This document describes differences between the 5.005 release and this one.
-=head1 Incompatible Changes
+=head1 Core Enhancements
-=head2 Perl Source Incompatibilities
+=head2 Interpreter cloning, threads, and concurrency
-Beware that any new warnings that have been added or old ones
-that have been enhanced are B<not> considered incompatible changes.
+Perl 5.005_63 introduces the beginnings of support for running multiple
+interpreters concurrently in different threads. In conjunction with
+the perl_clone() API call, which can be used to selectively duplicate
+the state of any given interpreter, it is possible to compile a
+piece of code once in an interpreter, clone that interpreter
+one or more times, and run all the resulting interpreters in distinct
+threads.
-Since all new warnings must be explicitly requested via the C<-w>
-switch or the C<warnings> pragma, it is ultimately the programmer's
-responsibility to ensure that warnings are enabled judiciously.
+On the Windows platform, this feature is used to emulate fork() at the
+interpreter level. See L<perlfork> for details about that.
-=over 4
+This feature is still in evolution. It is eventually meant to be used
+to selectively clone a subroutine and data reachable from that
+subroutine in a separate interpreter and run the cloned subroutine
+in a separate thread. Since there is no shared data between the
+interpreters, little or no locking will be needed (unless parts of
+the symbol table are explicitly shared). This is obviously intended
+to be an easy-to-use replacement for the existing threads support.
-=item CHECK is a new keyword
+Support for cloning interpreters and interpreter concurrency can be
+enabled using the -Dusethreads Configure option (see win32/Makefile for
+how to enable it on Windows.) The resulting perl executable will be
+functionally identical to one that was built with -Dmultiplicity, but
+the perl_clone() API call will only be available in the former.
-In addition to C<BEGIN>, C<INIT>, C<END>, C<DESTROY> and C<AUTOLOAD>,
-subroutines named C<CHECK> are now special. These are queued up during
-compilation and behave similar to END blocks, except they are called at
-the end of compilation rather than at the end of execution. They cannot
-be called directly.
+-Dusethreads enables the cpp macro USE_ITHREADS by default, which in turn
+enables Perl source code changes that provide a clear separation between
+the op tree and the data it operates with. The former is immutable, and
+can therefore be shared between an interpreter and all of its clones,
+while the latter is considered local to each interpreter, and is therefore
+copied for each clone.
-=item Treatment of list slices of undef has changed
+Note that building Perl with the -Dusemultiplicity Configure option
+is adequate if you wish to run multiple B<independent> interpreters
+concurrently in different threads. -Dusethreads only provides the
+additional functionality of the perl_clone() API call and other
+support for running B<cloned> interpreters concurrently.
-When taking a slice of a literal list (as opposed to a slice of
-an array or hash), Perl used to return an empty list if the
-result happened to be composed of all undef values.
+ NOTE: This is an experimental feature. Implementation details are
+ subject to change.
-The new behavior is to produce an empty list if (and only if)
-the original list was empty. Consider the following example:
+=head2 Lexically scoped warning categories
- @a = (1,undef,undef,2)[2,1,2];
+You can now control the granularity of warnings emitted by perl at a finer
+level using the C<use warnings> pragma. L<warnings> and L<perllexwarn>
+have copious documentation on this feature.
-The old behavior would have resulted in @a having no elements.
-The new behavior ensures it has three undefined elements.
+=head2 Unicode and UTF-8 support
-Note in particular that the behavior of slices of the following
-cases remains unchanged:
+Perl now uses UTF-8 as its internal representation for character
+strings. The C<utf8> and C<bytes> pragmas are used to control this support
+in the current lexical scope. See L<perlunicode>, L<utf8> and L<bytes> for
+more information.
- @a = ()[1,2];
- @a = (getpwent)[7,0];
- @a = (anything_returning_empty_list())[2,1,2];
- @a = @b[2,1,2];
- @a = @c{'a','b','c'};
+This feature is expected to evolve quickly to support some form of I/O
+disciplines that can be used to specify the kind of input and output data
+(bytes or characters). Until that happens, additional modules from CPAN
+will be needed to complete the toolkit for dealing with Unicode.
-See L<perldata>.
+ NOTE: This should be considered an experimental feature. Implementation
+ details are subject to change.
+
+=head2 Support for interpolating named characters
+
+The new C<\N> escape interpolates named characters within strings.
+For example, C<"Hi! \N{WHITE SMILING FACE}"> evaluates to a string
+with a unicode smiley face at the end.
+
+=head2 "our" declarations
+
+An "our" declaration introduces a value that can be best understood
+as a lexically scoped symbolic alias to a global variable in the
+package that was current where the variable was declared. This is
+mostly useful as an alternative to the C<vars> pragma, but also provides
+the opportunity to introduce typing and other attributes for such
+variables. See L<perlfunc/our>.
+
+=head2 Support for strings represented as a vector of ordinals
+
+Literals of the form C<v1.2.3.4> are now parsed as a string composed
+of characters with the specified ordinals. This is an alternative, more
+readable way to construct (possibly unicode) strings instead of
+interpolating characters, as in C<"\x{1}\x{2}\x{3}\x{4}">. The leading
+C<v> may be omitted if there are more than two ordinals, so C<1.2.3> is
+parsed the same as C<v1.2.3>.
+
+Strings written in this form are also useful to represent version "numbers".
+It is easy to compare such version "numbers" (which are really just plain
+strings) using any of the usual string comparison operators C<eq>, C<ne>,
+C<lt>, C<gt>, etc., or perform bitwise string operations on them using C<|>,
+C<&>, etc.
+
+In conjunction with the new C<$^V> magic variable (which contains
+the perl version as a string), such literals can be used as a readable way
+to check if you're running a particular version of Perl:
+
+ # this will parse in older versions of Perl also
+ if ($^V and $^V gt v5.6.0) {
+ # new features supported
+ }
+
+C<require> and C<use> also have some special magic to support such literals.
+They will be interpreted as a version rather than as a module name:
+
+ require v5.6.0; # croak if $^V lt v5.6.0
+ use v5.6.0; # same, but croaks at compile-time
+
+Alternatively, the C<v> may be omitted if there is more than one dot:
+
+ require 5.6.0;
+ use 5.6.0;
+
+Also, C<sprintf> and C<printf> support the Perl-specific format flag C<%v>
+to print ordinals of characters in arbitrary strings:
+
+ printf "v%vd", $^V; # prints current version, such as "v5.5.650"
+ printf "%*vX", ":", $addr; # formats IPv6 address
+ printf "%*vb", " ", $bits; # displays bitstring
+
+See L<perldata/"Scalar value constructors"> for additional information.
-=head2 Perl's version numbering has changed
+=head2 Improved Perl version numbering system
Beginning with Perl version 5.6.0, the version number convention has been
changed to a "dotted integer" scheme that is more commonly found in open
equivalent to a floating point value of 5.006_001 in the older format,
stored in C<$]>).
-=item Literals of the form C<1.2.3> parse differently
+=head2 New syntax for declaring subroutine attributes
-Previously, numeric literals with more than one dot in them were
-interpreted as a floating point number concatenated with one or more
-numbers. Such "numbers" are now parsed as strings composed of the
-specified ordinals.
+Formerly, if you wanted to mark a subroutine as being a method call or
+as requiring an automatic lock() when it is entered, you had to declare
+that with a C<use attrs> pragma in the body of the subroutine.
+That can now be accomplished with declaration syntax, like this:
-For example, C<print 97.98.99> used to output C<97.9899> in earlier
-versions, but now prints C<abc>.
