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 No Autovivification of Subroutine Parameters
32 In Perl versions 5.002 and 5.003, array and hash elements used as
33 subroutine parameters were "autovivified"; that is, they were brought
34 into existence if they did not already exist. For example, calling
35 C<func($h{foo})> would create C<$h{foo}> if it did not already exist,
36 causing C<exists $h{foo}> to become true and C<keys %h> to return
39 Perl 5.004 returns to the pre-5.002 behavior of I<not> autovivifying
40 array and hash elements used as subroutine parameters.
42 =head2 Fixed Parsing of $$<digit>, &$<digit>, etc.
44 A bug in previous versions of Perl 5.0 prevented proper parsing of
45 numeric special variables as symbolic references. That bug has been
46 fixed. As a result, the string "$$0" is no longer equivalent to
47 C<$$."0">, but rather to C<${$0}>. To get the old behavior, change
48 "$$" followed by a digit to "${$}".
50 =head2 No Resetting of $. on Implicit Close
52 The documentation for Perl 5.0 has always stated that C<$.> is I<not>
53 reset when an already-open file handle is re-opened with no intervening
54 call to C<close>. Due to a bug, perl versions 5.000 through 5.0003
55 I<did> reset C<$.> under that circumstance; Perl 5.004 does not.
57 =head2 Changes to Tainting Checks
59 A bug in previous versions may have failed to detect some insecure
60 conditions when taint checks are turned on. (Taint checks are used
61 in setuid or setgid scripts, or when explicitly turned on with the
62 C<-T> invocation option.) Although it's unlikely, this may cause a
63 previously-working script to now fail -- which should be construed
64 as a blessing, since that indicates a potentially-serious security
65 hole was just plugged.
67 =head2 New Opcode Module and Revised Safe Module
69 A new Opcode module supports the creation, manipulation and
70 application of opcode masks. The revised Safe module has a new API
71 and is implemented using the new Opcode module. Please read the new
72 Opcode and Safe documentation.
74 =head2 Embedding Improvements
76 In older versions of Perl it was not possible to create more than one
77 Perl interpreter instance inside a single process without leaking like a
78 sieve and/or crashing. The bugs that caused this behavior have all been
79 fixed. However, you still must take care when embedding Perl in a C
80 program. See the updated perlembed manpage for tips on how to manage
83 =head2 Internal Change: FileHandle Class Based on IO::* Classes
85 File handles are now stored internally as type IO::Handle. The
86 FileHandle module is still supported for backwards compatibility, but
87 it is now merely a front end to the IO::* modules -- specifically,
88 IO::Handle, IO::Seekable, and IO::File. We suggest, but do not
89 require, that you use the IO::* modules in new code.
91 In harmony with this change, C<*GLOB{FILEHANDLE}> is now a
92 backward-compatible synonym for C<*STDOUT{IO}>.
94 =head2 Internal Change: PerlIO internal IO abstraction interface
96 It is now possible to build Perl with AT&T's sfio IO package
97 instead of stdio. See L<perlapio> for more details, and
98 the F<INSTALL> file for how to use it.
100 =head2 New and Changed Built-in Variables
106 Extended error message on some platforms. (Also known as
107 $EXTENDED_OS_ERROR if you C<use English>).
111 The current set of syntax checks enabled by C<use strict>. See the
112 documentation of C<strict> for more details. Not actually new, but
114 Because it is intended for internal use by Perl core components,
115 there is no C<use English> long name for this variable.
119 By default, running out of memory it is not trappable. However, if
120 compiled for this, Perl may use the contents of C<$^M> as an emergency
121 pool after die()ing with this message. Suppose that your Perl were
122 compiled with -DEMERGENCY_SBRK and used Perl's malloc. Then
126 would allocate a 64K buffer for use when in emergency.
127 See the F<INSTALL> file for information on how to enable this option.
128 As a disincentive to casual use of this advanced feature,
129 there is no C<use English> long name for this variable.
133 =head2 New and Changed Built-in Functions
137 =item delete on slices
139 This now works. (e.g. C<delete @ENV{'PATH', 'MANPATH'}>)
143 is now supported on more platforms, prefers fcntl to lockf when
144 emulating, and always flushes before (un)locking.
