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 $PERL5OPT environment variable
32 You may now put Perl options in the $PERL5OPT environment variable.
33 Unless Perl is running with taint checks, it will interpret this
34 variable as if its contents had appeared on a "#!perl" line at the
35 beginning of your script, except that hyphens are optional. PERL5OPT
36 may only be used to set the following switches: B<-[DIMUdmw]>.
38 =head2 More precise warnings
40 If you removed the B<-w> option from your Perl 5.003 scripts because it
41 made Perl too verbose, we recommend that you try putting it back when
42 you upgrade to Perl 5.004. Each new perl version tends to remove some
43 undesirable warnings, while adding new warnings that may catch bugs in
46 =head2 Deprecated: Inherited C<AUTOLOAD> for non-methods
48 Before Perl 5.004, C<AUTOLOAD> functions were looked up as methods
49 (using the C<@ISA> hierarchy), even when the function to be autoloaded
50 was called as a plain function (e.g. C<Foo::bar()>), not a method
51 (e.g. C<Foo->bar()> or C<$obj->bar()>).
53 Perl 5.005 will use method lookup only for methods' C<AUTOLOAD>s.
54 However, there is a significant base of existing code that may be using
55 the old behavior. So, as an interim step, Perl 5.004 issues an optional
56 warning when a non-method uses an inherited C<AUTOLOAD>.
58 The simple rule is: Inheritance will not work when autoloading
59 non-methods. The simple fix for old code is: In any module that used to
60 depend on inheriting C<AUTOLOAD> for non-methods from a base class named
61 C<BaseClass>, execute C<*AUTOLOAD = \&BaseClass::AUTOLOAD> during startup.
63 =head2 Subroutine arguments created only when they're modified
65 In Perl 5.004, nonexistent array and hash elements used as subroutine
66 parameters are brought into existence only if they are actually
67 assigned to (via C<@_>).
69 Earlier versions of Perl vary in their handling of such arguments.
70 Perl versions 5.002 and 5.003 always brought them into existence.
71 Perl versions 5.000, 5.001, and 5.002 brought them into existence only
72 if they were not the first argument (which was almost certainly a
73 bug). Earlier versions of Perl never brought them into existence.
75 For example, given this code:
78 sub show { print $_[0] };
79 sub change { $_[0]++ };
83 After this code executes in Perl 5.004, $a{b} exists but $a[2] does
84 not. In Perl 5.002 and 5.003, both $a{b} and $a[2] would have existed
85 (but $a[2]'s value would have been undefined).
87 =head2 C<$)> is writable
89 The C<$)> special variable has always (well, in Perl 5, at least)
90 reflected not only the current effective group, but also the group
91 list as returned by the C<getgroups()> C function (if there is one).
92 However, due to an oversight, assigning to C<$)> has not called
93 C<setgroups()>, only C<setegid()>.
95 In Perl 5.004, assigning to C<$)> is exactly symmetrical with
96 examining it: The first number in its string value is used as the
97 effective gid, and all the others are passed to the C<setgroups()> C
98 function (if there is one).
100 =head2 Fixed parsing of $$<digit>, &$<digit>, etc.
102 Perl versions before 5.004 misinterpreted any type marker followed by
103 "$" and a digit. For example, "$$0" was incorrectly taken to mean
104 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
106 However, the developers of Perl 5.004 could not fix this bug completely,
107 because at least two widely-used modules depend on the old meaning of
108 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
109 old (broken) way inside strings; but it generates this message as a
110 warning. And in Perl 5.005, this special treatment will cease.
112 =head2 No resetting of $. on implicit close
114 The documentation for Perl 5.0 has always stated that C<$.> is I<not>
115 reset when an already-open file handle is reopened with no intervening
116 call to C<close>. Due to a bug, perl versions 5.000 through 5.003
117 I<did> reset C<$.> under that circumstance; Perl 5.004 does not.
119 =head2 C<wantarray> may return undef
121 The C<wantarray> operator returns true if a subroutine is expected to
122 return a list, and false otherwise. In Perl 5.004, C<wantarray> can
123 also return the undefined value if a subroutine's return value will
124 not be used at all, which allows subroutines to avoid a time-consuming
125 calculation of a return value if it isn't going to be used.
127 =head2 Changes to tainting checks
129 A bug in previous versions may have failed to detect some insecure
130 conditions when taint checks are turned on. (Taint checks are used
131 in setuid or setgid scripts, or when explicitly turned on with the
132 C<-T> invocation option.) Although it's unlikely, this may cause a
133 previously-working script to now fail -- which should be construed
134 as a blessing, since that indicates a potentially-serious security
135 hole was just plugged.
137 =head2 New Opcode module and revised Safe module
139 A new Opcode module supports the creation, manipulation and
140 application of opcode masks. The revised Safe module has a new API
141 and is implemented using the new Opcode module. Please read the new
142 Opcode and Safe documentation.
144 =head2 Embedding improvements
146 In older versions of Perl it was not possible to create more than one
147 Perl interpreter instance inside a single process without leaking like a
148 sieve and/or crashing. The bugs that caused this behavior have all been
149 fixed. However, you still must take care when embedding Perl in a C
150 program. See the updated perlembed manpage for tips on how to manage
153 =head2 Internal change: FileHandle class based on IO::* classes
155 File handles are now stored internally as type IO::Handle. The
156 FileHandle module is still supported for backwards compatibility, but
157 it is now merely a front end to the IO::* modules -- specifically,
158 IO::Handle, IO::Seekable, and IO::File. We suggest, but do not
159 require, that you use the IO::* modules in new code.
