1126 analysed by the Perl porting team.
1130 The F<Changes> file for exhaustive details on what changed.
1132 The F<INSTALL> file for how to build Perl. This file has been
1133 significantly updated for 5.004, so even veteran users should
1136 The F<README> file for general stuff.
1138 The F<Copying> file for copyright information.
1142 Constructed by Tom Christiansen, grabbing material with permission
1143 from innumerable contributors, with kibitzing by more than a few Perl
1146 Last update: Sat Mar 8 19:51:26 EST 1997