146 =item printf and sprintf
148 now support "%i" as a synonym for "%d", and the "h" modifier.
149 So "%hi" means "short integer in decimal", and "%ho" means
150 "unsigned short integer as octal".
152 =item keys as an lvalue
154 As an lvalue, C<keys> allows you to increase the number of hash buckets
155 allocated for the given hash. This can gain you a measure of efficiency if
156 you know the hash is going to get big. (This is similar to pre-extending
157 an array by assigning a larger number to $#array.) If you say
161 then C<%hash> will have at least 200 buckets allocated for it. These
162 buckets will be retained even if you do C<%hash = ()>; use C<undef
163 %hash> if you want to free the storage while C<%hash> is still in scope.
164 You can't shrink the number of buckets allocated for the hash using
165 C<keys> in this way (but you needn't worry about doing this by accident,
166 as trying has no effect).
168 =item my() in Control Structures
170 You can now use my() (with or without the parentheses) in the control
171 expressions of control structures such as:
173 while (defined(my $line = <>)) {
179 if ((my $answer = <STDIN>) =~ /^y(es)?$/i) {
181 } elsif ($answer =~ /^n(o)?$/i) {
185 die "`$answer' is neither `yes' nor `no'";
188 Also, you can declare a foreach loop control variable as lexical by
189 preceding it with the word "my". For example, in:
191 foreach my $i (1, 2, 3) {
195 $i is a lexical variable, and the scope of $i extends to the end of
196 the loop, but not beyond it.
198 Note that you still cannot use my() on global punctuation variables
199 such as $_ and the like.
201 =item unpack() and pack()
203 A new format 'w' represents a BER compressed integer (as defined in
204 ASN.1). Its format is a sequence of one or more bytes, each of which
205 provides seven bits of the total value, with the most significant
206 first. Bit eight of each byte is set, except for the last byte, in
207 which bit eight is clear.
211 If the first argument to C<use> is a number, it is treated as a version
212 number instead of a module name. If the version of the Perl interpreter
213 is less than VERSION, then an error message is printed and Perl exits
214 immediately. Because C<use> occurs at compile time, this check happens
215 immediately during the compilation process, unlike C<require VERSION>,
216 which waits until run-time for the check. This is often useful if you
217 need to check the current Perl version before C<use>ing library modules
218 which have changed in incompatible ways from older versions of Perl.
219 (We try not to do this more than we have to.)
221 =item use Module VERSION LIST
223 If the VERSION argument is present between Module and LIST, then the
224 C<use> will call the VERSION method in class Module with the given
225 version as an argument. The default VERSION method, inherited from
226 the Universal class, croaks if the given version is larger than the
227 value of the variable $Module::VERSION. (Note that there is not a
228 comma after VERSION!)
230 This version-checking mechanism is similar to the one currently used
231 in the Exporter module, but it is faster and can be used with modules
232 that don't use the Exporter. It is the recommended method for new
235 =item prototype(FUNCTION)
237 Returns the prototype of a function as a string (or C<undef> if the
238 function has no prototype). FUNCTION is a reference to or the name of the
239 function whose prototype you want to retrieve.
240 (Not actually new; just never documented before.)
244 The default seed for C<srand>, which used to be C<time>, has been changed.
245 Now it's a heady mix of difficult-to-predict system-dependent values,
246 which should be sufficient for most everyday purposes.
248 Previous to version 5.004, calling C<rand> without first calling C<srand>
249 would yield the same sequence of random numbers on most or all machines.
250 Now, when perl sees that you're calling C<rand> and haven't yet called
251 C<srand>, it calls C<srand> with the default seed. You should still call
252 C<srand> manually if your code might ever be run on a pre-5.004 system,
253 of course, or if you want a seed other than the default.
257 Functions documented in the Camel to default to $_ now in
258 fact do, and all those that do are so documented in L<perlfunc>.
260 =item C<m//g> does not trigger a pos() reset on failure
262 The C<m//g> match iteration construct used to reset the iteration
263 when it failed to match (so that the next C<m//g> match would start at
264 the beginning of the string). You now have to explicitly do a
265 C<pos $str = 0;> to reset the "last match" position, or modify the
266 string in some way. This change makes it practical to chain C<m//g>
267 matches together in conjunction with ordinary matches using the C<\G>
268 zero-width assertion. See L<perlop> and L<perlre>.
