3 perldelta - what's new for perl v5.6 (as of v5.005_61)
7 This is an unsupported alpha release, meant for intrepid Perl developers
8 only. The included sources may not even build correctly on some platforms.
9 Subscribing to perl5-porters is the best way to monitor and contribute
10 to the progress of development releases (see www.perl.org for info).
12 This document describes differences between the 5.005 release and this one.
14 =head1 Incompatible Changes
16 =head2 Perl Source Incompatibilities
22 =item Possibly changed pseudo-random number generator
24 In 5.005_0x and earlier, perl's rand() function used the C library
25 rand(3) function. As of 5.005_52, Configure tests for drand48(),
26 random(), and rand() (in that order) and picks the first one it finds.
27 Perl programs that depend on reproducing a specific set of pseudo-random
28 numbers will now likely produce different output.
32 =head2 C Source Incompatibilities
38 Release 5.005 grandfathered old global symbol names by providing preprocessor
39 macros for extension source compatibility. As of release 5.6, these
40 preprocessor definitions are not available by default. You need to explicitly
41 compile perl with C<-DPERL_POLLUTE> to get these definitions. For
42 extensions still using the old symbols, this option can be
43 specified via MakeMaker:
45 perl Makefile.PL POLLUTE=1
47 =item C<PERL_IMPLICIT_CONTEXT>
49 This new build option provides a set of macros for all API functions
50 such that an implicit interpreter/thread context argument is passed to
51 every API function. As a result of this, something like C<sv_setsv(foo,bar)>
52 amounts to a macro invocation that actually translates to something like
53 C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
54 to not have any significant source compatibility issues, the difference
55 between a macro and a real function call will need to be considered.
57 This means that there B<is> a source compatibility issue as a result of
58 this if your extensions attempt to use pointers to any of the Perl API
61 Note that the above issue is not relevant to the default build of
62 Perl, whose interfaces continue to match those of prior versions
63 (but subject to the other options described here).
65 PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built
66 with one of -Dusethreads, -Dusemultiplicity, or both.
68 See L<perlguts/"The Perl API"> for detailed information on the
69 ramifications of building Perl using this option.
71 =item C<PERL_POLLUTE_MALLOC>
73 Enabling Perl's malloc in release 5.005 and earlier caused
74 the namespace of system versions of the malloc family of functions to
75 be usurped by the Perl versions, since by default they used the
78 Besides causing problems on platforms that do not allow these functions to
79 be cleanly replaced, this also meant that the system versions could not
80 be called in programs that used Perl's malloc. Previous versions of Perl
81 have allowed this behaviour to be suppressed with the HIDEMYMALLOC and
82 EMBEDMYMALLOC preprocessor definitions.
84 As of release 5.6, Perl's malloc family of functions have default names
85 distinct from the system versions. You need to explicitly compile perl with
86 C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
87 and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
90 Note that these functions do B<not> constitute Perl's memory allocation API.
91 See L<perlguts/"Memory Allocation"> for further information about that.
93 =item C<PL_na> and C<dTHR> Issues
95 The C<PL_na> global is now thread local, so a C<dTHR> declaration is needed
96 in the scope in which the global appears. XSUBs should handle this automatically,
97 but if you have used C<PL_na> in support functions, you either need to
98 change the C<PL_na> to a local variable (which is recommended), or put in
103 =head2 Compatible C Source API Changes
107 =item C<PATCHLEVEL> is now C<PERL_VERSION>
109 The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
110 are now available by default from perl.h, and reflect the base revision,
111 patchlevel, and subversion respectively. C<PERL_REVISION> had no
112 prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
113 previously available as C<PATCHLEVEL> and C<SUBVERSION>.
115 The new names cause less pollution of the B<cpp> namespace and reflect what
116 the numbers have come to stand for in common practice. For compatibility,
117 the old names are still supported when F<patchlevel.h> is explicitly
118 included (as required before), so there is no source incompatibility
123 =head2 Binary Incompatibilities
125 The default build of this release is binary compatible with the 5.005
126 release or its maintenance versions.