161 In harmony with this change, C<*GLOB{FILEHANDLE}> is now a
162 backward-compatible synonym for C<*STDOUT{IO}>.
164 =head2 Internal change: PerlIO abstraction interface
166 It is now possible to build Perl with AT&T's sfio IO package
167 instead of stdio. See L<perlapio> for more details, and
168 the F<INSTALL> file for how to use it.
170 =head2 New and changed builtin variables
176 Extended error message on some platforms. (Also known as
177 $EXTENDED_OS_ERROR if you C<use English>).
181 The current set of syntax checks enabled by C<use strict>. See the
182 documentation of C<strict> for more details. Not actually new, but
184 Because it is intended for internal use by Perl core components,
185 there is no C<use English> long name for this variable.
189 By default, running out of memory it is not trappable. However, if
190 compiled for this, Perl may use the contents of C<$^M> as an emergency
191 pool after die()ing with this message. Suppose that your Perl were
192 compiled with -DEMERGENCY_SBRK and used Perl's malloc. Then
196 would allocate a 64K buffer for use when in emergency.
197 See the F<INSTALL> file for information on how to enable this option.
198 As a disincentive to casual use of this advanced feature,
199 there is no C<use English> long name for this variable.
203 =head2 New and changed builtin functions
207 =item delete on slices
209 This now works. (e.g. C<delete @ENV{'PATH', 'MANPATH'}>)
213 is now supported on more platforms, prefers fcntl to lockf when
214 emulating, and always flushes before (un)locking.
216 =item printf and sprintf
218 now support "%i" as a synonym for "%d", and the "h" modifier.
219 So "%hi" means "short integer in decimal", and "%ho" means
220 "unsigned short integer as octal".
222 =item keys as an lvalue
224 As an lvalue, C<keys> allows you to increase the number of hash buckets
225 allocated for the given hash. This can gain you a measure of efficiency if
226 you know the hash is going to get big. (This is similar to pre-extending
227 an array by assigning a larger number to $#array.) If you say
231 then C<%hash> will have at least 200 buckets allocated for it. These
232 buckets will be retained even if you do C<%hash = ()>; use C<undef
233 %hash> if you want to free the storage while C<%hash> is still in scope.
234 You can't shrink the number of buckets allocated for the hash using
235 C<keys> in this way (but you needn't worry about doing this by accident,
236 as trying has no effect).
238 =item my() in Control Structures
240 You can now use my() (with or without the parentheses) in the control
241 expressions of control structures such as:
243 while (defined(my $line = <>)) {
249 if ((my $answer = <STDIN>) =~ /^y(es)?$/i) {
251 } elsif ($answer =~ /^n(o)?$/i) {
255 die "`$answer' is neither `yes' nor `no'";
258 Also, you can declare a foreach loop control variable as lexical by
259 preceding it with the word "my". For example, in:
261 foreach my $i (1, 2, 3) {
265 $i is a lexical variable, and the scope of $i extends to the end of
266 the loop, but not beyond it.
268 Note that you still cannot use my() on global punctuation variables
269 such as $_ and the like.
271 =item unpack() and pack()
273 A new format 'w' represents a BER compressed integer (as defined in
274 ASN.1). Its format is a sequence of one or more bytes, each of which
275 provides seven bits of the total value, with the most significant
276 first. Bit eight of each byte is set, except for the last byte, in
277 which bit eight is clear.
281 If the first argument to C<use> is a number, it is treated as a version
282 number instead of a module name. If the version of the Perl interpreter
283 is less than VERSION, then an error message is printed and Perl exits
284 immediately. Because C<use> occurs at compile time, this check happens
285 immediately during the compilation process, unlike C<require VERSION>,
286 which waits until runtime for the check. This is often useful if you
287 need to check the current Perl version before C<use>ing library modules
288 which have changed in incompatible ways from older versions of Perl.
289 (We try not to do this more than we have to.)
291 =item use Module VERSION LIST
293 If the VERSION argument is present between Module and LIST, then the
294 C<use> will call the VERSION method in class Module with the given
295 version as an argument. The default VERSION method, inherited from
296 the UNIVERSAL class, croaks if the given version is larger than the
297 value of the variable $Module::VERSION. (Note that there is not a
298 comma after VERSION!)
300 This version-checking mechanism is similar to the one currently used
301 in the Exporter module, but it is faster and can be used with modules
302 that don't use the Exporter. It is the recommended method for new
305 =item prototype(FUNCTION)
307 Returns the prototype of a function as a string (or C<undef> if the
308 function has no prototype). FUNCTION is a reference to or the name of the
309 function whose prototype you want to retrieve.
310 (Not actually new; just never documented before.)