+ sub mymethod : locked method ;
+ ...
+ sub mymethod : locked method {
+ ...
+ }
-See L<Support for strings represented as a vector of ordinals> below.
+ sub othermethod :locked :method ;
+ ...
+ sub othermethod :locked :method {
+ ...
+ }
-=item Possibly changed pseudo-random number generator
-In 5.005_0x and earlier, perl's rand() function used the C library
-rand(3) function. As of 5.005_52, Configure tests for drand48(),
-random(), and rand() (in that order) and picks the first one it finds.
-Perl programs that depend on reproducing a specific set of pseudo-random
-numbers will now likely produce different output. You can use
-C<sh Configure -Drandfunc=rand> to obtain the old behavior.
+(Note how only the first C<:> is mandatory, and whitespace surrounding
+the C<:> is optional.)
-=item Hashing function for hash keys has changed
+F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes
+with the stubs they provide. See L<attributes>.
-Perl hashes are not order preserving. The apparently random order
-encountered when iterating on the contents of a hash is determined
-by the hashing algorithm used. To improve the distribution of lower
-bits in the hashed value, the algorithm has changed slightly as of
-5.005_52. When iterating over hashes, this may yield a random order
-that is B<different> from that of previous versions.
+=head2 File and directory handles can be autovivified
-=item C<undef> fails on read only values
+Similar to how constructs such as C<< $x->[0] >> autovivify a reference,
+handle constructors (open(), opendir(), pipe(), socketpair(), sysopen(),
+socket(), and accept()) now autovivify a file or directory handle
+if the handle passed to them is an uninitialized scalar variable. This
+allows the constructs such as C<open(my $fh, ...)> and C<open(local $fh,...)>
+to be used to create filehandles that will conveniently be closed
+automatically when the scope ends, provided there are no other references
+to them. This largely eliminates the need for typeglobs when opening
+filehandles that must be passed around, as in the following example:
-Using the C<undef> operator on a readonly value (such as $1) has
-the same effect as assigning C<undef> to the readonly value--it
-throws an exception.
+ sub myopen {
+ open my $fh, "@_"
+ or die "Can't open '@_': $!";
+ return $fh;
+ }
-=item Close-on-exec bit may be set on pipe and socket handles
+ {
+ my $f = myopen("</etc/motd");
+ print <$f>;
+ # $f implicitly closed here
+ }
-On systems that support a close-on-exec flag on filehandles, the
-flag will be set for any handles created by pipe(), socketpair(),
-socket(), and accept(), if that is warranted by the value of $^F
-that may be in effect. Earlier versions neglected to set the flag
-for handles created with these operators. See L<perlfunc/pipe>,
-L<perlfunc/socketpair>, L<perlfunc/socket>, L<perlfunc/accept>,
-and L<perlvar/$^F>.
+=head2 open() with more than two arguments
-=item Writing C<"$$1"> to mean C<"${$}1"> is unsupported
+If open() is passed three arguments instead of two, the second argument
+is used as the mode and the third argument is taken to be the file name.
+This is primarily useful for protecting against unintended magic behavior
+of the traditional two-argument form. See L<perlfunc/open>.
-Perl 5.004 deprecated the interpretation of C<$$1> and
-similar within interpolated strings to mean C<$$ . "1">,
-but still allowed it.
+=head2 64-bit support
-In Perl 5.6.0 and later, C<"$$1"> always means C<"${$1}">.
+Any platform that has 64-bit integers either
-=item delete(), values() and C<\(%h)> operate on aliases to values, not copies
+ (1) natively as longs or ints
+ (2) via special compiler flags
+ (3) using long long or int64_t
-delete(), each(), values() and hashes in a list context return the actual
-values in the hash, instead of copies (as they used to in earlier
-versions). Typical idioms for using these constructs copy the
-returned values, but this can make a significant difference when
-creating references to the returned values.
+is able to use "quads" (64-bit integers) as follows:
-Keys in the hash are still returned as copies when iterating on
-a hash.
+=over 4
-=item vec(EXPR,OFFSET,BITS) enforces powers-of-two BITS
+=item *
-vec() generates a run-time error if the BITS argument is not
-a valid power-of-two integer.
+constants (decimal, hexadecimal, octal, binary) in the code
-=item Text of some diagnostic output has changed
+=item *
-Most references to internal Perl operations in diagnostics
-have been changed to be more descriptive. This may be an
-issue for programs that may incorrectly rely on the exact
-text of diagnostics for proper functioning.
+arguments to oct() and hex()
-=item C<%@> has been removed
+=item *
-The undocumented special variable C<%@> that used to accumulate
-"background" errors (such as those that happen in DESTROY())
-has been removed, because it could potentially result in memory
-leaks.
+arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)
-=item Parenthesized not() behaves like a list operator
+=item *
-The C<not> operator now falls under the "if it looks like a function,
-it behaves like a function" rule.
+printed as such
-As a result, the parenthesized form can be used with C<grep> and C<map>.
-The following construct used to be a syntax error before, but it works
-as expected now:
+=item *
- grep not($_), @things;
+pack() and unpack() "q" and "Q" formats
-On the other hand, using C<not> with a literal list slice may not
-work. The following previously allowed construct:
+=item *
- print not (1,2,3)[0];
+in basic arithmetics: + - * / % (NOTE: operating close to the limits
+of the integer values may produce surprising results)
-needs to be written with additional parentheses now:
+=item *
- print not((1,2,3)[0]);
+in bit arithmetics: & | ^ ~ << >> (NOTE: these used to be forced
+to be 32 bits wide but now operate on the full native width.)
-The behavior remains unaffected when C<not> is not followed by parentheses.
+=item *
-=item Semantics of bareword prototype C<(*)> have changed
-
-Arguments prototyped as C<*> will now be visible within the subroutine
-as either a simple scalar or as a reference to a typeglob. Perl 5.005
-always coerced simple scalar arguments to a typeglob, which wasn't useful
-in situations where the subroutine must distinguish between a simple
-scalar and a typeglob. See L<perlsub/Prototypes>.
-
-=head2 On 64-bit platforms the semantics of bit operators have changed
-
-If your platform is either natively 64-bit or your Perl has been
-configured to used 64-bit integers, i.e., $Config{ivsize} is 8,
-be warned that the semantics of all the bitwise numeric operators
-(& | ^ ~ << >>) have been changed. These operators used to strictly
-operate on the lower 32 bits of integers, but now operate over the
-entire width of native integers. In particular, note that unary C<~>
-will produce different results on platforms that have different
-$Config{ivsize}. For portability, be sure to mask off the excess bits
-in the result of unary C<~>, e.g., C<~$x & 0xffffffff>.
-
-=head2 More builtins taint their results
-
-The C<passwd> and C<shell> fields returned by the getpwent(), getpwnam(),
-and getpwuid() are now tainted, because the user can affect their own
-encrypted password and login shell.
-
-The variable modified by shmread(), and messages returned by msgrcv()
-(and its object-oriented interface IPC::SysV::Msg::rcv) are also tainted,
-because other untrusted processes can modify messages and shared memory
-segments for their own nefarious purposes.
-
-To avoid these new tainting behaviors, you can build Perl with the
-Configure option C<-Accflags=-DINCOMPLETE_TAINTS>. Beware that the
-ensuing perl binary may be insecure.