128 The usethreads or usemultiplicity builds are B<not> binary compatible
129 with the corresponding builds in 5.005.
133 =head2 Unicode and UTF-8 support
135 Perl can optionally use UTF-8 as its internal representation for character
136 strings. The C<use utf8> pragma enables this support in the current lexical
137 scope. See L<utf8> for more information.
139 =head2 Lexically scoped warning categories
141 You can now control the granularity of warnings emitted by perl at a finer
142 level using the C<use warnings> pragma. See L<warnings> and L<perllexwarn>
145 =head2 Binary numbers supported
147 Binary numbers are now supported as literals, in s?printf formats, and
151 printf "The answer is: %b\n", oct("0b101010");
153 =head2 syswrite() ease-of-use
155 The length argument of C<syswrite()> is now optional.
157 =head2 64-bit support
159 All platforms that have 64-bit integers either (a) natively as longs
160 or ints (b) via special compiler flags (c) using long long are able to
161 use "quads" (64-integers) as follows:
165 =item constants (decimal, hexadecimal, octal, binary) in the code
167 =item arguments to oct() and hex()
169 =item arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)
171 =item printed as such
173 =item pack() and unpack() "q" and "Q" formats
175 =item in basic arithmetics: + - * / %
177 =item vec() (but see the below note about bit arithmetics)
181 Note that unless you have the case (a) you will have to configure
182 and compile Perl using the -Duse64bits Configure flag.
184 Unfortunately bit arithmetics (&, |, ^, ~, <<, >>) for numbers are not
185 64-bit clean, they are explictly forced to be 32-bit. Bit arithmetics
186 for bit vectors (created by vec()) are not limited in their width.
188 Last but not least: note that due to Perl's habit of always using
189 floating point numbers the quads are still not true integers.
190 When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned,
191 -9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they
192 are silently promoted to floating point numbers, after which they will
193 start losing precision (their lower digits).
195 =head2 Large file support
197 If you have filesystems that support "large files" (files larger than
198 2 gigabytes), you may now also be able to create and access them from
199 Perl. You have to use Configure -Duselargefiles. Turning on the
200 large file support turns on also the 64-bit support, for obvious reasons.
202 Note that in addition to requiring a proper file system to do large
203 files you may also need to adjust your per-process (or your
204 per-system, or per-process-group, or per-user-group) maximum filesize
205 limits before running Perl scripts that try to handle large files,
206 especially if you intend to write such files.
208 Finally, in addition to your process/process group maximum filesize
209 limits, you may have quota limits on your filesystems that stop you
210 (your user id or your user group id) from using large files.
212 Adjusting your process/user/group/file system/operating system limits
213 is outside the scope of Perl core language. For process limits, you
214 may try increasing the limits using your shell's limits/limit/ulimit
215 command before running Perl. The BSD::Resource extension (not
216 included with the standard Perl distribution) may also be of use, it
217 offers the getrlimit/setrlimit interface that can be used to adjust
218 process resource usage limits, including the maximum filesize limit.
222 In some systems you may be able to use long doubles to enhance the
223 range of precision of your double precision floating point numbers
224 (that is, Perl's numbers). Use Configure -Duselongdouble to enable
225 this support (if it is available).
229 You can Configure -Dusemorebits to turn on both the 64-bit support
230 and the long double support.
232 =head2 Better syntax checks on parenthesized unary operators
236 print defined(&foo,&bar,&baz);
237 print uc("foo","bar","baz");
240 used to be accidentally allowed in earlier versions, and produced
241 unpredictable behaviour. Some produced ancillary warnings
242 when used in this way; others silently did the wrong thing.
244 The parenthesized forms of most unary operators that expect a single
245 argument now ensure that they are not called with more than one
246 argument, making the cases shown above syntax errors. The usual
249 print defined &foo, &bar, &baz;
250 print uc "foo", "bar", "baz";
253 remains unchanged. See L<perlop>.