270 =item nested C<sub{}> closures work now
272 Prior to the 5.004 release, nested anonymous functions didn't work
275 =item formats work right on changing lexicals
277 Just like anonymous functions that contain lexical variables
278 that change (like a lexical index variable for a C<foreach> loop),
279 formats now work properly. For example, this silently failed
280 before, and is fine now:
283 foreach $i ( 1 .. 10 ) {
293 =head2 New Built-in Methods
295 The C<UNIVERSAL> package automatically contains the following methods that
296 are inherited by all other classes:
302 C<isa> returns I<true> if its object is blessed into a subclass of C<CLASS>
304 C<isa> is also exportable and can be called as a sub with two arguments. This
305 allows the ability to check what a reference points to. Example:
307 use UNIVERSAL qw(isa);
309 if(isa($ref, 'ARRAY')) {
315 C<can> checks to see if its object has a method called C<METHOD>,
316 if it does then a reference to the sub is returned; if it does not then
317 I<undef> is returned.
319 =item VERSION( [NEED] )
321 C<VERSION> returns the version number of the class (package). If the
322 NEED argument is given then it will check that the current version (as
323 defined by the $VERSION variable in the given package) not less than
324 NEED; it will die if this is not the case. This method is normally
325 called as a class method. This method is called automatically by the
326 C<VERSION> form of C<use>.
328 use A 1.2 qw(some imported subs);
334 B<NOTE:> C<can> directly uses Perl's internal code for method lookup, and
335 C<isa> uses a very similar method and caching strategy. This may cause
336 strange effects if the Perl code dynamically changes @ISA in any package.
338 You may add other methods to the UNIVERSAL class via Perl or XS code.
339 You do not need to C<use UNIVERSAL> in order to make these methods
340 available to your program. This is necessary only if you wish to
341 have C<isa> available as a plain subroutine in the current package.
343 =head2 TIEHANDLE Now Supported
345 See L<perltie> for other kinds of tie()s.
349 =item TIEHANDLE classname, LIST
351 This is the constructor for the class. That means it is expected to
352 return an object of some sort. The reference can be used to
353 hold some internal information.
358 return bless \$i, shift;
361 =item PRINT this, LIST
363 This method will be triggered every time the tied handle is printed to.
364 Beyond its self reference it also expects the list that was passed to
370 return print join( $, => map {uc} @_), $\;
375 This method will be called when the handle is read from via the C<read>
376 or C<sysread> functions.
380 my($buf,$len,$offset) = @_;
381 print "READ called, \$buf=$buf, \$len=$len, \$offset=$offset";
386 This method will be called when the handle is read from. The method
387 should return undef when there is no more data.
391 return "PRINT called $$r times\n"
396 This method will be called when the C<getc> function is called.
398 sub GETC { print "Don't GETC, Get Perl"; return "a"; }
402 As with the other types of ties, this method will be called when the
403 tied handle is about to be destroyed. This is useful for debugging and
404 possibly for cleaning up.
412 =head2 Malloc Enhancements
414 Four new compilation flags are recognized by malloc.c. (They have no
415 effect if perl is compiled with system malloc().)
419 =item -DDEBUGGING_MSTATS
421 If perl is compiled with C<DEBUGGING_MSTATS> defined, you can print
422 memory statistics at runtime by running Perl thusly:
424 env PERL_DEBUG_MSTATS=2 perl your_script_here
426 The value of 2 means to print statistics after compilation and on
427 exit; with a value of 1, the statistics ares printed only on exit.
428 (If you want the statistics at an arbitrary time, you'll need to
429 install the optional module Devel::Peek.)
431 =item -DEMERGENCY_SBRK
433 If this macro is defined, running out of memory need not be a fatal
434 error: a memory pool can allocated by assigning to the special
435 variable C<$^M>. See L<"$^M">.
439 Perl memory allocation is by bucket with sizes close to powers of two.
440 Because of these malloc overhead may be big, especially for data of
441 size exactly a power of two. If C<PACK_MALLOC> is defined, perl uses
442 a slightly different algorithm for small allocations (up to 64 bytes
443 long), which makes it possible to have overhead down to 1 byte for
444 allocations which are powers of two (and appear quite often).