-
-=back
-
-=head2 C Source Incompatibilities
-
-=over 4
-
-=item C<PERL_POLLUTE>
-
-Release 5.005 grandfathered old global symbol names by providing preprocessor
-macros for extension source compatibility. As of release 5.6.0, these
-preprocessor definitions are not available by default. You need to explicitly
-compile perl with C<-DPERL_POLLUTE> to get these definitions. For
-extensions still using the old symbols, this option can be
-specified via MakeMaker:
-
- perl Makefile.PL POLLUTE=1
-
-=item C<PERL_IMPLICIT_CONTEXT>
-
- NOTE: PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built
- with one of -Dusethreads, -Dusemultiplicity, or both. It is not
- intended to be enabled by users at this time.
-
-This new build option provides a set of macros for all API functions
-such that an implicit interpreter/thread context argument is passed to
-every API function. As a result of this, something like C<sv_setsv(foo,bar)>
-amounts to a macro invocation that actually translates to something like
-C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
-to not have any significant source compatibility issues, the difference
-between a macro and a real function call will need to be considered.
-
-This means that there B<is> a source compatibility issue as a result of
-this if your extensions attempt to use pointers to any of the Perl API
-functions.
-
-Note that the above issue is not relevant to the default build of
-Perl, whose interfaces continue to match those of prior versions
-(but subject to the other options described here).
-
-See L<perlguts/"The Perl API"> for detailed information on the
-ramifications of building Perl with this option.
-
-=item C<PERL_POLLUTE_MALLOC>
-
-Enabling Perl's malloc in release 5.005 and earlier caused the namespace of
-the system's malloc family of functions to be usurped by the Perl versions,
-since by default they used the same names. Besides causing problems on
-platforms that do not allow these functions to be cleanly replaced, this
-also meant that the system versions could not be called in programs that
-used Perl's malloc. Previous versions of Perl have allowed this behaviour
-to be suppressed with the HIDEMYMALLOC and EMBEDMYMALLOC preprocessor
-definitions.
-
-As of release 5.6.0, Perl's malloc family of functions have default names
-distinct from the system versions. You need to explicitly compile perl with
-C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
-and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
-the default.
-
-Note that these functions do B<not> constitute Perl's memory allocation API.
-See L<perlguts/"Memory Allocation"> for further information about that.
-
-=back
-
-=head2 Compatible C Source API Changes
-
-=over
-
-=item C<PATCHLEVEL> is now C<PERL_VERSION>
-
-The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
-are now available by default from perl.h, and reflect the base revision,
-patchlevel, and subversion respectively. C<PERL_REVISION> had no
-prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
-previously available as C<PATCHLEVEL> and C<SUBVERSION>.
-
-The new names cause less pollution of the B<cpp> namespace and reflect what
-the numbers have come to stand for in common practice. For compatibility,
-the old names are still supported when F<patchlevel.h> is explicitly
-included (as required before), so there is no source incompatibility
-from the change.
+vec()
=back
-=head2 Binary Incompatibilities
-
-In general, the default build of this release is expected to be binary
-compatible for extensions built with the 5.005 release or its maintenance
-versions. However, specific platforms may have broken binary compatibility
-due to changes in the defaults used in hints files. Therefore, please be
-sure to always check the platform-specific README files for any notes to
-the contrary.
-
-The usethreads or usemultiplicity builds are B<not> binary compatible
-with the corresponding builds in 5.005.
-
-On platforms that require an explicit list of exports (AIX, OS/2 and Windows,
-among others), purely internal symbols such as parser functions and the
-run time opcodes are not exported by default. Perl 5.005 used to export
-all functions irrespective of whether they were considered part of the
-public API or not.
-
-For the full list of public API functions, see L<perlapi>.
-
-=head1 Installation and Configuration Improvements
-
-=head2 -Dusethreads means something different
-
- WARNING: Support for threads continues to be an experimental feature.
- Interfaces and implementation are subject to sudden and drastic changes.
-
-The -Dusethreads flag now enables the experimental interpreter-based thread
-support by default. To get the flavor of experimental threads that was in
-5.005 instead, you need to run Configure with "-Dusethreads -Duse5005threads".
-
-As of v5.6.0, interpreter-threads support is still lacking a way to
-create new threads from Perl (i.e., C<use Thread;> will not work with
-interpreter threads). C<use Thread;> continues to be available when you
-specify the -Duse5005threads option to Configure, bugs and all.
-
-=head2 New Configure flags
-
-The following new flags may be enabled on the Configure command line
-by running Configure with C<-Dflag>.
-
- usemultiplicity
- usethreads useithreads (new interpreter threads: no Perl API yet)
- usethreads use5005threads (threads as they were in 5.005)
-
- use64bitint (equal to now deprecated 'use64bits')
- use64bitall
-
- uselongdouble
- usemorebits
- uselargefiles
- usesocks (only SOCKS v5 supported)
-
-=head2 Threadedness and 64-bitness now more daring
+Note that unless you have the case (a) you will have to configure
+and compile Perl using the -Duse64bitint Configure flag.
-The Configure options enabling the use of threads and the use of
-64-bitness are now more daring in the sense that they no more have an
-explicit list of operating systems of known threads/64-bit
-capabilities. In other words: if your operating system has the
-necessary APIs and datatypes, you should be able just to go ahead and
-use them, for threads by Configure -Dusethreads, and for 64 bits
-either explicitly by Configure -Duse64bitint or implicitly if your
-system has 64-bit wide datatypes. See also L<"64-bit support">.
+ NOTE: The Configure flags -Duselonglong and -Duse64bits have been
+ deprecated. Use -Duse64bitint instead.
-=head2 Long Doubles
+There are actually two modes of 64-bitness: the first one is achieved
+using Configure -Duse64bitint and the second one using Configure
+-Duse64bitall. The difference is that the first one is minimal and
+the second one maximal. The first works in more places than the second.
-Some platforms have "long doubles", floating point numbers of even
-larger range than ordinary "doubles". To enable using long doubles for
-Perl's scalars, use -Duselongdouble.
+The C<use64bitint> does only as much as is required to get 64-bit
+integers into Perl (this may mean, for example, using "long longs")
+while your memory may still be limited to 2 gigabytes (because your
+pointers could still be 32-bit). Note that the name C<64bitint> does
+not imply that your C compiler will be using 64-bit C<int>s (it might,
+but it doesn't have to): the C<use64bitint> means that you will be
+able to have 64 bits wide scalar values.
-=head2 -Dusemorebits
+The C<use64bitall> goes all the way by attempting to switch also
+integers (if it can), longs (and pointers) to being 64-bit. This may
+create an even more binary incompatible Perl than -Duse64bitint: the
+resulting executable may not run at all in a 32-bit box, or you may
+have to reboot/reconfigure/rebuild your operating system to be 64-bit
+aware.
-You can enable both -Duse64bitint and -Duselongdouble with -Dusemorebits.
-See also L<"64-bit support">.
+Natively 64-bit systems like Alpha and Cray need neither -Duse64bitint
+nor -Duse64bitall.
-=head2 -Duselargefiles
+Last but not least: note that due to Perl's habit of always using
+floating point numbers, the quads are still not true integers.
+When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned,
+-9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they
+are silently promoted to floating point numbers, after which they will
+start losing precision (in their lower digits).
-Some platforms support system APIs that are capable of handling large files
-(typically, files larger than two gigabytes). Perl will try to use these
-APIs if you ask for -Duselargefiles.
+ NOTE: 64-bit support is still experimental on most platforms.
+ Existing support only covers the LP64 data model. In particular, the
+ LLP64 data model is not yet supported. 64-bit libraries and system
+ APIs on many platforms have not stabilized--your mileage may vary.
-See L<"Large file support"> for more information.