314 The default seed for C<srand>, which used to be C<time>, has been changed.
315 Now it's a heady mix of difficult-to-predict system-dependent values,
316 which should be sufficient for most everyday purposes.
318 Previous to version 5.004, calling C<rand> without first calling C<srand>
319 would yield the same sequence of random numbers on most or all machines.
320 Now, when perl sees that you're calling C<rand> and haven't yet called
321 C<srand>, it calls C<srand> with the default seed. You should still call
322 C<srand> manually if your code might ever be run on a pre-5.004 system,
323 of course, or if you want a seed other than the default.
327 Functions documented in the Camel to default to $_ now in
328 fact do, and all those that do are so documented in L<perlfunc>.
330 =item C<m//g> does not trigger a pos() reset on failure
332 The C<m//g> match iteration construct used to reset the iteration
333 when it failed to match (so that the next C<m//g> match would start at
334 the beginning of the string). You now have to explicitly do a
335 C<pos $str = 0;> to reset the "last match" position, or modify the
336 string in some way. This change makes it practical to chain C<m//g>
337 matches together in conjunction with ordinary matches using the C<\G>
338 zero-width assertion. See L<perlop> and L<perlre>.
340 =item C<m//x> ignores whitespace before ?*+{}
342 The C<m//x> construct has always been intended to ignore all unescaped
343 whitespace. However, before Perl 5.004, whitespace had the effect of
344 escaping repeat modifiers like "*" or "?"; for example, C</a *b/x> was
345 (mis)interpreted as C</a\*b/x>. This bug has been fixed in 5.004.
347 =item nested C<sub{}> closures work now
349 Prior to the 5.004 release, nested anonymous functions didn't work
352 =item formats work right on changing lexicals
354 Just like anonymous functions that contain lexical variables
355 that change (like a lexical index variable for a C<foreach> loop),
356 formats now work properly. For example, this silently failed
357 before, and is fine now:
360 foreach $i ( 1 .. 10 ) {
370 =head2 New builtin methods
372 The C<UNIVERSAL> package automatically contains the following methods that
373 are inherited by all other classes:
379 C<isa> returns I<true> if its object is blessed into a subclass of C<CLASS>
381 C<isa> is also exportable and can be called as a sub with two arguments. This
382 allows the ability to check what a reference points to. Example:
384 use UNIVERSAL qw(isa);
386 if(isa($ref, 'ARRAY')) {
392 C<can> checks to see if its object has a method called C<METHOD>,
393 if it does then a reference to the sub is returned; if it does not then
394 I<undef> is returned.
396 =item VERSION( [NEED] )
398 C<VERSION> returns the version number of the class (package). If the
399 NEED argument is given then it will check that the current version (as
400 defined by the $VERSION variable in the given package) not less than
401 NEED; it will die if this is not the case. This method is normally
402 called as a class method. This method is called automatically by the
403 C<VERSION> form of C<use>.
405 use A 1.2 qw(some imported subs);
411 B<NOTE:> C<can> directly uses Perl's internal code for method lookup, and
412 C<isa> uses a very similar method and caching strategy. This may cause
413 strange effects if the Perl code dynamically changes @ISA in any package.
415 You may add other methods to the UNIVERSAL class via Perl or XS code.
416 You do not need to C<use UNIVERSAL> in order to make these methods
417 available to your program. This is necessary only if you wish to
418 have C<isa> available as a plain subroutine in the current package.
420 =head2 TIEHANDLE now supported
422 See L<perltie> for other kinds of tie()s.
426 =item TIEHANDLE classname, LIST
428 This is the constructor for the class. That means it is expected to
429 return an object of some sort. The reference can be used to
430 hold some internal information.
435 return bless \$i, shift;
438 =item PRINT this, LIST
440 This method will be triggered every time the tied handle is printed to.
441 Beyond its self reference it also expects the list that was passed to
447 return print join( $, => map {uc} @_), $\;
452 This method will be called when the handle is read from via the C<read>
453 or C<sysread> functions.
457 my($buf,$len,$offset) = @_;
458 print "READ called, \$buf=$buf, \$len=$len, \$offset=$offset";
463 This method will be called when the handle is read from. The method
464 should return undef when there is no more data.
468 return "PRINT called $$r times\n"
473 This method will be called when the C<getc> function is called.
475 sub GETC { print "Don't GETC, Get Perl"; return "a"; }
479 As with the other types of ties, this method will be called when the
480 tied handle is about to be destroyed. This is useful for debugging and
481 possibly for cleaning up.
489 =head2 Malloc enhancements
491 Four new compilation flags are recognized by malloc.c. (They have no
492 effect if perl is compiled with system malloc().)
496 =item -DDEBUGGING_MSTATS
498 If perl is compiled with C<DEBUGGING_MSTATS> defined, you can print
499 memory statistics at runtime by running Perl thusly:
501 env PERL_DEBUG_MSTATS=2 perl your_script_here
503 The value of 2 means to print statistics after compilation and on
504 exit; with a value of 1, the statistics ares printed only on exit.
505 (If you want the statistics at an arbitrary time, you'll need to
506 install the optional module Devel::Peek.)