446 Expected memory savings (with 8-byte alignment in C<alignbytes>) is
447 about 20% for typical Perl usage. Expected slowdown due to additional
448 malloc overhead is in fractions of a percent (hard to measure, because
449 of the effect of saved memory on speed).
451 =item -DTWO_POT_OPTIMIZE
453 Similarly to C<PACK_MALLOC>, this macro improves allocations of data
454 with size close to a power of two; but this works for big allocations
455 (starting with 16K by default). Such allocations are typical for big
456 hashes and special-purpose scripts, especially image processing.
458 On recent systems, the fact that perl requires 2M from system for 1M
459 allocation will not affect speed of execution, since the tail of such
460 a chunk is not going to be touched (and thus will not require real
461 memory). However, it may result in a premature out-of-memory error.
462 So if you will be manipulating very large blocks with sizes close to
463 powers of two, it would be wise to define this macro.
465 Expected saving of memory is 0-100% (100% in applications which
466 require most memory in such 2**n chunks); expected slowdown is
471 =head2 Miscellaneous Efficiency Enhancements
473 Functions that have an empty prototype and that do nothing but return
474 a fixed value are now inlined (e.g. C<sub PI () { 3.14159 }>).
476 Each unique hash key is only allocated once, no matter how many hashes
477 have an entry with that key. So even if you have 100 copies of the
478 same hash, the hash keys never have to be reallocated.
482 Four new pragmatic modules exist:
490 Looks for MakeMaker-like I<'blib'> directory structure starting in
491 I<dir> (or current directory) and working back up to five levels of
494 Intended for use on command line with B<-M> option as a way of testing
495 arbitrary scripts against an uninstalled version of a package.
499 Tells the compiler to enable (or disable) the use of POSIX locales for
502 When C<use locale> is in effect, the current LC_CTYPE locale is used
503 for regular expressions and case mapping; LC_COLLATE for string
504 ordering; and LC_NUMERIC for numeric formating in printf and sprintf
505 (but B<not> in print). LC_NUMERIC is always used in write, since
506 lexical scoping of formats is problematic at best.
508 Each C<use locale> or C<no locale> affects statements to the end of
509 the enclosing BLOCK or, if not inside a BLOCK, to the end of the
510 current file. Locales can be switched and queried with
513 See L<perllocale> for more information.
517 Disable unsafe opcodes, or any named opcodes, when compiling Perl code.
521 Enable VMS-specific language features. Currently, there are three
522 VMS-specific features available: 'status', which makes C<$?> and
523 C<system> return genuine VMS status values instead of emulating POSIX;
524 'exit', which makes C<exit> take a genuine VMS status value instead of
525 assuming that C<exit 1> is an error; and 'time', which makes all times
526 relative to the local time zone, in the VMS tradition.
532 =head2 Installation Directories
534 The I<installperl> script now places the Perl source files for
535 extensions in the architecture-specific library directory, which is
536 where the shared libraries for extensions have always been. This
537 change is intended to allow administrators to keep the Perl 5.004
538 library directory unchanged from a previous version, without running
539 the risk of binary incompatibility between extensions' Perl source and
544 New constants in the existing Fcntl modules are now supported,
545 provided that your operating system happens to support them:
548 O_ASYNC O_DEFER O_DSYNC O_FSYNC O_SYNC
551 These constants are intended for use with the Perl operators sysopen()
552 and fcntl() and the basic database modules like SDBM_File. For the
553 exact meaning of these and other Fcntl constants please refer to your
554 operating system's documentation for fcntl() and open().
556 In addition, the Fcntl module now provides these constants for use
557 with the Perl operator flock():
559 LOCK_SH LOCK_EX LOCK_NB LOCK_UN
561 These constants are defined in all environments (because where there is
562 no flock() system call, Perl emulates it). However, for historical
563 reasons, these constants are not exported unless they are explicitly
564 requested with the ":flock" tag (e.g. C<use Fcntl ':flock'>).
