3 perldelta - what's new for perl v5.6 (as of v5.005_58)
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
20 =head2 C Source Incompatibilities
26 Release 5.005 grandfathered old global symbol names by providing preprocessor
27 macros for extension source compatibility. As of release 5.6, these
28 preprocessor definitions are not available by default. You need to explicitly
29 compile perl with C<-DPERL_POLLUTE> to get these definitions. For
30 extensions still using the old symbols, this option can be
31 specified via MakeMaker:
33 perl Makefile.PL POLLUTE=1
35 =item C<PERL_IMPLICIT_CONTEXT>
37 This new build option provides a set of macros for all API functions
38 such that an implicit interpreter/thread context argument is passed to
39 every API function. As a result of this, something like C<sv_setsv(foo,bar)>
40 amounts to a macro invocation that actually translates to something like
41 C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
42 to not have any significant source compatibility issues, the difference
43 between a macro and a real function call will need to be considered.
45 This means that there B<is> a source compatibility issue as a result of
46 this if your extensions attempt to use pointers to any of the Perl API
49 Note that the above issue is not relevant to the default build of
50 Perl, whose interfaces continue to match those of prior versions
51 (but subject to the other options described here).
53 For testing purposes, the 5.005_58 release automatically enables
54 PERL_IMPLICIT_CONTEXT whenever Perl is built with -Dusethreads or
57 See L<perlguts/"The Perl API"> for detailed information on the
58 ramifications of building Perl using this option.
60 =item C<PERL_POLLUTE_MALLOC>
62 Enabling Perl's malloc in release 5.005 and earlier caused
63 the namespace of system versions of the malloc family of functions to
64 be usurped by the Perl versions, since by default they used the
67 Besides causing problems on platforms that do not allow these functions to
68 be cleanly replaced, this also meant that the system versions could not
69 be called in programs that used Perl's malloc. Previous versions of Perl
70 have allowed this behaviour to be suppressed with the HIDEMYMALLOC and
71 EMBEDMYMALLOC preprocessor definitions.
73 As of release 5.6, Perl's malloc family of functions have default names
74 distinct from the system versions. You need to explicitly compile perl with
75 C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
76 and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
79 Note that these functions do B<not> constitute Perl's memory allocation API.
80 See L<perlguts/"Memory Allocation"> for further information about that.
82 =item C<PL_na> and C<dTHR> Issues
84 The C<PL_na> global is now thread local, so a C<dTHR> declaration is needed
85 in the scope in which the global appears. XSUBs should handle this automatically,
86 but if you have used C<PL_na> in support functions, you either need to
87 change the C<PL_na> to a local variable (which is recommended), or put in
92 =head2 Compatible C Source API Changes
96 =item C<PATCHLEVEL> is now C<PERL_VERSION>
98 The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
99 are now available by default from perl.h, and reflect the base revision,
100 patchlevel, and subversion respectively. C<PERL_REVISION> had no
101 prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
102 previously available as C<PATCHLEVEL> and C<SUBVERSION>.
104 The new names cause less pollution of the B<cpp> namespace and reflect what
105 the numbers have come to stand for in common practice. For compatibility,
106 the old names are still supported when F<patchlevel.h> is explicitly
107 included (as required before), so there is no source incompatibility
112 =head2 Binary Incompatibilities
114 The default build of this release is binary compatible with the 5.005
115 release or its maintenance versions.
117 The usethreads or usemultiplicity builds are B<not> binary compatible
118 with the corresponding builds in 5.005.
122 =head2 Unicode and UTF-8 support
124 Perl can optionally use UTF-8 as its internal representation for character
125 strings. The C<use utf8> pragma enables this support in the current lexical
126 scope. See L<utf8> for more information.
128 =head2 Lexically scoped warning categories
130 You can now control the granularity of warnings emitted by perl at a finer
131 level using the C<use warning> pragma. See L<warning> and L<perllexwarn>
134 =head2 Binary numbers supported
136 Binary numbers are now supported as literals, in s?printf formats, and
140 printf "The answer is: %b\n", oct("0b101010");
142 =head2 syswrite() ease-of-use
144 The length argument of C<syswrite()> is now optional.
146 =head2 64-bit support
148 All platforms that have 64-bit integers either (a) natively as longs
149 or ints (b) via special compiler flags (c) using long long are able to
150 use "quads" (64-integers) as follows:
154 =item constants in the code
156 =item arguments to oct() and hex()
158 =item arguments to print(), printf() and sprintf()
160 =item pack() and unpack() "q" format
162 =item in basic arithmetics
166 Note that unless you have the case (a) you will have to configure
167 and compile Perl using the -Duse64bits Configure flag.