508 =item -DEMERGENCY_SBRK
510 If this macro is defined, running out of memory need not be a fatal
511 error: a memory pool can allocated by assigning to the special
512 variable C<$^M>. See L<"$^M">.
516 Perl memory allocation is by bucket with sizes close to powers of two.
517 Because of these malloc overhead may be big, especially for data of
518 size exactly a power of two. If C<PACK_MALLOC> is defined, perl uses
519 a slightly different algorithm for small allocations (up to 64 bytes
520 long), which makes it possible to have overhead down to 1 byte for
521 allocations which are powers of two (and appear quite often).
523 Expected memory savings (with 8-byte alignment in C<alignbytes>) is
524 about 20% for typical Perl usage. Expected slowdown due to additional
525 malloc overhead is in fractions of a percent (hard to measure, because
526 of the effect of saved memory on speed).
528 =item -DTWO_POT_OPTIMIZE
530 Similarly to C<PACK_MALLOC>, this macro improves allocations of data
531 with size close to a power of two; but this works for big allocations
532 (starting with 16K by default). Such allocations are typical for big
533 hashes and special-purpose scripts, especially image processing.
535 On recent systems, the fact that perl requires 2M from system for 1M
536 allocation will not affect speed of execution, since the tail of such
537 a chunk is not going to be touched (and thus will not require real
538 memory). However, it may result in a premature out-of-memory error.
539 So if you will be manipulating very large blocks with sizes close to
540 powers of two, it would be wise to define this macro.
542 Expected saving of memory is 0-100% (100% in applications which
543 require most memory in such 2**n chunks); expected slowdown is
548 =head2 Miscellaneous efficiency enhancements
550 Functions that have an empty prototype and that do nothing but return
551 a fixed value are now inlined (e.g. C<sub PI () { 3.14159 }>).
553 Each unique hash key is only allocated once, no matter how many hashes
554 have an entry with that key. So even if you have 100 copies of the
555 same hash, the hash keys never have to be reallocated.
559 Six new pragmatic modules exist:
563 =item use autouse MODULE => qw(sub1 sub2 sub3)
565 Defers C<require MODULE> until someone calls one of the specified
566 subroutines (which must be exported by MODULE). This pragma should be
567 used with caution, and only when necessary.
573 Looks for MakeMaker-like I<'blib'> directory structure starting in
574 I<dir> (or current directory) and working back up to five levels of
577 Intended for use on command line with B<-M> option as a way of testing
578 arbitrary scripts against an uninstalled version of a package.
580 =item use constant NAME => VALUE
582 Provides a convenient interface for creating compile-time constants,
583 See L<perlsub/"Constant Functions">.
587 Tells the compiler to enable (or disable) the use of POSIX locales for
590 When C<use locale> is in effect, the current LC_CTYPE locale is used
591 for regular expressions and case mapping; LC_COLLATE for string
592 ordering; and LC_NUMERIC for numeric formating in printf and sprintf
593 (but B<not> in print). LC_NUMERIC is always used in write, since
594 lexical scoping of formats is problematic at best.
596 Each C<use locale> or C<no locale> affects statements to the end of
597 the enclosing BLOCK or, if not inside a BLOCK, to the end of the
598 current file. Locales can be switched and queried with
601 See L<perllocale> for more information.
605 Disable unsafe opcodes, or any named opcodes, when compiling Perl code.
609 Enable VMS-specific language features. Currently, there are three
610 VMS-specific features available: 'status', which makes C<$?> and
611 C<system> return genuine VMS status values instead of emulating POSIX;
612 'exit', which makes C<exit> take a genuine VMS status value instead of
613 assuming that C<exit 1> is an error; and 'time', which makes all times
614 relative to the local time zone, in the VMS tradition.