566 =head2 Module Information Summary
568 Brand new modules, arranged by topic rather than strictly
571 CPAN interface to Comprehensive Perl Archive Network
572 CPAN::FirstTime create a CPAN configuration file
573 CPAN::Nox run CPAN while avoiding compiled extensions
575 IO.pm Top-level interface to IO::* classes
576 IO/File.pm IO::File extension Perl module
577 IO/Handle.pm IO::Handle extension Perl module
578 IO/Pipe.pm IO::Pipe extension Perl module
579 IO/Seekable.pm IO::Seekable extension Perl module
580 IO/Select.pm IO::Select extension Perl module
581 IO/Socket.pm IO::Socket extension Perl module
583 Opcode.pm Disable named opcodes when compiling Perl code
585 ExtUtils/Embed.pm Utilities for embedding Perl in C programs
586 ExtUtils/testlib.pm Fixes up @INC to use just-built extension
588 FindBin.pm Find path of currently executing program
590 Class/Template.pm Structure/member template builder
591 File/stat.pm Object-oriented wrapper around CORE::stat
592 Net/hostent.pm Object-oriented wrapper around CORE::gethost*
593 Net/netent.pm Object-oriented wrapper around CORE::getnet*
594 Net/protoent.pm Object-oriented wrapper around CORE::getproto*
595 Net/servent.pm Object-oriented wrapper around CORE::getserv*
596 Time/gmtime.pm Object-oriented wrapper around CORE::gmtime
597 Time/localtime.pm Object-oriented wrapper around CORE::localtime
598 Time/tm.pm Perl implementation of "struct tm" for {gm,local}time
599 User/grent.pm Object-oriented wrapper around CORE::getgr*
600 User/pwent.pm Object-oriented wrapper around CORE::getpw*
602 Tie/RefHash.pm Base class for tied hashes with references as keys
604 UNIVERSAL.pm Base class for *ALL* classes
608 The IO module provides a simple mechanism to load all of the IO modules at one
609 go. Currently this includes:
617 For more information on any of these modules, please see its
618 respective documentation.
622 The Math::Complex module has been totally rewritten, and now supports
623 more operations. These are overloaded:
625 + - * / ** <=> neg ~ abs sqrt exp log sin cos atan2 "" (stringify)
627 And these functions are now exported:
631 tan cotan asin acos atan acotan
632 sinh cosh tanh cotanh asinh acosh atanh acotanh
637 There have been quite a few changes made to DB_File. Here are a few of
644 Fixed a handful of bugs.
648 By public demand, added support for the standard hash function exists().
652 Made it compatible with Berkeley DB 1.86.
656 Made negative subscripts work with RECNO interface.
660 Changed the default flags from O_RDWR to O_CREAT|O_RDWR and the default
661 mode from 0640 to 0666.
665 Made DB_File automatically import the open() constants (O_RDWR,
666 O_CREAT etc.) from Fcntl, if available.
670 Updated documentation.
674 Refer to the HISTORY section in DB_File.pm for a complete list of
675 changes. Everything after DB_File 1.01 has been added since 5.003.
679 Major rewrite - support added for both udp echo and real icmp pings.
681 =head2 Overridden Built-ins
683 Many of the Perl built-ins returning lists now have
684 object-oriented overrides. These are:
696 For example, you can now say
700 $his = (stat($filename)->st_uid == pwent($whoever)->pw_uid);
702 =head1 Utility Changes
708 =item C<void> XSUBs now default to returning nothing
710 Due to a documentation/implementation bug in previous versions of
711 Perl, XSUBs with a return type of C<void> have actually been
712 returning one value. Usually that value was the GV for the XSUB,
713 but sometimes it was some already freed or reused value, which would
714 sometimes lead to program failure.
716 In Perl 5.004, if an XSUB is declared as returning C<void>, it
717 actually returns no value, i.e. an empty list (though there is a
718 backward-compatibility exception; see below). If your XSUB really
719 does return an SV, you should give it a return type of C<SV *>.
721 For backward compatibility, I<xsubpp> tries to guess whether a
722 C<void> XSUB is really C<void> or if it wants to return an C<SV *>.
723 It does so by examining the text of the XSUB: if I<xsubpp> finds
724 what looks like an assignment to C<ST(0)>, it assumes that the
725 XSUB's return type is really C<SV *>.
729 =head1 C Language API Changes
733 =item C<gv_fetchmethod> and C<perl_call_sv>
735 The C<gv_fetchmethod> function finds a method for an object, just like
736 in Perl 5.003. The GV it returns may be a method cache entry.