+=head2 Large file support
-=head2 installusrbinperl
+If you have filesystems that support "large files" (files larger than
+2 gigabytes), you may now also be able to create and access them from
+Perl.
-You can use "Configure -Uinstallusrbinperl" which causes installperl
-to skip installing perl also as /usr/bin/perl. This is useful if you
-prefer not to modify /usr/bin for some reason or another but harmful
-because many scripts assume to find Perl in /usr/bin/perl.
+ NOTE: The default action is to enable large file support, if
+ available on the platform.
-=head2 SOCKS support
+If the large file support is on, and you have a Fcntl constant
+O_LARGEFILE, the O_LARGEFILE is automatically added to the flags
+of sysopen().
-You can use "Configure -Dusesocks" which causes Perl to probe
-for the SOCKS proxy protocol library (v5, not v4). For more information
-on SOCKS, see:
+Beware that unless your filesystem also supports "sparse files" seeking
+to umpteen petabytes may be inadvisable.
- http://www.socks.nec.com/
+Note that in addition to requiring a proper file system to do large
+files you may also need to adjust your per-process (or your
+per-system, or per-process-group, or per-user-group) maximum filesize
+limits before running Perl scripts that try to handle large files,
+especially if you intend to write such files.
-=head2 C<-A> flag
+Finally, in addition to your process/process group maximum filesize
+limits, you may have quota limits on your filesystems that stop you
+(your user id or your user group id) from using large files.
-You can "post-edit" the Configure variables using the Configure C<-A>
-switch. The editing happens immediately after the platform specific
-hints files have been processed but before the actual configuration
-process starts. Run C<Configure -h> to find out the full C<-A> syntax.
+Adjusting your process/user/group/file system/operating system limits
+is outside the scope of Perl core language. For process limits, you
+may try increasing the limits using your shell's limits/limit/ulimit
+command before running Perl. The BSD::Resource extension (not
+included with the standard Perl distribution) may also be of use, it
+offers the getrlimit/setrlimit interface that can be used to adjust
+process resource usage limits, including the maximum filesize limit.
-=head2 Enhanced Installation Directories
+=head2 Long doubles
-The installation structure has been enriched to improve the support
-for maintaining multiple versions of perl, to provide locations for
-vendor-supplied modules, scripts, and manpages, and to ease maintenance
-of locally-added modules, scripts, and manpages. See the section on
-Installation Directories in the INSTALL file for complete details.
-For most users building and installing from source, the defaults should
-be fine.
+In some systems you may be able to use long doubles to enhance the
+range and precision of your double precision floating point numbers
+(that is, Perl's numbers). Use Configure -Duselongdouble to enable
+this support (if it is available).
-If you previously used C<Configure -Dsitelib> or C<-Dsitearch> to set
-special values for library directories, you might wish to consider using
-the new C<-Dsiteprefix> setting instead. Also, if you wish to re-use a
-config.sh file from an earlier version of perl, you should be sure to
-check that Configure makes sensible choices for the new directories.
-See INSTALL for complete details.
+=head2 "more bits"
-=head1 Core Changes
+You can "Configure -Dusemorebits" to turn on both the 64-bit support
+and the long double support.
-=head2 Unicode and UTF-8 support
+=head2 Enhanced support for sort() subroutines
- WARNING: This is an experimental feature. Implementation details are
- subject to change.
+Perl subroutines with a prototype of C<($$)>, and XSUBs in general, can
+now be used as sort subroutines. In either case, the two elements to
+be compared are passed as normal parameters in @_. See L<perlfunc/sort>.
-Perl now uses UTF-8 as its internal representation for character
-strings. The C<utf8> and C<bytes> pragmas are used to control this support
-in the current lexical scope. See L<perlunicode>, L<utf8> and L<bytes> for
-more information.
+For unprototyped sort subroutines, the historical behavior of passing
+the elements to be compared as the global variables $a and $b remains
+unchanged.
-=head2 Interpreter cloning, threads, and concurrency
+=head2 C<sort $coderef @foo> allowed
- WARNING: This is an experimental feature. Implementation details are
- subject to change.
+sort() did not accept a subroutine reference as the comparison
+function in earlier versions. This is now permitted.
-Perl 5.005_63 introduces the beginnings of support for running multiple
-interpreters concurrently in different threads. In conjunction with
-the perl_clone() API call, which can be used to selectively duplicate
-the state of any given interpreter, it is possible to compile a
-piece of code once in an interpreter, clone that interpreter
-one or more times, and run all the resulting interpreters in distinct
-threads.
+=head2 File globbing implemented internally
-On Windows, this feature is used to emulate fork() at the interpreter
-level. See L<perlfork>.
+Perl now uses the File::Glob implementation of the glob() operator
+automatically. This avoids using an external csh process and the
+problems associated with it.
-This feature is still in evolution. It is eventually meant to be used
-to selectively clone a subroutine and data reachable from that
-subroutine in a separate interpreter and run the cloned subroutine
-in a separate thread. Since there is no shared data between the
-interpreters, little or no locking will be needed (unless parts of
-the symbol table are explicitly shared). This is obviously intended
-to be an easy-to-use replacement for the existing threads support.
+ NOTE: This is currently an experimental feature. Interfaces and
+ implementation are subject to change.
-Support for cloning interpreters and interpreter concurrency can be
-enabled using the -Dusethreads Configure option (see win32/Makefile for
-how to enable it on Windows.) The resulting perl executable will be
-functionally identical to one that was built with -Dmultiplicity, but
-the perl_clone() API call will only be available in the former.
+=item Support for CHECK blocks
--Dusethreads enables the cpp macro USE_ITHREADS by default, which in turn
-enables Perl source code changes that provide a clear separation between
-the op tree and the data it operates with. The former is immutable, and
-can therefore be shared between an interpreter and all of its clones,
-while the latter is considered local to each interpreter, and is therefore
-copied for each clone.
+In addition to C<BEGIN>, C<INIT>, C<END>, C<DESTROY> and C<AUTOLOAD>,
+subroutines named C<CHECK> are now special. These are queued up during
+compilation and behave similar to END blocks, except they are called at
+the end of compilation rather than at the end of execution. They cannot
+be called directly.
-Note that building Perl with the -Dusemultiplicity Configure option
-is adequate if you wish to run multiple B<independent> interpreters
-concurrently in different threads. -Dusethreads only provides the
-additional functionality of the perl_clone() API call and other
-support for running B<cloned> interpreters concurrently.
+=head2 POSIX character class syntax [: :] supported
-=head2 Lexically scoped warning categories
+For example to match alphabetic characters use /[[:alpha:]]/.
+See L<perlre> for details.
-You can now control the granularity of warnings emitted by perl at a finer
-level using the C<use warnings> pragma. See L<warnings> and L<perllexwarn>
-for details.
+=item Better pseudo-random number generator
-=head2 Lvalue subroutines
+In 5.005_0x and earlier, perl's rand() function used the C library
+rand(3) function. As of 5.005_52, Configure tests for drand48(),
+random(), and rand() (in that order) and picks the first one it finds.
- WARNING: This is an experimental feature. Details are subject to change.
+These changes should result in better random numbers from rand().
-Subroutines can now return modifiable lvalues.
-See L<perlsub/"Lvalue subroutines">.
+=head2 Improved C<qw//> operator
-=head2 "our" declarations
+The C<qw//> operator is now evaluated at compile time into a true list
+instead of being replaced with a run time call to C<split()>. This
+removes the confusing misbehaviour of C<qw//> in scalar context, which
+had inherited that behaviour from split().