169 Unfortunately, bit operations (&, <<, ...) and vec() do not work,
170 they are limited to 32 bits.
172 Last but not least: note that due to Perl's tendency to always use
173 floating point numbers the quads are not true integers. They may lose
174 their precision due to rounding errors, and when they get large their
175 less significant digits will fall off.
177 =head2 Better syntax checks on parenthesized unary operators
181 print defined(&foo,&bar,&baz);
182 print uc("foo","bar","baz");
185 used to be accidentally allowed in earlier versions, and produced
186 unpredictable behaviour. Some produced ancillary warnings
187 when used in this way; others silently did the wrong thing.
189 The parenthesized forms of most unary operators that expect a single
190 argument now ensure that they are not called with more than one
191 argument, making the cases shown above syntax errors. The usual
194 print defined &foo, &bar, &baz;
195 print uc "foo", "bar", "baz";
198 remains unchanged. See L<perlop>.
200 =head2 POSIX character class syntax [: :] supported
202 For example to match alphabetic characters use /[[:alpha:]]/.
203 See L<perlre> for details.
205 =head2 Improved C<qw//> operator
207 The C<qw//> operator is now evaluated at compile time into a true list
208 instead of being replaced with a run time call to C<split()>. This
209 removes the confusing misbehaviour of C<qw//> in scalar context, which
210 had inherited that behaviour from split().
214 $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
216 now correctly prints "3|a", instead of "2|a".
218 =head2 pack() format 'Z' supported
220 The new format type 'Z' is useful for packing and unpacking null-terminated
221 strings. See L<perlfunc/"pack">.
223 =head2 pack() format modifier '!' supported
225 The new format type modifier '!' is useful for packing and unpacking
226 native shorts, ints, and longs. See L<perlfunc/"pack">.
228 =head2 pack() and unpack() support counted strings
230 The template character '#' can be used to specify a counted string
231 type to be packed or unpacked. See L<perlfunc/"pack">.
233 =head2 $^X variables may now have names longer than one character
235 Formerly, $^X was synonymous with ${"\cX"}, but $^XY was a syntax
236 error. Now variable names that begin with a control character may be
237 arbitrarily long. However, for compatibility reasons, these variables
238 I<must> be written with explicit braces, as C<${^XY}> for example.
239 C<${^XYZ}> is synonymous with ${"\cXYZ"}. Variable names with more
240 than one control character, such as C<${^XY^Z}>, are illegal.
242 The old syntax has not changed. As before, `^X' may be either a
243 literal control-X character or the two-character sequence `caret' plus
244 `X'. When braces are omitted, the variable name stops after the
245 control character. Thus C<"$^XYZ"> continues to be synonymous with
246 C<$^X . "YZ"> as before.
248 As before, lexical variables may not have names beginning with control
249 characters. As before, variables whose names begin with a control
250 character are always forced to be in package `main'. All such variables
251 are reserved for future extensions, except those that begin with
252 C<^_>, which may be used by user programs and is guaranteed not to
253 acquire special meaning in any future version of Perl.
255 =head1 Significant bug fixes
257 =head2 E<lt>HANDLEE<gt> on empty files
259 With C<$/> set to C<undef>, slurping an empty file returns a string of
260 zero length (instead of C<undef>, as it used to) the first time the
261 HANDLE is read. Further reads yield C<undef>.
263 This means that the following will append "foo" to an empty file (it used
266 perl -0777 -pi -e 's/^/foo/' empty_file
270 perl -pi -e 's/^/foo/' empty_file
272 is unchanged (it continues to leave the file empty).
274 =head2 C<eval '...'> improvements
276 Line numbers (as reflected by caller() and most diagnostics) within
277 C<eval '...'> were often incorrect when here documents were involved.
278 This has been corrected.
280 Lexical lookups for variables appearing in C<eval '...'> within
281 functions that were themselves called within an C<eval '...'> were
282 searching the wrong place for lexicals. The lexical search now
283 correctly ends at the subroutine's block boundary.