620 =head2 Required Updates
622 Though Perl 5.004 is compatible with almost all modules that work
623 with Perl 5.003, there are a few exceptions:
625 Module Required Version for Perl 5.004
626 ------ -------------------------------
629 Tk Tk400.202 (-w makes noise)
631 =head2 Installation directories
633 The I<installperl> script now places the Perl source files for
634 extensions in the architecture-specific library directory, which is
635 where the shared libraries for extensions have always been. This
636 change is intended to allow administrators to keep the Perl 5.004
637 library directory unchanged from a previous version, without running
638 the risk of binary incompatibility between extensions' Perl source and
641 =head2 Module information summary
643 Brand new modules, arranged by topic rather than strictly
646 CPAN interface to Comprehensive Perl Archive Network
647 CPAN::FirstTime create a CPAN configuration file
648 CPAN::Nox run CPAN while avoiding compiled extensions
650 IO.pm Top-level interface to IO::* classes
651 IO/File.pm IO::File extension Perl module
652 IO/Handle.pm IO::Handle extension Perl module
653 IO/Pipe.pm IO::Pipe extension Perl module
654 IO/Seekable.pm IO::Seekable extension Perl module
655 IO/Select.pm IO::Select extension Perl module
656 IO/Socket.pm IO::Socket extension Perl module
658 Opcode.pm Disable named opcodes when compiling Perl code
660 ExtUtils/Embed.pm Utilities for embedding Perl in C programs
661 ExtUtils/testlib.pm Fixes up @INC to use just-built extension
663 FindBin.pm Find path of currently executing program
665 Class/Template.pm Structure/member template builder
666 File/stat.pm Object-oriented wrapper around CORE::stat
667 Net/hostent.pm Object-oriented wrapper around CORE::gethost*
668 Net/netent.pm Object-oriented wrapper around CORE::getnet*
669 Net/protoent.pm Object-oriented wrapper around CORE::getproto*
670 Net/servent.pm Object-oriented wrapper around CORE::getserv*
671 Time/gmtime.pm Object-oriented wrapper around CORE::gmtime
672 Time/localtime.pm Object-oriented wrapper around CORE::localtime
673 Time/tm.pm Perl implementation of "struct tm" for {gm,local}time
674 User/grent.pm Object-oriented wrapper around CORE::getgr*
675 User/pwent.pm Object-oriented wrapper around CORE::getpw*
677 Tie/RefHash.pm Base class for tied hashes with references as keys
679 UNIVERSAL.pm Base class for *ALL* classes
683 New constants in the existing Fcntl modules are now supported,
684 provided that your operating system happens to support them:
687 O_ASYNC O_DEFER O_DSYNC O_FSYNC O_SYNC
690 These constants are intended for use with the Perl operators sysopen()
691 and fcntl() and the basic database modules like SDBM_File. For the
692 exact meaning of these and other Fcntl constants please refer to your
693 operating system's documentation for fcntl() and open().
695 In addition, the Fcntl module now provides these constants for use
696 with the Perl operator flock():
698 LOCK_SH LOCK_EX LOCK_NB LOCK_UN
700 These constants are defined in all environments (because where there is
701 no flock() system call, Perl emulates it). However, for historical
702 reasons, these constants are not exported unless they are explicitly
703 requested with the ":flock" tag (e.g. C<use Fcntl ':flock'>).
707 The IO module provides a simple mechanism to load all of the IO modules at one
708 go. Currently this includes:
716 For more information on any of these modules, please see its
717 respective documentation.
721 The Math::Complex module has been totally rewritten, and now supports
722 more operations. These are overloaded:
724 + - * / ** <=> neg ~ abs sqrt exp log sin cos atan2 "" (stringify)
726 And these functions are now exported:
729 log10 logn ln cbrt root
742 This new module provides a simpler interface to parts of Math::Complex for
743 those who need trigonometric functions only for real numbers.
747 There have been quite a few changes made to DB_File. Here are a few of
754 Fixed a handful of bugs.
758 By public demand, added support for the standard hash function exists().
762 Made it compatible with Berkeley DB 1.86.
766 Made negative subscripts work with RECNO interface.
770 Changed the default flags from O_RDWR to O_CREAT|O_RDWR and the default
771 mode from 0640 to 0666.
775 Made DB_File automatically import the open() constants (O_RDWR,
776 O_CREAT etc.) from Fcntl, if available.
780 Updated documentation.
784 Refer to the HISTORY section in DB_File.pm for a complete list of
785 changes. Everything after DB_File 1.01 has been added since 5.003.
789 Major rewrite - support added for both udp echo and real icmp pings.
791 =head2 Object-oriented overrides for builtin operators
793 Many of the Perl builtins returning lists now have
794 object-oriented overrides. These are:
806 For example, you can now say
810 $his = (stat($filename)->st_uid == pwent($whoever)->pw_uid);
812 =head1 Utility Changes
818 =item C<void> XSUBs now default to returning nothing
820 Due to a documentation/implementation bug in previous versions of
821 Perl, XSUBs with a return type of C<void> have actually been
822 returning one value. Usually that value was the GV for the XSUB,
823 but sometimes it was some already freed or reused value, which would
824 sometimes lead to program failure.
826 In Perl 5.004, if an XSUB is declared as returning C<void>, it
827 actually returns no value, i.e. an empty list (though there is a
828 backward-compatibility exception; see below). If your XSUB really
829 does return an SV, you should give it a return type of C<SV *>.
831 For backward compatibility, I<xsubpp> tries to guess whether a
832 C<void> XSUB is really C<void> or if it wants to return an C<SV *>.
833 It does so by examining the text of the XSUB: if I<xsubpp> finds
834 what looks like an assignment to C<ST(0)>, it assumes that the
835 XSUB's return type is really C<SV *>.
839 =head1 C Language API Changes
843 =item C<gv_fetchmethod> and C<perl_call_sv>
845 The C<gv_fetchmethod> function finds a method for an object, just like
846 in Perl 5.003. The GV it returns may be a method cache entry.
847 However, in Perl 5.004, method cache entries are not visible to users;
848 therefore, they can no longer be passed directly to C<perl_call_sv>.
849 Instead, you should use the C<GvCV> macro on the GV to extract its CV,
850 and pass the CV to C<perl_call_sv>.