-An "our" declaration introduces a value that can be best understood
-as a lexically scoped symbolic alias to a global variable in the
-package that was current where the variable was declared. This is
-mostly useful as an alternative to the C<vars> pragma, but also provides
-the opportunity to introduce typing and other attributes for such
-variables. See L<perlfunc/our>.
+Thus:
-=head2 Support for strings represented as a vector of ordinals
+ $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
-Literals of the form C<v1.2.3.4> are now parsed as a string composed of
-of characters with the specified ordinals. This is an alternative, more
-readable way to construct (possibly unicode) strings instead of
-interpolating characters, as in C<"\x{1}\x{2}\x{3}\x{4}">. The leading
-C<v> may be omitted if there are more than two ordinals, so C<1.2.3> is
-parsed the same as C<v1.2.3>.
+now correctly prints "3|a", instead of "2|a".
-Strings written in this form are also useful to represent version "numbers".
-It is easy to compare such version "numbers" (which are really just plain
-strings) using any of the usual string comparison operators C<eq>, C<ne>,
-C<lt>, C<gt>, etc., or perform bitwise string operations on them using C<|>,
-C<&>, etc.
+=item Better worst-case behavior of hashes
-In conjunction with the new C<$^V> magic variable (which contains
-the perl version as a string), such literals can be used as a readable way
-to check if you're running a particular version of Perl:
+Small changes in the hashing algorithm have been implemented in
+order to improve the distribution of lower order bits in the
+hashed value. This is expected to yield better performance on
+keys that are repeated sequences.
- # this will parse in older versions of Perl also
- if ($^V and $^V gt v5.6.0) {
- # new features supported
- }
+=head2 pack() format 'Z' supported
-C<require> and C<use> also have some special magic to support such literals.
-They will be interpreted as a version rather than as a module name:
+The new format type 'Z' is useful for packing and unpacking null-terminated
+strings. See L<perlfunc/"pack">.
- require v5.6.0; # croak if $^V lt v5.6.0
- use v5.6.0; # same, but croaks at compile-time
+=head2 pack() format modifier '!' supported
-Alternatively, the C<v> may be omitted if there is more than one dot:
+The new format type modifier '!' is useful for packing and unpacking
+native shorts, ints, and longs. See L<perlfunc/"pack">.
- require 5.6.0;
- use 5.6.0;
+=head2 pack() and unpack() support counted strings
-Also, C<sprintf> and C<printf> support the Perl-specific format flag C<%v>
-to print ordinals of characters in arbitrary strings:
+The template character '/' can be used to specify a counted string
+type to be packed or unpacked. See L<perlfunc/"pack">.
- printf "v%vd", $^V; # prints current version, such as "v5.5.650"
- printf "%*vX", ":", $addr; # formats IPv6 address
- printf "%*vb", " ", $bits; # displays bitstring
+=head2 Comments in pack() templates
-See L<perldata/"Scalar value constructors"> for additional information.
+The '#' character in a template introduces a comment up to
+end of the line. This facilitates documentation of pack()
+templates.
=head2 Weak references
- WARNING: This is an experimental feature. Details are subject to change.
-
In previous versions of Perl, you couldn't cache objects so as
to allow them to be deleted if the last reference from outside
the cache is deleted. The reference in the cache would hold a
To use this feature, you need the WeakRef package from CPAN, which
contains additional documentation.
-=head2 File globbing implemented internally
-
- WARNING: This is currently an experimental feature. Interfaces and
- implementation are likely to change.
-
-Perl now uses the File::Glob implementation of the glob() operator
-automatically. This avoids using an external csh process and the
-problems associated with it.
+ NOTE: This is an experimental feature. Details are subject to change.
=head2 Binary numbers supported
$answer = 0b101010;
printf "The answer is: %b\n", oct("0b101010");
+=head2 Lvalue subroutines
+
+Subroutines can now return modifiable lvalues.
+See L<perlsub/"Lvalue subroutines">.
+
+ NOTE: This is an experimental feature. Details are subject to change.
+
=head2 Some arrows may be omitted in calls through references
Perl now allows the arrow to be omitted in many constructs
C<< $foo[10]->{'foo'} >>. Note however, that the arrow is still
required for C<< foo(10)->('bar') >>.
+=head2 Boolean assignment operators are legal lvalues
+
+Constructs such as C<($a ||= 2) += 1> are now allowed.
+
=head2 exists() is supported on subroutine names
The exists() builtin now works on subroutine names. A subroutine
See L<perlfunc/exists> and L<perlfunc/delete> for examples.
-=head2 syswrite() ease-of-use
-
-The length argument of C<syswrite()> has become optional.
-
-=head2 File and directory handles can be autovivified
-
-Similar to how constructs such as C<< $x->[0] >> autovivify a reference,
-handle constructors (open(), opendir(), pipe(), socketpair(), sysopen(),
-socket(), and accept()) now autovivify a file or directory handle
-if the handle passed to them is an uninitialized scalar variable. This
-allows the constructs such as C<open(my $fh, ...)> and C<open(local $fh,...)>
-to be used to create filehandles that will conveniently be closed
-automatically when the scope ends, provided there are no other references
-to them. This largely eliminates the need for typeglobs when opening
-filehandles that must be passed around, as in the following example:
-
- sub myopen {
- open my $fh, "@_"
- or die "Can't open '@_': $!";
- return $fh;
- }
-
- {
- my $f = myopen("</etc/motd");
- print <$f>;
- # $f implicitly closed here
- }
-
-=head2 open() with more than two arguments
-
-If open() is passed three arguments instead of two, the second arguments
-is used as the mode and the third argument is taken to be the file name.
-This is primarily useful for protecting against unintended magic behavior
-of the traditional two-argument form. See L<perlfunc/open>.
-
-=head2 64-bit support
-
- WARNING: 64-bit support is still experimental on most platforms.
- Existing support only covers the LP64 data model. In particular, the
- LLP64 data model is not yet supported. 64-bit libraries and system
- APIs on many platforms have not stabilized--your mileage may vary.
-
-Any platform that has 64-bit integers either
-
- (1) natively as longs or ints
- (2) via special compiler flags
- (3) using long long or int64_t
-
-are able to use "quads" (64-bit integers) as follows:
+=head2 Pseudo-hashes work better
- NOTE: The Configure flags -Duselonglong and -Duse64bits have been
- deprecated. Use -Duse64bitint instead.
+Dereferencing some types of reference values in a pseudo-hash,
+such as C<< $ph->{foo}[1] >>, was accidentally disallowed. This has
+been corrected.
-=over 4
+When applied to a pseudo-hash element, exists() now reports whether
+the specified value exists, not merely if the key is valid.
-=item *
+delete() now works on pseudo-hashes. When given a pseudo-hash element
+or slice it deletes the values corresponding to the keys (but not the keys
+themselves). See L<perlref/"Pseudo-hashes: Using an array as a hash">.
-constants (decimal, hexadecimal, octal, binary) in the code
+Pseudo-hash slices with constant keys are now optimized to array lookups
+at compile-time.
-=item *
+List assignments to pseudo-hash slices are now supported.
-arguments to oct() and hex()
+The C<fields> pragma now provides ways to create pseudo-hashes, via
+fields::new() and fields::phash(). See L<fields>.
-=item *
+ NOTE: The pseudo-hash data type continues to be experimental.
+ Limiting oneself to the interface elements provided by the
+ fields pragma will provide protection from any future changes.
-arguments to print(), printf() and sprintf() (flag prefixes ll, L, q)
+=head2 Automatic flushing of output buffers
-=item *
+fork(), exec(), system(), qx//, and pipe open()s now flush buffers
+of all files opened for output when the operation was attempted. This
+mostly eliminates confusing buffering mishaps suffered by users unaware
+of how Perl internally handles I/O.