255 =head2 POSIX character class syntax [: :] supported
257 For example to match alphabetic characters use /[[:alpha:]]/.
258 See L<perlre> for details.
260 =head2 Improved C<qw//> operator
262 The C<qw//> operator is now evaluated at compile time into a true list
263 instead of being replaced with a run time call to C<split()>. This
264 removes the confusing misbehaviour of C<qw//> in scalar context, which
265 had inherited that behaviour from split().
269 $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
271 now correctly prints "3|a", instead of "2|a".
273 =head2 pack() format 'Z' supported
275 The new format type 'Z' is useful for packing and unpacking null-terminated
276 strings. See L<perlfunc/"pack">.
278 =head2 pack() format modifier '!' supported
280 The new format type modifier '!' is useful for packing and unpacking
281 native shorts, ints, and longs. See L<perlfunc/"pack">.
283 =head2 pack() and unpack() support counted strings
285 The template character '#' can be used to specify a counted string
286 type to be packed or unpacked. See L<perlfunc/"pack">.
288 =head2 $^X variables may now have names longer than one character
290 Formerly, $^X was synonymous with ${"\cX"}, but $^XY was a syntax
291 error. Now variable names that begin with a control character may be
292 arbitrarily long. However, for compatibility reasons, these variables
293 I<must> be written with explicit braces, as C<${^XY}> for example.
294 C<${^XYZ}> is synonymous with ${"\cXYZ"}. Variable names with more
295 than one control character, such as C<${^XY^Z}>, are illegal.
297 The old syntax has not changed. As before, `^X' may be either a
298 literal control-X character or the two-character sequence `caret' plus
299 `X'. When braces are omitted, the variable name stops after the
300 control character. Thus C<"$^XYZ"> continues to be synonymous with
301 C<$^X . "YZ"> as before.
303 As before, lexical variables may not have names beginning with control
304 characters. As before, variables whose names begin with a control
305 character are always forced to be in package `main'. All such variables
306 are reserved for future extensions, except those that begin with
307 C<^_>, which may be used by user programs and are guaranteed not to
308 acquire special meaning in any future version of Perl.
310 =head2 C<use attrs> implicit in subroutine attributes
312 Formerly, if you wanted to mark a subroutine as being a method call or
313 as requiring an automatic lock() when it is entered, you had to declare
314 that with a C<use attrs> pragma in the body of the subroutine.
315 That can now be accomplished with a declaration syntax, like this:
317 sub mymethod : locked, method ;
319 sub mymethod : locked, method {
323 F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes
324 with the stubs they provide. See L<attributes>.
326 =head1 Significant bug fixes
328 =head2 E<lt>HANDLEE<gt> on empty files
330 With C<$/> set to C<undef>, slurping an empty file returns a string of
331 zero length (instead of C<undef>, as it used to) the first time the
332 HANDLE is read. Further reads yield C<undef>.
334 This means that the following will append "foo" to an empty file (it used
337 perl -0777 -pi -e 's/^/foo/' empty_file
341 perl -pi -e 's/^/foo/' empty_file
343 is unchanged (it continues to leave the file empty).
345 =head2 C<eval '...'> improvements
347 Line numbers (as reflected by caller() and most diagnostics) within
348 C<eval '...'> were often incorrect when here documents were involved.
349 This has been corrected.
351 Lexical lookups for variables appearing in C<eval '...'> within
352 functions that were themselves called within an C<eval '...'> were
353 searching the wrong place for lexicals. The lexical search now
354 correctly ends at the subroutine's block boundary.
356 Parsing of here documents used to be flawed when they appeared as
357 the replacement expression in C<eval 's/.../.../e'>. This has
360 =head2 Automatic flushing of output buffers
362 fork(), exec(), system(), qx//, and pipe open()s now flush buffers
363 of all files opened for output when the operation
364 was attempted. This mostly eliminates confusing
365 buffering mishaps suffered by users unaware of how Perl internally
368 =head2 Better diagnostics on meaningless filehandle operations
370 Constructs such as C<open(E<lt>FHE<gt>)> and C<close(E<lt>FHE<gt>)>
371 are compile time errors. Attempting to read from filehandles that
372 were opened only for writing will now produce warnings (just as
373 writing to read-only filehandles does).