737 However, in Perl 5.004, method cache entries are not visible to users;
738 therefore, they can no longer be passed directly to C<perl_call_sv>.
739 Instead, you should use the C<GvCV> macro on the GV to extract its CV,
740 and pass the CV to C<perl_call_sv>.
742 The most likely symptom of passing the result of C<gv_fetchmethod> to
743 C<perl_call_sv> is Perl's producing an "Undefined subroutine called"
744 error on the I<second> call to a given method (since there is no cache
747 =item Extended API for manipulating hashes
749 Internal handling of hash keys has changed. The old hashtable API is
750 still fully supported, and will likely remain so. The additions to the
751 API allow passing keys as C<SV*>s, so that C<tied> hashes can be given
752 real scalars as keys rather than plain strings (non-tied hashes still
753 can only use strings as keys). New extensions must use the new hash
754 access functions and macros if they wish to use C<SV*> keys. These
755 additions also make it feasible to manipulate C<HE*>s (hash entries),
756 which can be more efficient. See L<perlguts> for details.
760 =head1 Documentation Changes
762 Many of the base and library pods were updated. These
763 new pods are included in section 1:
773 Locale support (internationalization and localization).
777 Tutorial on Perl OO programming.
781 Perl internal IO abstraction interface.
785 Although not new, this has been massively updated.
789 Although not new, this has been massively updated.
793 =head1 New Diagnostics
795 Several new conditions will trigger warnings that were
796 silent before. Some only affect certain platforms.
797 The following new warnings and errors outline these.
798 These messages are classified as follows (listed in
799 increasing order of desperation):
801 (W) A warning (optional).
802 (D) A deprecation (optional).
803 (S) A severe warning (mandatory).
804 (F) A fatal error (trappable).
805 (P) An internal error you should never see (trappable).
806 (X) A very fatal error (non-trappable).
807 (A) An alien error message (not generated by Perl).
811 =item "my" variable %s masks earlier declaration in same scope
813 (S) A lexical variable has been redeclared in the same scope, effectively
814 eliminating all access to the previous instance. This is almost always
815 a typographical error. Note that the earlier variable will still exist
816 until the end of the scope or until all closure referents to it are
819 =item %s argument is not a HASH element or slice
821 (F) The argument to delete() must be either a hash element, such as
824 $ref->[12]->{"susie"}
826 or a hash slice, such as
828 @foo{$bar, $baz, $xyzzy}
829 @{$ref->[12]}{"susie", "queue"}
831 =item Allocation too large: %lx
833 (X) You can't allocate more than 64K on an MSDOS machine.
835 =item Allocation too large
837 (F) You can't allocate more than 2^31+"small amount" bytes.
839 =item Attempt to free non-existent shared string
841 (P) Perl maintains a reference counted internal table of strings to
842 optimize the storage and access of hash keys and other strings. This
843 indicates someone tried to decrement the reference count of a string
844 that can no longer be found in the table.
846 =item Attempt to use reference as lvalue in substr
848 (W) You supplied a reference as the first argument to substr() used
849 as an lvalue, which is pretty strange. Perhaps you forgot to
850 dereference it first. See L<perlfunc/substr>.
852 =item Unsupported function fork
854 (F) Your version of executable does not support forking.
856 Note that under some systems, like OS/2, there may be different flavors of
857 Perl executables, some of which may support fork, some not. Try changing
858 the name you call Perl by to C<perl_>, C<perl__>, and so on.
860 =item Ill-formed logical name |%s| in prime_env_iter
862 (W) A warning peculiar to VMS. A logical name was encountered when preparing
863 to iterate over %ENV which violates the syntactic rules governing logical
864 names. Since it cannot be translated normally, it is skipped, and will not
865 appear in %ENV. This may be a benign occurrence, as some software packages
866 might directly modify logical name tables and introduce non-standard names,
867 or it may indicate that a logical name table has been corrupted.
869 =item Can't use bareword ("%s") as %s ref while "strict refs" in use
871 (F) Only hard references are allowed by "strict refs". Symbolic references
872 are disallowed. See L<perlref>.