-printed as such
+This is not supported on some platforms like Solaris where a suitably
+correct implementation of fflush(NULL) isn't available.
-=item *
+=head2 Better diagnostics on meaningless filehandle operations
-pack() and unpack() "q" and "Q" formats
+Constructs such as C<< open(<FH>) >> and C<< close(<FH>) >>
+are compile time errors. Attempting to read from filehandles that
+were opened only for writing will now produce warnings (just as
+writing to read-only filehandles does).
-=item *
+=head2 Where possible, buffered data discarded from duped input filehandle
-in basic arithmetics: + - * / % (NOTE: operating close to the limits
-of the integer values may produce surprising results)
+C<< open(NEW, "<&OLD") >> now attempts to discard any data that
+was previously read and buffered in C<OLD> before duping the handle.
+On platforms where doing this is allowed, the next read operation
+on C<NEW> will return the same data as the corresponding operation
+on C<OLD>. Formerly, it would have returned the data from the start
+of the following disk block instead.
-=item *
+=head2 eof() has the same old magic as <>
-in bit arithmetics: & | ^ ~ << >> (NOTE: these used to be forced
-to be 32 bits wide but now operate on the full native width.)
+C<eof()> would return true if no attempt to read from C<< <> >> had
+yet been made. C<eof()> has been changed to have a little magic of its
+own, it now opens the C<< <> >> files.
-=item *
+=head2 binmode() can be used to set :crlf and :raw modes
-vec()
+binmode() now accepts a second argument that specifies a discipline
+for the handle in question. The two pseudo-disciplines ":raw" and
+":crlf" are currently supported on DOS-derivative platforms.
+See L<perlfunc/"binmode"> and L<open>.
-=back
+=head2 C<-T> filetest recognizes UTF-8 encoded files as "text"
-Note that unless you have the case (a) you will have to configure
-and compile Perl using the -Duse64bitint Configure flag.
+The algorithm used for the C<-T> filetest has been enhanced to
+correctly identify UTF-8 content as "text".
-There are actually two modes of 64-bitness: the first one is achieved
-using Configure -Duse64bitint and the second one using Configure
--Duse64bitall. The difference is that the first one is minimal and
-the second one maximal. The first works in more places than the second.
+=head2 system(), backticks and pipe open now reflect exec() failure
-The C<use64bitint> does only as much as is required to get 64-bit
-integers into Perl (this may mean, for example, using "long longs")
-while your memory may still be limited to 2 gigabytes (because your
-pointers could still be 32-bit). Note that the name C<64bitint> does
-not imply that your C compiler will be using 64-bit C<int>s (it might,
-but it doesn't have to): the C<use64bitint> means that you will be
-able to have 64 bits wide scalar values.
+On Unix and similar platforms, system(), qx() and open(FOO, "cmd |")
+etc., are implemented via fork() and exec(). When the underlying
+exec() fails, earlier versions did not report the error properly,
+since the exec() happened to be in a different process.
-The C<use64bitall> goes all the way by attempting to switch also
-integers (if it can), longs (and pointers) to being 64-bit. This may
-create an even more binary incompatible Perl than -Duse64bitint: the
-resulting executable may not run at all in a 32-bit box, or you may
-have to reboot/reconfigure/rebuild your operating system to be 64-bit
-aware.
+The child process now communicates with the parent about the
+error in launching the external command, which allows these
+constructs to return with their usual error value and set $!.
-Natively 64-bit systems like Alpha and Cray need neither -Duse64bitint
-nor -Duse64bitall.
+=head2 Improved diagnostics
-Last but not least: note that due to Perl's habit of always using
-floating point numbers, the quads are still not true integers.
-When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned,
--9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they
-are silently promoted to floating point numbers, after which they will
-start losing precision (in their lower digits).
+Line numbers are no longer suppressed (under most likely circumstances)
+during the global destruction phase.
-=head2 Large file support
+Diagnostics emitted from code running in threads other than the main
+thread are now accompanied by the thread ID.
-If you have filesystems that support "large files" (files larger than
-2 gigabytes), you may now also be able to create and access them from
-Perl. NOTE: the default action is to use the large file support, if
-available on the platform.
+Embedded null characters in diagnostics now actually show up. They
+used to truncate the message in prior versions.
-If the large file support is on, and you have a Fcntl constant
-O_LARGEFILE, the O_LARGEFILE is automatically added to the flags
-of sysopen().
+$foo::a and $foo::b are now exempt from "possible typo" warnings only
+if sort() is encountered in package C<foo>.
-Beware: unless your filesystem also supports "sparse files" seeking to
-umpteen petabytes may be unadvisable.
+Unrecognized alphabetic escapes encountered when parsing quote
+constructs now generate a warning, since they may take on new
+semantics in later versions of Perl.
-Note that in addition to requiring a proper file system to do large
-files you may also need to adjust your per-process (or your
-per-system, or per-process-group, or per-user-group) maximum filesize
-limits before running Perl scripts that try to handle large files,
-especially if you intend to write such files.
+Many diagnostics now report the internal operation in which the warning
+was provoked, like so:
-Finally, in addition to your process/process group maximum filesize
-limits, you may have quota limits on your filesystems that stop you
-(your user id or your user group id) from using large files.
+ Use of uninitialized value in concatenation (.) at (eval 1) line 1.
+ Use of uninitialized value in print at (eval 1) line 1.
-Adjusting your process/user/group/file system/operating system limits
-is outside the scope of Perl core language. For process limits, you
-may try increasing the limits using your shell's limits/limit/ulimit
-command before running Perl. The BSD::Resource extension (not
-included with the standard Perl distribution) may also be of use, it
-offers the getrlimit/setrlimit interface that can be used to adjust
-process resource usage limits, including the maximum filesize limit.
+Diagnostics that occur within eval may also report the file and line
+number where the eval is located, in addition to the eval sequence
+number and the line number within the evaluated text itself. For
+example:
-=head2 Long doubles
+ Not enough arguments for scalar at (eval 4)[newlib/perl5db.pl:1411] line 2, at EOF
-In some systems you may be able to use long doubles to enhance the
-range and precision of your double precision floating point numbers
-(that is, Perl's numbers). Use Configure -Duselongdouble to enable
-this support (if it is available).
+=head2 Diagnostics follow STDERR
-=head2 "more bits"
+Diagnostic output now goes to whichever file the C<STDERR> handle
+is pointing at, instead of always going to the underlying C runtime
+library's C<stderr>.
-You can "Configure -Dusemorebits" to turn on both the 64-bit support
-and the long double support.
+=item More consistent close-on-exec behavior
-=head2 Enhanced support for sort() subroutines
+On systems that support a close-on-exec flag on filehandles, the
+flag is now set for any handles created by pipe(), socketpair(),
+socket(), and accept(), if that is warranted by the value of $^F
+that may be in effect. Earlier versions neglected to set the flag
+for handles created with these operators. See L<perlfunc/pipe>,
+L<perlfunc/socketpair>, L<perlfunc/socket>, L<perlfunc/accept>,
+and L<perlvar/$^F>.
-Perl subroutines with a prototype of C<($$)>, and XSUBs in general, can
-now be used as sort subroutines. In either case, the two elements to
-be compared are passed as normal parameters in @_. See L<perlfunc/sort>.
+=head2 syswrite() ease-of-use
-For unprototyped sort subroutines, the historical behavior of passing
-the elements to be compared as the global variables $a and $b remains
-unchanged.