852 The most likely symptom of passing the result of C<gv_fetchmethod> to
853 C<perl_call_sv> is Perl's producing an "Undefined subroutine called"
854 error on the I<second> call to a given method (since there is no cache
857 =item Extended API for manipulating hashes
859 Internal handling of hash keys has changed. The old hashtable API is
860 still fully supported, and will likely remain so. The additions to the
861 API allow passing keys as C<SV*>s, so that C<tied> hashes can be given
862 real scalars as keys rather than plain strings (nontied hashes still
863 can only use strings as keys). New extensions must use the new hash
864 access functions and macros if they wish to use C<SV*> keys. These
865 additions also make it feasible to manipulate C<HE*>s (hash entries),
866 which can be more efficient. See L<perlguts> for details.
870 =head1 Documentation Changes
872 Many of the base and library pods were updated. These
873 new pods are included in section 1:
883 Locale support (internationalization and localization).
887 Tutorial on Perl OO programming.
891 Perl internal IO abstraction interface.
895 Although not new, this has been massively updated.
899 Although not new, this has been massively updated.
903 =head1 New Diagnostics
905 Several new conditions will trigger warnings that were
906 silent before. Some only affect certain platforms.
907 The following new warnings and errors outline these.
908 These messages are classified as follows (listed in
909 increasing order of desperation):
911 (W) A warning (optional).
912 (D) A deprecation (optional).
913 (S) A severe warning (mandatory).
914 (F) A fatal error (trappable).
915 (P) An internal error you should never see (trappable).
916 (X) A very fatal error (nontrappable).
917 (A) An alien error message (not generated by Perl).
921 =item "my" variable %s masks earlier declaration in same scope
923 (S) A lexical variable has been redeclared in the same scope, effectively
924 eliminating all access to the previous instance. This is almost always
925 a typographical error. Note that the earlier variable will still exist
926 until the end of the scope or until all closure referents to it are
929 =item %s argument is not a HASH element or slice
931 (F) The argument to delete() must be either a hash element, such as
934 $ref->[12]->{"susie"}
936 or a hash slice, such as
938 @foo{$bar, $baz, $xyzzy}
939 @{$ref->[12]}{"susie", "queue"}
941 =item Allocation too large: %lx
943 (X) You can't allocate more than 64K on an MS-DOS machine.
945 =item Allocation too large
947 (F) You can't allocate more than 2^31+"small amount" bytes.
949 =item Applying %s to %s will act on scalar(%s)
951 (W) The pattern match (//), substitution (s///), and translation (tr///)
952 operators work on scalar values. If you apply one of them to an array
953 or a hash, it will convert the array or hash to a scalar value -- the
954 length of an array, or the population info of a hash -- and then work on
955 that scalar value. This is probably not what you meant to do. See
956 L<perlfunc/grep> and L<perlfunc/map> for alternatives.
958 =item Attempt to free nonexistent shared string
960 (P) Perl maintains a reference counted internal table of strings to
961 optimize the storage and access of hash keys and other strings. This
962 indicates someone tried to decrement the reference count of a string
963 that can no longer be found in the table.
965 =item Attempt to use reference as lvalue in substr
967 (W) You supplied a reference as the first argument to substr() used
968 as an lvalue, which is pretty strange. Perhaps you forgot to
969 dereference it first. See L<perlfunc/substr>.
971 =item Can't use bareword ("%s") as %s ref while "strict refs" in use
973 (F) Only hard references are allowed by "strict refs". Symbolic references
974 are disallowed. See L<perlref>.
976 =item Cannot resolve method `%s' overloading `%s' in package `%s'
978 (P) Internal error trying to resolve overloading specified by a method
979 name (as opposed to a subroutine reference).
981 =item Constant subroutine %s redefined
983 (S) You redefined a subroutine which had previously been eligible for
984 inlining. See L<perlsub/"Constant Functions"> for commentary and
987 =item Constant subroutine %s undefined
989 (S) You undefined a subroutine which had previously been eligible for
990 inlining. See L<perlsub/"Constant Functions"> for commentary and
993 =item Copy method did not return a reference
995 (F) The method which overloads "=" is buggy. See L<overload/Copy Constructor>.
999 (F) You passed die() an empty string (the equivalent of C<die "">) or
1000 you called it with no args and both C<$@> and C<$_> were empty.
1002 =item Exiting pseudo-block via %s
1004 (W) You are exiting a rather special block construct (like a sort block or
1005 subroutine) by unconventional means, such as a goto, or a loop control
1006 statement. See L<perlfunc/sort>.
1008 =item Illegal character %s (carriage return)
1010 (F) A carriage return character was found in the input. This is an
1011 error, and not a warning, because carriage return characters can break
1012 multi-line strings, including here documents (e.g., C<print E<lt>E<lt>EOF;>).
1014 =item Illegal switch in PERL5OPT: %s
1016 (X) The PERL5OPT environment variable may only be used to set the
1017 following switches: B<-[DIMUdmw]>.