874 =item Constant subroutine %s redefined
876 (S) You redefined a subroutine which had previously been eligible for
877 inlining. See L<perlsub/"Constant Functions"> for commentary and
882 (F) You passed die() an empty string (the equivalent of C<die "">) or
883 you called it with no args and both C<$@> and C<$_> were empty.
885 =item Integer overflow in hex number
887 (S) The literal hex number you have specified is too big for your
888 architecture. On a 32-bit architecture the largest hex literal is
891 =item Integer overflow in octal number
893 (S) The literal octal number you have specified is too big for your
894 architecture. On a 32-bit architecture the largest octal literal is
897 =item Name "%s::%s" used only once: possible typo
899 (W) Typographical errors often show up as unique variable names.
900 If you had a good reason for having a unique name, then just mention
901 it again somehow to suppress the message (the C<use vars> pragma is
902 provided for just this purpose).
904 =item Null picture in formline
906 (F) The first argument to formline must be a valid format picture
907 specification. It was found to be empty, which probably means you
908 supplied it an uninitialized value. See L<perlform>.
910 =item Offset outside string
912 (F) You tried to do a read/write/send/recv operation with an offset
913 pointing outside the buffer. This is difficult to imagine.
914 The sole exception to this is that C<sysread()>ing past the buffer
915 will extend the buffer and zero pad the new area.
917 =item Stub found while resolving method `%s' overloading `%s' in package `%s'
919 (P) Overloading resolution over @ISA tree may be broken by importing stubs.
920 Stubs should never be implicitely created, but explicit calls to C<can>
923 =item Cannot resolve method `%s' overloading `%s' in package `s'
925 (P) Internal error trying to resolve overloading specified by a method
926 name (as opposed to a subroutine reference).
930 (X|F) The malloc() function returned 0, indicating there was insufficient
931 remaining memory (or virtual memory) to satisfy the request.
933 The request was judged to be small, so the possibility to trap it
934 depends on the way Perl was compiled. By default it is not trappable.
935 However, if compiled for this, Perl may use the contents of C<$^M> as
936 an emergency pool after die()ing with this message. In this case the
937 error is trappable I<once>.
939 =item Out of memory during request for %s
941 (F) The malloc() function returned 0, indicating there was insufficient
942 remaining memory (or virtual memory) to satisfy the request. However,
943 the request was judged large enough (compile-time default is 64K), so
944 a possibility to shut down by trapping this error is granted.
946 =item Possible attempt to put comments in qw() list
948 (W) qw() lists contain items separated by whitespace; as with literal
949 strings, comment characters are not ignored, but are instead treated
950 as literal data. (You may have used different delimiters than the
951 exclamation marks parentheses shown here; braces are also frequently
954 You probably wrote something like this:
961 when you should have written this:
968 If you really want comments, build your list the
969 old-fashioned way, with quotes and commas:
973 'b', # another comment
976 =item Possible attempt to separate words with commas
978 (W) qw() lists contain items separated by whitespace; therefore commas
979 aren't needed to separate the items. (You may have used different
980 delimiters than the parentheses shown here; braces are also frequently
983 You probably wrote something like this:
987 which puts literal commas into some of the list items. Write it without
988 commas if you don't want them to appear in your data:
992 =item Scalar value @%s{%s} better written as $%s{%s}
994 (W) You've used a hash slice (indicated by @) to select a single element of
995 a hash. Generally it's better to ask for a scalar value (indicated by $).
996 The difference is that C<$foo{&bar}> always behaves like a scalar, both when
997 assigning to it and when evaluating its argument, while C<@foo{&bar}> behaves
998 like a list when you assign to it, and provides a list context to its
999 subscript, which can do weird things if you're expecting only one subscript.
1001 =item untie attempted while %d inner references still exist
1003 (W) A copy of the object returned from C<tie> (or C<tied>) was still
1004 valid when C<untie> was called.
1006 =item Value of %s construct can be "0"; test with defined()
1008 (W) In a conditional expression, you used <HANDLE>, <*> (glob), or
1009 C<readdir> as a boolean value. Each of these constructs can return a
1010 value of "0"; that would make the conditional expression false, which
1011 is probably not what you intended. When using these constructs in
1012 conditional expressions, test their values with the C<defined> operator.