375 =head2 Buffered data discarded from input filehandle when dup'ed.
377 C<open(NEW, "E<lt>&OLD")> now discards any data that was previously
378 read and buffered in C<OLD>. The next read operation on C<NEW> will
379 return the same data as the corresponding operation on C<OLD>.
380 Formerly, it would have returned the data from the start of the
381 following disk block instead.
383 =head1 Supported Platforms
389 VM/ESA is now supported.
393 Siemens BS2000 is now supported under the POSIX Shell.
397 The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
402 GNU/Hurd is now supported.
406 Rhapsody is now supported.
410 EPOC is is now supported (on Psion 5).
420 Compatibility tests for C<sub : attrs> vs the older C<use attrs>.
424 IO constants (SEEK_*, _IO*).
428 Directory-related IO methods (new, read, close, rewind, tied delete).
430 =item lib/io_multihomed
432 INET sockets with multi-homed hosts.
444 Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>.
452 Verify operations that access pad objects (lexicals and temporaries).
456 =head1 Modules and Pragmata
464 While used internally by Perl as a pragma, this module also
465 provides a way to fetch subroutine and variable attributes.
470 The ByteLoader is a dedication extension to generate and run
471 Perl bytecode. See L<ByteLoader>.
475 The Perl Compiler suite has been extensively reworked for this
480 Devel::DProf, a Perl source code profiler has been added.
484 Added Dumpvalue module provides screen dumps of Perl data.
488 You can now run tests for I<n> seconds instead of guessing the right
489 number of tests to run: e.g. timethese(-5, ...) will run each
490 code for at least 5 CPU seconds. Zero as the "number of repetitions"
491 means "for at least 3 CPU seconds". The output format has also
492 changed. For example:
494 use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
496 will now output something like this:
498 Benchmark: running a, b, each for at least 5 CPU seconds...
499 a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
500 b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
502 New features: "each for at least N CPU seconds...", "wallclock secs",
503 and the "@ operations/CPU second (n=operations)".
507 The Devel::Peek module provides access to the internal representation
508 of Perl variables and data. It is a data debugging tool for the XS programmer.
512 More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
513 large (more than 4G) file access (64-bit support is not yet
514 working, though, so no need to get overly excited), Free/Net/OpenBSD
515 locking behaviour flags F_FLOCK, F_POSIX, Linux F_SHLCK, and
516 O_ACCMODE: the mask of O_RDONLY, O_WRONLY, and O_RDWR.
520 New methods have been added to the File::Spec module: devnull() returns
521 the name of the null device (/dev/null on Unix) and tmpdir() the name of
522 the temp directory (normally /tmp on Unix). There are now also methods
523 to convert between absolute and relative filenames: abs2rel() and
524 rel2abs(). For compatibility with operating systems that specify volume
525 names in file paths, the splitpath(), splitdir(), and catdir() methods
528 =item File::Spec::Functions
530 The new File::Spec::Functions modules provides a function interface
531 to the File::Spec module. Allows shorthand
533 $fullname = catfile($dir1, $dir2, $file);
537 $fullname = File::Spec->catfile($dir1, $dir2, $file);
541 The logical operations C<E<lt>E<lt>>, C<E<gt>E<gt>>, C<&>, C<|>,
542 and C<~> are now supported on bigints.
546 The accessor methods Re, Im, arg, abs, rho, and theta can now also
547 act as mutators (accessor $z->Re(), mutator $z->Re(3)).
551 A little bit of radial trigonometry (cylindrical and spherical),
552 radial coordinate conversions, and the great circle distance were added.