285 Parsing of here documents used to be flawed when they appeared as
286 the replacement expression in C<eval 's/.../.../e'>. This has
289 =head2 Automatic flushing of output buffers
291 fork(), exec(), system(), qx//, and pipe open()s now flush buffers
292 of all files opened for output when the operation
293 was attempted. This mostly eliminates confusing
294 buffering mishaps suffered by users unaware of how Perl internally
297 =head2 Better diagnostics on meaningless filehandle operations
299 Constructs such as C<open(E<lt>FHE<gt>)> and C<close(E<lt>FHE<gt>)>
300 are compile time errors. Attempting to read from filehandles that
301 were opened only for writing will now produce warnings (just as
302 writing to read-only filehandles does).
304 =head2 Buffered data discarded from input filehandle when dup'ed.
306 C<open(NEW, "E<lt>&OLD")> now discards any data that was previously
307 read and buffered in C<OLD>. The next read operation on C<NEW> will
308 return the same data as the corresponding operation on C<OLD>.
309 Formerly, it would have returned the data from the start of the
310 following disk block instead.
312 =head1 Supported Platforms
318 VM/ESA is now supported.
322 Siemens BS2000 is now supported under the POSIX Shell.
326 The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
331 GNU/Hurd is now supported.
335 Rhapsody is now supported.
339 EPOC is is now supported (on Psion 5).
349 IO constants (SEEK_*, _IO*).
353 Directory-related IO methods (new, read, close, rewind, tied delete).
355 =item op/io_multihomed
357 INET sockets with multi-homed hosts.
373 Verify operations that access pad objects (lexicals and temporaries).
377 =head1 Modules and Pragmata
385 The ByteLoader is a dedication extension to generate and run
386 Perl bytecode. See L<ByteLoader>.
390 The Perl Compiler suite has been extensively reworked for this
395 Devel::DProf, a Perl source code profiler has been added.
399 Added Dumpvalue module provides screen dumps of Perl data.
403 You can now run tests for I<n> seconds instead of guessing the right
404 number of tests to run: e.g. timethese(-5, ...) will run each
405 code for at least 5 CPU seconds. Zero as the "number of repetitions"
406 means "for at least 3 CPU seconds". The output format has also
407 changed. For example:
409 use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
411 will now output something like this:
413 Benchmark: running a, b, each for at least 5 CPU seconds...
414 a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
415 b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
417 New features: "each for at least N CPU seconds...", "wallclock secs",
418 and the "@ operations/CPU second (n=operations)".
422 The Devel::Peek module provides access to the internal representation
423 of Perl variables and data. It is a data debugging tool for the XS programmer.
427 More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
428 large (more than 4G) file access (64-bit support is not yet
429 working, though, so no need to get overly excited), Free/Net/OpenBSD
430 locking behaviour flags F_FLOCK, F_POSIX, Linux F_SHLCK, and
431 O_ACCMODE: the mask of O_RDONLY, O_WRONLY, and O_RDWR.
435 New methods have been added to the File::Spec module: devnull() returns
436 the name of the null device (/dev/null on Unix) and tmpdir() the name of
437 the temp directory (normally /tmp on Unix). There are now also methods
438 to convert between absolute and relative filenames: abs2rel() and
439 rel2abs(). For compatibility with operating systems that specify volume
440 names in file paths, the splitpath(), splitdir(), and catdir() methods
443 =item File::Spec::Functions
445 The new File::Spec::Functions modules provides a function interface
446 to the File::Spec module. Allows shorthand
448 $fullname = catfile($dir1, $dir2, $file);
452 $fullname = File::Spec->catfile($dir1, $dir2, $file);
456 The logical operations C<E<lt>E<lt>>, C<E<gt>E<gt>>, C<&>, C<|>,
457 and C<~> are now supported on bigints.
461 The accessor methods Re, Im, arg, abs, rho, and theta can now also
462 act as mutators (accessor $z->Re(), mutator $z->Re(3)).
466 A little bit of radial trigonometry (cylindrical and spherical),
467 radial coordinate conversions, and the great circle distance were added.
471 An EXISTS method has been added to this module (and sdbm_exists() has
472 been added to the underlying sdbm library), so one can now call exists
473 on an SDBM_File tied hash and get the correct result, rather than a
478 The timelocal() and timegm() functions used to silently return bogus
479 results when the date exceeded the machine's integer range. They
480 now consistently croak() if the date falls in an unsupported range.