1014 =item Variable "%s" may be unavailable
1016 (W) An inner (nested) I<anonymous> subroutine is inside a I<named>
1017 subroutine, and outside that is another subroutine; and the anonymous
1018 (innermost) subroutine is referencing a lexical variable defined in
1019 the outermost subroutine. For example:
1021 sub outermost { my $a; sub middle { sub { $a } } }
1023 If the anonymous subroutine is called or referenced (directly or
1024 indirectly) from the outermost subroutine, it will share the variable
1025 as you would expect. But if the anonymous subroutine is called or
1026 referenced when the outermost subroutine is not active, it will see
1027 the value of the shared variable as it was before and during the
1028 *first* call to the outermost subroutine, which is probably not what
1031 In these circumstances, it is usually best to make the middle
1032 subroutine anonymous, using the C<sub {}> syntax. Perl has specific
1033 support for shared variables in nested anonymous subroutines; a named
1034 subroutine in between interferes with this feature.
1036 =item Variable "%s" will not stay shared
1038 (W) An inner (nested) I<named> subroutine is referencing a lexical
1039 variable defined in an outer subroutine.
1041 When the inner subroutine is called, it will probably see the value of
1042 the outer subroutine's variable as it was before and during the
1043 *first* call to the outer subroutine; in this case, after the first
1044 call to the outer subroutine is complete, the inner and outer
1045 subroutines will no longer share a common value for the variable. In
1046 other words, the variable will no longer be shared.
1048 Furthermore, if the outer subroutine is anonymous and references a
1049 lexical variable outside itself, then the outer and inner subroutines
1050 will I<never> share the given variable.
1052 This problem can usually be solved by making the inner subroutine
1053 anonymous, using the C<sub {}> syntax. When inner anonymous subs that
1054 reference variables in outer subroutines are called or referenced,
1055 they are automatically re-bound to the current values of such
1058 =item Warning: something's wrong
1060 (W) You passed warn() an empty string (the equivalent of C<warn "">) or
1061 you called it with no args and C<$_> was empty.
1063 =item Got an error from DosAllocMem
1065 (P) An error peculiar to OS/2. Most probably you're using an obsolete
1066 version of Perl, and this should not happen anyway.
1068 =item Malformed PERLLIB_PREFIX
1070 (F) An error peculiar to OS/2. PERLLIB_PREFIX should be of the form
1078 with non-empty prefix1 and prefix2. If C<prefix1> is indeed a prefix of
1079 a builtin library search path, prefix2 is substituted. The error may appear
1080 if components are not found, or are too long. See L<perlos2/"PERLLIB_PREFIX">.
1082 =item PERL_SH_DIR too long
1084 (F) An error peculiar to OS/2. PERL_SH_DIR is the directory to find the
1085 C<sh>-shell in. See L<perlos2/"PERL_SH_DIR">.
1087 =item Process terminated by SIG%s
1089 (W) This is a standard message issued by OS/2 applications, while *nix
1090 applications die in silence. It is considered a feature of the OS/2
1091 port. One can easily disable this by appropriate sighandlers, see
1092 L<perlipc/"Signals">. See L<perlos2/"Process terminated by SIGTERM/SIGINT">.
1098 If you find what you think is a bug, you might check the headers of
1099 recently posted articles in the comp.lang.perl.misc newsgroup.
1100 There may also be information at http://www.perl.com/perl/, the Perl
1103 If you believe you have an unreported bug, please run the B<perlbug>
1104 program included with your release. Make sure you trim your bug down
1105 to a tiny but sufficient test case. Your bug report, along with the
1106 output of C<perl -V>, will be sent off to <F<perlbug@perl.com>> to be
1107 analysed by the Perl porting team.
1111 The F<Changes> file for exhaustive details on what changed.
1113 The F<INSTALL> file for how to build Perl. This file has been
1114 significantly updated for 5.004, so even veteran users should
1117 The F<README> file for general stuff.
1119 The F<Copying> file for copyright information.
1123 Constructed by Tom Christiansen, grabbing material with permission
1124 from innumerable contributors, with kibitzing by more than a few Perl
1127 Last update: Sat Mar 8 19:51:26 EST 1997