1019 =item Integer overflow in hex number
1021 (S) The literal hex number you have specified is too big for your
1022 architecture. On a 32-bit architecture the largest hex literal is
1025 =item Integer overflow in octal number
1027 (S) The literal octal number you have specified is too big for your
1028 architecture. On a 32-bit architecture the largest octal literal is
1031 =item internal error: glob failed
1033 (P) Something went wrong with the external program(s) used for C<glob>
1034 and C<E<lt>*.cE<gt>>. This may mean that your csh (C shell) is
1035 broken. If so, you should change all of the csh-related variables in
1036 config.sh: If you have tcsh, make the variables refer to it as if it
1037 were csh (e.g. C<full_csh='/usr/bin/tcsh'>); otherwise, make them all
1038 empty (except that C<d_csh> should be C<'undef'>) so that Perl will
1039 think csh is missing. In either case, after editing config.sh, run
1040 C<./Configure -S> and rebuild Perl.
1042 =item Name "%s::%s" used only once: possible typo
1044 (W) Typographical errors often show up as unique variable names.
1045 If you had a good reason for having a unique name, then just mention
1046 it again somehow to suppress the message (the C<use vars> pragma is
1047 provided for just this purpose).
1049 =item Null picture in formline
1051 (F) The first argument to formline must be a valid format picture
1052 specification. It was found to be empty, which probably means you
1053 supplied it an uninitialized value. See L<perlform>.
1055 =item Offset outside string
1057 (F) You tried to do a read/write/send/recv operation with an offset
1058 pointing outside the buffer. This is difficult to imagine.
1059 The sole exception to this is that C<sysread()>ing past the buffer
1060 will extend the buffer and zero pad the new area.
1062 =item Out of memory!
1064 (X|F) The malloc() function returned 0, indicating there was insufficient
1065 remaining memory (or virtual memory) to satisfy the request.
1067 The request was judged to be small, so the possibility to trap it
1068 depends on the way Perl was compiled. By default it is not trappable.
1069 However, if compiled for this, Perl may use the contents of C<$^M> as
1070 an emergency pool after die()ing with this message. In this case the
1071 error is trappable I<once>.
1073 =item Out of memory during request for %s
1075 (F) The malloc() function returned 0, indicating there was insufficient
1076 remaining memory (or virtual memory) to satisfy the request. However,
1077 the request was judged large enough (compile-time default is 64K), so
1078 a possibility to shut down by trapping this error is granted.
1080 =item Possible attempt to put comments in qw() list
1082 (W) qw() lists contain items separated by whitespace; as with literal
1083 strings, comment characters are not ignored, but are instead treated
1084 as literal data. (You may have used different delimiters than the
1085 exclamation marks parentheses shown here; braces are also frequently
1088 You probably wrote something like this:
1095 when you should have written this:
1102 If you really want comments, build your list the
1103 old-fashioned way, with quotes and commas:
1107 'b', # another comment
1110 =item Possible attempt to separate words with commas
1112 (W) qw() lists contain items separated by whitespace; therefore commas
1113 aren't needed to separate the items. (You may have used different
1114 delimiters than the parentheses shown here; braces are also frequently
1117 You probably wrote something like this:
1121 which puts literal commas into some of the list items. Write it without
1122 commas if you don't want them to appear in your data:
1126 =item Recursive substitution detected
1128 (F) The replacement string of a substitution caused the recursive
1129 execution of that very same substituion. Perl cannot keep track of
1130 special variables (C<$1>, etc.) under such circumstances.
1132 =item Scalar value @%s{%s} better written as $%s{%s}
1134 (W) You've used a hash slice (indicated by @) to select a single element of
1135 a hash. Generally it's better to ask for a scalar value (indicated by $).
1136 The difference is that C<$foo{&bar}> always behaves like a scalar, both when
1137 assigning to it and when evaluating its argument, while C<@foo{&bar}> behaves
1138 like a list when you assign to it, and provides a list context to its
1139 subscript, which can do weird things if you're expecting only one subscript.
1141 =item Stub found while resolving method `%s' overloading `%s' in package `%s'
1143 (P) Overloading resolution over @ISA tree may be broken by importing stubs.
1144 Stubs should never be implicitely created, but explicit calls to C<can>
1147 =item Too late for "B<-T>" option
1149 (X) The #! line (or local equivalent) in a Perl script contains the
1150 B<-T> option, but Perl was not invoked with B<-T> in its argument
1151 list. This is an error because, by the time Perl discovers a B<-T> in
1152 a script, it's too late to properly taint everything from the
1153 environment. So Perl gives up.
1155 =item untie attempted while %d inner references still exist
1157 (W) A copy of the object returned from C<tie> (or C<tied>) was still
1158 valid when C<untie> was called.
1160 =item Unrecognized character %s
1162 (F) The Perl parser has no idea what to do with the specified character
1163 in your Perl script (or eval). Perhaps you tried to run a compressed
1164 script, a binary program, or a directory as a Perl program.
1166 =item Unsupported function fork
1168 (F) Your version of executable does not support forking.
1170 Note that under some systems, like OS/2, there may be different flavors of
1171 Perl executables, some of which may support fork, some not. Try changing
1172 the name you call Perl by to C<perl_>, C<perl__>, and so on.
1174 =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
1176 (D) Perl versions before 5.004 misinterpreted any type marker followed
1177 by "$" and a digit. For example, "$$0" was incorrectly taken to mean
1178 "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
1180 However, the developers of Perl 5.004 could not fix this bug completely,
1181 because at least two widely-used modules depend on the old meaning of
1182 "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
1183 old (broken) way inside strings; but it generates this message as a
1184 warning. And in Perl 5.005, this special treatment will cease.