556 An EXISTS method has been added to this module (and sdbm_exists() has
557 been added to the underlying sdbm library), so one can now call exists
558 on an SDBM_File tied hash and get the correct result, rather than a
563 The timelocal() and timegm() functions used to silently return bogus
564 results when the date exceeded the machine's integer range. They
565 now consistently croak() if the date falls in an unsupported range--
566 but on the other hand they now accept "out-of-limits" day-of-month
567 to make "Julian date" conversions easier.
571 The error return value in list context has been changed for all functions
572 that return a list of values. Previously these functions returned a list
573 with a single element C<undef> if an error occurred. Now these functions
574 return the empty list in these situations. This applies to the following
580 The remaining functions are unchanged and continue to return C<undef> on
581 error even in list context.
583 The Win32::SetLastError(ERROR) function has been added as a complement
584 to the Win32::GetLastError() function.
586 The new Win32::GetFullPathName(FILENAME) returns the full absolute
587 pathname for FILENAME in scalar context. In list context it returns
588 a two-element list containing the fully qualified directory name and
593 A new feature called "DBM Filters" has been added to all the
594 DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
595 DBM Filters add four new methods to each DBM module:
602 These can be used to filter key-value pairs before the pairs are
603 written to the database or just after they are read from the database.
604 See L<perldbmfilter> for further information.
610 C<use attrs> is now obsolescent, and is only provided for
611 backward-compatibility. It's been replaced by the C<sub : attributes>
612 syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>.
614 C<use utf8> to enable UTF-8 and Unicode support.
616 C<use caller 'encoding'> allows modules to inherit pragmatic attributes
617 from the caller's context. C<encoding> is currently the only supported
620 Lexical warnings pragma, C<use warnings;>, to control optional warnings.
622 C<use filetest> to control the behaviour of filetests (C<-r> C<-w> ...).
623 Currently only one subpragma implemented, "use filetest 'access';",
624 that enables the use of access(2) or equivalent to check
625 permissions instead of using stat(2) as usual. This matters
626 in filesystems where there are ACLs (access control lists): the
627 stat(2) might lie, but access(2) knows better.
629 =head1 Utility Changes
633 =head1 Documentation Changes
637 =item perlopentut.pod
639 A tutorial on using open() effectively.
643 A tutorial that introduces the essentials of references.
647 A tutorial on managing class data for object modules.
651 =head1 New Diagnostics
653 =item "my sub" not yet implemented
655 (F) Lexically scoped subroutines are not yet implemented. Don't try that
658 =item %s package attribute may clash with future reserved word: %s
660 (W) A lowercase attribute name was used that had a package-specific handler.
661 That name might have a meaning to Perl itself some day, even though it
662 doesn't yet. Perhaps you should use a mixed-case attribute name, instead.
665 =item /%s/: Unrecognized escape \\%c passed through
667 (W) You used a backslash-character combination which is not recognized
668 by Perl. This combination appears in an interpolated variable or a
669 C<'>-delimited regular expression.
671 =item Filehandle %s opened only for output
673 (W) You tried to read from a filehandle opened only for writing. If you
674 intended it to be a read-write filehandle, you needed to open it with
675 "+E<lt>" or "+E<gt>" or "+E<gt>E<gt>" instead of with "E<lt>" or nothing. If
676 you intended only to read from the file, use "E<lt>". See
679 =item Invalid %s attribute: %s
681 The indicated attribute for a subroutine or variable was not recognized
682 by Perl or by a user-supplied handler. See L<attributes>.
684 =item Invalid %s attributes: %s
686 The indicated attributes for a subroutine or variable were not recognized
687 by Perl or by a user-supplied handler. See L<attributes>.
689 =item Invalid separator character %s in attribute list
691 (F) Something other than a comma or whitespace was seen between the
692 elements of an attribute list. If the previous attribute
693 had a parenthesised parameter list, perhaps that list was terminated
694 too soon. See L<attributes>.
696 =item Missing command in piped open
698 (W) You used the C<open(FH, "| command")> or C<open(FH, "command |")>
699 construction, but the command was missing or blank.
701 =item Missing name in "my sub"
703 (F) The reserved syntax for lexically scoped subroutines requires that they
704 have a name with which they can be found.