484 The error return value in list context has been changed for all functions
485 that return a list of values. Previously these functions returned a list
486 with a single element C<undef> if an error occurred. Now these functions
487 return the empty list in these situations. This applies to the following
493 The remaining functions are unchanged and continue to return C<undef> on
494 error even in list context.
496 The Win32::SetLastError(ERROR) function has been added as a complement
497 to the Win32::GetLastError() function.
499 The new Win32::GetFullPathName(FILENAME) returns the full absolute
500 pathname for FILENAME in scalar context. In list context it returns
501 a two-element list containing the fully qualified directory name and
506 A new feature called "DBM Filters" has been added to all the
507 DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
508 DBM Filters add four new methods to each DBM module:
515 These can be used to filter key-value pairs before the pairs are
516 written to the database or just after they are read from the database.
517 See L<perldbmfilter> for further information.
523 C<use utf8> to enable UTF-8 and Unicode support.
525 C<use caller 'encoding'> allows modules to inherit pragmatic attributes
526 from the caller's context. C<encoding> is currently the only supported
529 Lexical warnings pragma, C<use warning;>, to control optional warnings.
531 C<use filetest> to control the behaviour of filetests (C<-r> C<-w> ...).
532 Currently only one subpragma implemented, "use filetest 'access';",
533 that enables the use of access(2) or equivalent to check
534 permissions instead of using stat(2) as usual. This matters
535 in filesystems where there are ACLs (access control lists): the
536 stat(2) might lie, but access(2) knows better.
538 =head1 Utility Changes
542 =head1 Documentation Changes
546 =item perlopentut.pod
548 A tutorial on using open() effectively.
552 A tutorial that introduces the essentials of references.
556 A tutorial on managing class data for object modules.
560 =head1 New Diagnostics
562 =item /%s/: Unrecognized escape \\%c passed through
564 (W) You used a backslash-character combination which is not recognized
565 by Perl. This combination appears in an interpolated variable or a
566 C<'>-delimited regular expression.
568 =item Filehandle %s opened only for output
570 (W) You tried to read from a filehandle opened only for writing. If you
571 intended it to be a read-write filehandle, you needed to open it with
572 "+E<lt>" or "+E<gt>" or "+E<gt>E<gt>" instead of with "E<lt>" or nothing. If
573 you intended only to read from the file, use "E<lt>". See
576 =item Missing command in piped open
578 (W) You used the C<open(FH, "| command")> or C<open(FH, "command |")>
579 construction, but the command was missing or blank.
581 =item Unrecognized escape \\%c passed through
583 (W) You used a backslash-character combination which is not recognized
586 =item defined(@array) is deprecated
588 (D) defined() is not usually useful on arrays because it checks for an
589 undefined I<scalar> value. If you want to see if the array is empty,
590 just use C<if (@array) { # not empty }> for example.
592 =item defined(%hash) is deprecated
594 (D) defined() is not usually useful on hashes because it checks for an
595 undefined I<scalar> value. If you want to see if the hash is empty,
596 just use C<if (%hash) { # not empty }> for example.
598 =head1 Obsolete Diagnostics
602 =head1 Configuration Changes
604 =head2 installusrbinperl
606 You can use "Configure -Uinstallusrbinperl" which causes installperl
607 to skip installing perl also as /usr/bin/perl. This is useful if you
608 prefer not to modify /usr/bin for some reason or another but harmful
609 because many scripts assume to find Perl in /usr/bin/perl.
613 You can use "Configure -Dusesocks" which causes Perl to probe
614 for the SOCKS proxy protocol library, http://www.socks.nec.com/
618 If you find what you think is a bug, you might check the headers of
619 articles recently posted to the comp.lang.perl.misc newsgroup.
620 There may also be information at http://www.perl.com/perl/, the Perl
623 If you believe you have an unreported bug, please run the B<perlbug>
624 program included with your release. Make sure to trim your bug down
625 to a tiny but sufficient test case. Your bug report, along with the
626 output of C<perl -V>, will be sent off to perlbug@perl.com to be
627 analysed by the Perl porting team.
631 The F<Changes> file for exhaustive details on what changed.
633 The F<INSTALL> file for how to build Perl.
635 The F<README> file for general stuff.
637 The F<Artistic> and F<Copying> files for copyright information.
641 Written by Gurusamy Sarathy <F<gsar@umich.edu>>, with many contributions
642 from The Perl Porters.
644 Send omissions or corrections to <F<perlbug@perl.com>>.