1186 =item Value of %s can be "0"; test with defined()
1188 (W) In a conditional expression, you used <HANDLE>, <*> (glob), C<each()>,
1189 or C<readdir()> as a boolean value. Each of these constructs can return a
1190 value of "0"; that would make the conditional expression false, which is
1191 probably not what you intended. When using these constructs in conditional
1192 expressions, test their values with the C<defined> operator.
1194 =item Variable "%s" may be unavailable
1196 (W) An inner (nested) I<anonymous> subroutine is inside a I<named>
1197 subroutine, and outside that is another subroutine; and the anonymous
1198 (innermost) subroutine is referencing a lexical variable defined in
1199 the outermost subroutine. For example:
1201 sub outermost { my $a; sub middle { sub { $a } } }
1203 If the anonymous subroutine is called or referenced (directly or
1204 indirectly) from the outermost subroutine, it will share the variable
1205 as you would expect. But if the anonymous subroutine is called or
1206 referenced when the outermost subroutine is not active, it will see
1207 the value of the shared variable as it was before and during the
1208 *first* call to the outermost subroutine, which is probably not what
1211 In these circumstances, it is usually best to make the middle
1212 subroutine anonymous, using the C<sub {}> syntax. Perl has specific
1213 support for shared variables in nested anonymous subroutines; a named
1214 subroutine in between interferes with this feature.
1216 =item Variable "%s" will not stay shared
1218 (W) An inner (nested) I<named> subroutine is referencing a lexical
1219 variable defined in an outer subroutine.
1221 When the inner subroutine is called, it will probably see the value of
1222 the outer subroutine's variable as it was before and during the
1223 *first* call to the outer subroutine; in this case, after the first
1224 call to the outer subroutine is complete, the inner and outer
1225 subroutines will no longer share a common value for the variable. In
1226 other words, the variable will no longer be shared.
1228 Furthermore, if the outer subroutine is anonymous and references a
1229 lexical variable outside itself, then the outer and inner subroutines
1230 will I<never> share the given variable.
1232 This problem can usually be solved by making the inner subroutine
1233 anonymous, using the C<sub {}> syntax. When inner anonymous subs that
1234 reference variables in outer subroutines are called or referenced,
1235 they are automatically rebound to the current values of such
1238 =item Warning: something's wrong
1240 (W) You passed warn() an empty string (the equivalent of C<warn "">) or
1241 you called it with no args and C<$_> was empty.
1243 =item Ill-formed logical name |%s| in prime_env_iter
1245 (W) A warning peculiar to VMS. A logical name was encountered when preparing
1246 to iterate over %ENV which violates the syntactic rules governing logical
1247 names. Since it cannot be translated normally, it is skipped, and will not
1248 appear in %ENV. This may be a benign occurrence, as some software packages
1249 might directly modify logical name tables and introduce nonstandard names,
1250 or it may indicate that a logical name table has been corrupted.
1252 =item Got an error from DosAllocMem
1254 (P) An error peculiar to OS/2. Most probably you're using an obsolete
1255 version of Perl, and this should not happen anyway.
1257 =item Malformed PERLLIB_PREFIX
1259 (F) An error peculiar to OS/2. PERLLIB_PREFIX should be of the form
1267 with nonempty prefix1 and prefix2. If C<prefix1> is indeed a prefix
1268 of a builtin library search path, prefix2 is substituted. The error
1269 may appear if components are not found, or are too long. See
1270 "PERLLIB_PREFIX" in F<README.os2>.
1272 =item PERL_SH_DIR too long
1274 (F) An error peculiar to OS/2. PERL_SH_DIR is the directory to find the
1275 C<sh>-shell in. See "PERL_SH_DIR" in F<README.os2>.
1277 =item Process terminated by SIG%s
1279 (W) This is a standard message issued by OS/2 applications, while *nix
1280 applications die in silence. It is considered a feature of the OS/2
1281 port. One can easily disable this by appropriate sighandlers, see
1282 L<perlipc/"Signals">. See also "Process terminated by SIGTERM/SIGINT"
1289 If you find what you think is a bug, you might check the headers of
1290 recently posted articles in the comp.lang.perl.misc newsgroup.
1291 There may also be information at http://www.perl.com/perl/, the Perl
1294 If you believe you have an unreported bug, please run the B<perlbug>
1295 program included with your release. Make sure you trim your bug down
1296 to a tiny but sufficient test case. Your bug report, along with the
1297 output of C<perl -V>, will be sent off to <F<perlbug@perl.com>> to be
1298 analysed by the Perl porting team.
1302 The F<Changes> file for exhaustive details on what changed.
1304 The F<INSTALL> file for how to build Perl. This file has been
1305 significantly updated for 5.004, so even veteran users should
1308 The F<README> file for general stuff.
1310 The F<Copying> file for copyright information.
1314 Constructed by Tom Christiansen, grabbing material with permission
1315 from innumerable contributors, with kibitzing by more than a few Perl
1318 Last update: Sat Mar 8 19:51:26 EST 1997