706 =item Unrecognized escape \\%c passed through
708 (W) You used a backslash-character combination which is not recognized
711 =item Unterminated attribute parameter in attribute list
713 (F) The lexer saw an opening (left) parenthesis character while parsing an
714 attribute list, but the matching closing (right) parenthesis
715 character was not found. You may need to add (or remove) a backslash
716 character to get your parentheses to balance. See L<attributes>.
718 =item Unterminated attribute list
720 (F) The lexer found something other than a simple identifier at the start
721 of an attribute, and it wasn't a semicolon or the start of a
722 block. Perhaps you terminated the parameter list of the previous attribute
723 too soon. See L<attributes>.
725 =item defined(@array) is deprecated
727 (D) defined() is not usually useful on arrays because it checks for an
728 undefined I<scalar> value. If you want to see if the array is empty,
729 just use C<if (@array) { # not empty }> for example.
731 =item defined(%hash) is deprecated
733 (D) defined() is not usually useful on hashes because it checks for an
734 undefined I<scalar> value. If you want to see if the hash is empty,
735 just use C<if (%hash) { # not empty }> for example.
737 =item Invalid separator character %s in subroutine attribute list
739 (F) Something other than a comma or whitespace was seen between the
740 elements of a subroutine attribute list. If the previous attribute
741 had a parenthesised parameter list, perhaps that list was terminated
744 =item Possible Y2K bug: %s
746 (W) You are concatenating the number 19 with another number, which
747 could be a potential Year 2000 problem.
749 =item Unterminated attribute parameter in subroutine attribute list
751 (F) The lexer saw an opening (left) parenthesis character while parsing a
752 subroutine attribute list, but the matching closing (right) parenthesis
753 character was not found. You may need to add (or remove) a backslash
754 character to get your parentheses to balance.
756 =item Unterminated subroutine attribute list
758 (F) The lexer found something other than a simple identifier at the start
759 of a subroutine attribute, and it wasn't a semicolon or the start of a
760 block. Perhaps you terminated the parameter list of the previous attribute
763 =item /%s/ should probably be written as "%s"
765 (W) You have used a pattern where Perl expected to find a string,
766 like in the first argument to C<join>. Perl will treat the true
767 or false result of matching the pattern against $_ as the string,
768 which is probably not what you had in mind.
770 =head1 Obsolete Diagnostics
774 =head1 Configuration Changes
776 =head2 installusrbinperl
778 You can use "Configure -Uinstallusrbinperl" which causes installperl
779 to skip installing perl also as /usr/bin/perl. This is useful if you
780 prefer not to modify /usr/bin for some reason or another but harmful
781 because many scripts assume to find Perl in /usr/bin/perl.
785 You can use "Configure -Dusesocks" which causes Perl to probe
786 for the SOCKS proxy protocol library, http://www.socks.nec.com/
790 You can "post-edit" the Configure variables using the Configure -A
791 flag. The editing happens immediately after the platform specific
792 hints files have been processed but before the actual configuration
793 process starts. Run Configure -h to find out the full -A syntax.
797 If you find what you think is a bug, you might check the headers of
798 articles recently posted to the comp.lang.perl.misc newsgroup.
799 There may also be information at http://www.perl.com/perl/, the Perl
802 If you believe you have an unreported bug, please run the B<perlbug>
803 program included with your release. Make sure to trim your bug down
804 to a tiny but sufficient test case. Your bug report, along with the
805 output of C<perl -V>, will be sent off to perlbug@perl.com to be
806 analysed by the Perl porting team.
810 The F<Changes> file for exhaustive details on what changed.
812 The F<INSTALL> file for how to build Perl.
814 The F<README> file for general stuff.
816 The F<Artistic> and F<Copying> files for copyright information.
820 Written by Gurusamy Sarathy <F<gsar@umich.edu>>, with many contributions
821 from The Perl Porters.
823 Send omissions or corrections to <F<perlbug@perl.com>>.