+The length argument of C<syswrite()> has become optional.
=head2 Better syntax checks on parenthesized unary operators
remains unchanged. See L<perlop>.
-=head2 POSIX character class syntax [: :] supported
-
-For example to match alphabetic characters use /[[:alpha:]]/.
-See L<perlre> for details.
-
-=head2 Improved C<qw//> operator
-
-The C<qw//> operator is now evaluated at compile time into a true list
-instead of being replaced with a run time call to C<split()>. This
-removes the confusing misbehaviour of C<qw//> in scalar context, which
-had inherited that behaviour from split().
-
-Thus:
+=head2 Bit operators support full native integer width
- $foo = ($bar) = qw(a b c); print "$foo|$bar\n";
+The bit operators (& | ^ ~ << >>) now operate on the full native
+integral width (the exact size of which is available in $Config{ivsize}).
+For example, if your platform is either natively 64-bit or if Perl
+has been configured to use 64-bit integers, these operations apply
+to 8 bytes (as opposed to 4 bytes on 32-bit platforms).
+For portability, be sure to mask off the excess bits in the result of
+unary C<~>, e.g., C<~$x & 0xffffffff>.
-now correctly prints "3|a", instead of "2|a".
+=head2 Improved security features
-=head2 pack() format 'Z' supported
+More potentially unsafe operations taint their results for improved
+security.
-The new format type 'Z' is useful for packing and unpacking null-terminated
-strings. See L<perlfunc/"pack">.
+The C<passwd> and C<shell> fields returned by the getpwent(), getpwnam(),
+and getpwuid() are now tainted, because the user can affect their own
+encrypted password and login shell.
-=head2 pack() format modifier '!' supported
+The variable modified by shmread(), and messages returned by msgrcv()
+(and its object-oriented interface IPC::SysV::Msg::rcv) are also tainted,
+because other untrusted processes can modify messages and shared memory
+segments for their own nefarious purposes.
-The new format type modifier '!' is useful for packing and unpacking
-native shorts, ints, and longs. See L<perlfunc/"pack">.
+=item More functional bareword prototype (*)
-=head2 pack() and unpack() support counted strings
+Bareword prototypes have been rationalized to enable them to be used
+to override builtins that accept barewords and interpret them in
+a special way, such as C<require> or C<do>.
-The template character '/' can be used to specify a counted string
-type to be packed or unpacked. See L<perlfunc/"pack">.
+Arguments prototyped as C<*> will now be visible within the subroutine
+as either a simple scalar or as a reference to a typeglob.
+See L<perlsub/Prototypes>.
-=head2 Comments in pack() templates
+=head2 C<require> and C<do> may be overridden
-The '#' character in a template introduces a comment up to
-end of the line. This facilitates documentation of pack()
-templates.
+C<require> and C<do 'file'> operations may be overridden locally
+by importing subroutines of the same name into the current package
+(or globally by importing them into the CORE::GLOBAL:: namespace).
+Overriding C<require> will also affect C<use>, provided the override
+is visible at compile-time.
+See L<perlsub/"Overriding Built-in Functions">.
=head2 $^X variables may now have names longer than one character
C<^_>, which may be used by user programs and are guaranteed not to
acquire special meaning in any future version of Perl.
-=head2 C<use attrs> implicit in subroutine attributes
-
-Formerly, if you wanted to mark a subroutine as being a method call or
-as requiring an automatic lock() when it is entered, you had to declare
-that with a C<use attrs> pragma in the body of the subroutine.
-That can now be accomplished with declaration syntax, like this:
-
- sub mymethod : locked method ;
- ...
- sub mymethod : locked method {
- ...
- }
-
- sub othermethod :locked :method ;
- ...
- sub othermethod :locked :method {
- ...
- }
-
-
-(Note how only the first C<:> is mandatory, and whitespace surrounding
-the C<:> is optional.)
-
-F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes
-with the stubs they provide. See L<attributes>.
-
-=head2 Support for interpolating named characters
-
-The new C<\N> escape interpolates named characters within strings.
-For example, C<"Hi! \N{WHITE SMILING FACE}"> evaluates to a string
-with a unicode smiley face at the end.
-
-=head2 C<require> and C<do> may be overridden
-
-C<require> and C<do 'file'> operations may be overridden locally
-by importing subroutines of the same name into the current package
-(or globally by importing them into the CORE::GLOBAL:: namespace).
-Overriding C<require> will also affect C<use>, provided the override
-is visible at compile-time.
-See L<perlsub/"Overriding Built-in Functions">.
-
=head2 New variable $^C reflects C<-c> switch
C<$^C> has a boolean value that reflects whether perl is being run
This behavior must be specifically enabled when running Configure.
See F<INSTALL> and F<README.Y2K>.
-=head1 Significant bug fixes
-
-=head2 <HANDLE> on empty files
-
-With C<$/> set to C<undef>, "slurping" an empty file returns a string of
-zero length (instead of C<undef>, as it used to) the first time the
-HANDLE is read after C<$/> is set to C<undef>. Further reads yield
-C<undef>.
-
-This means that the following will append "foo" to an empty file (it used
-to do nothing):
-
- perl -0777 -pi -e 's/^/foo/' empty_file
-
-The behaviour of:
-
- perl -pi -e 's/^/foo/' empty_file
-
-is unchanged (it continues to leave the file empty).
+=head1 Modules and Pragmata
-=head2 C<eval '...'> improvements
+=head2 Modules
-Line numbers (as reflected by caller() and most diagnostics) within
-C<eval '...'> were often incorrect where here documents were involved.
-This has been corrected.
+=over 4
-Lexical lookups for variables appearing in C<eval '...'> within
-functions that were themselves called within an C<eval '...'> were
-searching the wrong place for lexicals. The lexical search now
-correctly ends at the subroutine's block boundary.
+=item attributes
-Parsing of here documents used to be flawed when they appeared as
-the replacement expression in C<eval 's/.../.../e'>. This has
-been fixed.
+While used internally by Perl as a pragma, this module also
+provides a way to fetch subroutine and variable attributes.
+See L<attributes>.
-=head2 All compilation errors are true errors
+=item B
-Some "errors" encountered at compile time were by neccessity
-generated as warnings followed by eventual termination of the
-program. This enabled more such errors to be reported in a
-single run, rather than causing a hard stop at the first error
-that was encountered.
+The Perl Compiler suite has been extensively reworked for this
+release. More of the standard Perl testsuite passes when run
+under the Compiler, but there is still a significant way to
+go to achieve production quality compiled executables.
-The mechanism for reporting such errors has been reimplemented
-to queue compile-time errors and report them at the end of the
-compilation as true errors rather than as warnings. This fixes
-cases where error messages leaked through in the form of warnings
-when code was compiled at run time using C<eval STRING>, and
-also allows such errors to be reliably trapped using C<eval "...">.
+ NOTE: The Compiler suite remains highly experimental. The
+ generated code may not be correct, even it manages to execute
+ without errors.
-=head2 Automatic flushing of output buffers
+=item Benchmark
-fork(), exec(), system(), qx//, and pipe open()s now flush buffers
-of all files opened for output when the operation was attempted. This
-mostly eliminates confusing buffering mishaps suffered by users unaware
-of how Perl internally handles I/O.
+Overall, Benchmark results exhibit lower average error and better timing
+accuracy.
-This is not supported on some platforms like Solaris where a suitably
-correct implementation of fflush(NULL) isn't available.
+You can now run tests for I<n> seconds instead of guessing the right
+number of tests to run: e.g., timethese(-5, ...) will run each
+code for at least 5 CPU seconds. Zero as the "number of repetitions"
+means "for at least 3 CPU seconds". The output format has also
+changed. For example:
-=head2 Better diagnostics on meaningless filehandle operations
+ use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
-Constructs such as C<< open(<FH>) >> and C<< close(<FH>) >>
-are compile time errors. Attempting to read from filehandles that
-were opened only for writing will now produce warnings (just as
-writing to read-only filehandles does).
+will now output something like this:
-=head2 Where possible, buffered data discarded from duped input filehandle
+ Benchmark: running a, b, each for at least 5 CPU seconds...
+ a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
+ b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
-C<< open(NEW, "<&OLD") >> now attempts to discard any data that
-was previously read and buffered in C<OLD> before duping the handle.
-On platforms where doing this is allowed, the next read operation
-on C<NEW> will return the same data as the corresponding operation
-on C<OLD>. Formerly, it would have returned the data from the start
-of the following disk block instead.
+New features: "each for at least N CPU seconds...", "wallclock secs",
+and the "@ operations/CPU second (n=operations)".
-=head2 eof() has the same old magic as <>
+timethese() now returns a reference to a hash of Benchmark objects containing
+the test results, keyed on the names of the tests.
-C<eof()> would return true if no attempt to read from C<< <> >> had
-yet been made. C<eof()> has been changed to have a little magic of its
-own, it now opens the C<< <> >> files.
+timethis() now returns the iterations field in the Benchmark result object
+instead of 0.
-=head2 system(), backticks and pipe open now reflect exec() failure
+timethese(), timethis(), and the new cmpthese() (see below) can also take
+a format specifier of 'none' to suppress output.
-On Unix and similar platforms, system(), qx() and open(FOO, "cmd |")
-etc., are implemented via fork() and exec(). When the underlying
-exec() fails, earlier versions did not report the error properly,
-since the exec() happened to be in a different process.
+A new function countit() is just like timeit() except that it takes a
+TIME instead of a COUNT.
-The child process now communicates with the parent about the
-error in launching the external command, which allows these
-constructs to return with their usual error value and set $!.
+A new function cmpthese() prints a chart comparing the results of each test
+returned from a timethese() call. For each possible pair of tests, the
+percentage speed difference (iters/sec or seconds/iter) is shown.
-=head2 Implicitly closed filehandles are safer
+For other details, see L<Benchmark>.
-Sometimes implicitly closed filehandles (as when they are localized,
-and Perl automatically closes them on exiting the scope) could
-inadvertently set $? or $!. This has been corrected.
+=item ByteLoader
-=head2 C<(\$)> prototype and C<$foo{a}>
+The ByteLoader is a dedicated extension to generate and run
+Perl bytecode. See L<ByteLoader>.
-A scalar reference prototype now correctly allows a hash or
-array element in that slot.
+=item constant
-=head2 Pseudo-hashes work better
+References can now be used.
- WARNING: The pseudo-hash data type continues to be experimental.
- Limiting oneself to the interface elements provided by the
- fields pragma will provide protection from any future changes.
+The new version also allows a leading underscore in constant names, but
+disallows a double leading underscore (as in "__LINE__"). Some other names
+are disallowed or warned against, including BEGIN, END, etc. Some names
+which were forced into main:: used to fail silently in some cases; now they're
+fatal (outside of main::) and an optional warning (inside of main::).
+The ability to detect whether a constant had been set with a given name has
+been added.
-Dereferencing some types of reference values in a pseudo-hash,
-such as C<< $ph->{foo}[1] >>, was accidentally disallowed. This has
-been corrected.
+See L<constant>.
-When applied to a pseudo-hash element, exists() now reports whether
-the specified value exists, not merely if the key is valid.
+=item charnames
-delete() now works on pseudo-hashes. When given a pseudo-hash element
-or slice it deletes the values corresponding to the keys (but not the keys
-themselves). See L<perlref/"Pseudo-hashes: Using an array as a hash">.
+This pragma implements the C<\N> string escape. See L<charnames>.
-Pseudo-hash slices with constant keys are now optimized to array lookups
-at compile-time.
+=item Data::Dumper
-The C<fields> pragma now provides ways to create pseudo-hashes, via
-fields::new() and fields::phash(). See L<fields>.
+A C<Maxdepth> setting can be specified to avoid venturing
+too deeply into deep data structures. See L<Data::Dumper>.
-=head2 C<goto &sub> and AUTOLOAD
+The XSUB implementation of Dump() is now automatically called if the
+C<Useqq> setting is not in use.
-The C<goto &sub> construct works correctly when C<&sub> happens
-to be autoloaded.
+Dumping C<qr//> objects works correctly.
-=head2 C<-bareword> allowed under C<use integer>
+=item DB
-The autoquoting of barewords preceded by C<-> did not work
-in prior versions when the C<integer> pragma was enabled.
-This has been fixed.
+C<DB> is an experimental module that exposes a clean abstraction
+to Perl's debugging API.
-=head2 Boolean assignment operators are legal lvalues
+=item DB_File
-Constructs such as C<($a ||= 2) += 1> are now allowed.
+DB_File can now be built with Berkeley DB versions 1, 2 or 3.
+See C<ext/DB_File/Changes>.
-=head2 C<sort $coderef @foo> allowed
+=item Devel::DProf
-sort() did not accept a subroutine reference as the comparison
-function in earlier versions. This is now permitted.
+Devel::DProf, a Perl source code profiler has been added. See
+L<Devel::DProf> and L<dprofpp>.
-=head2 Failures in DESTROY()
+=item Devel::Peek
-When code in a destructor threw an exception, it went unnoticed
-in earlier versions of Perl, unless someone happened to be
-looking in $@ just after the point the destructor happened to
-run. Such failures are now visible as warnings when warnings are
-enabled.
+The Devel::Peek module provides access to the internal representation
+of Perl variables and data. It is a data debugging tool for the XS programmer.
-=head2 Locale bugs fixed
+=item Dumpvalue
-printf() and sprintf() previously reset the numeric locale
-back to the default "C" locale. This has been fixed.
+The Dumpvalue module provides screen dumps of Perl data.
-Numbers formatted according to the local numeric locale
-(such as using a decimal comma instead of a decimal dot) caused
-"isn't numeric" warnings, even while the operations accessing
-those numbers produced correct results. These warnings have been
-discontinued.
+=item DynaLoader
-=head2 Memory leaks
+DynaLoader now supports a dl_unload_file() function on platforms that
+support unloading shared objects using dlclose().
-The C<eval 'return sub {...}'> construct could sometimes leak
-memory. This has been fixed.
+Perl can also optionally arrange to unload all extension shared objects
+loaded by Perl. To enable this, build Perl with the Configure option
+C<-Accflags=-DDL_UNLOAD_ALL_AT_EXIT>. (This maybe useful if you are
+using Apache with mod_perl.)
-Operations that aren't filehandle constructors used to leak memory
-when used on invalid filehandles. This has been fixed.
+=item English
-Constructs that modified C<@_> could fail to deallocate values
-in C<@_> and thus leak memory. This has been corrected.
+$PERL_VERSION now stands for C<$^V> (a string value) rather than for C<$]>
+(a numeric value).
-=head2 Spurious subroutine stubs after failed subroutine calls
+=item Env
-Perl could sometimes create empty subroutine stubs when a
-subroutine was not found in the package. Such cases stopped
-later method lookups from progressing into base packages.
-This has been corrected.
+Env now supports accessing environment variables like PATH as array
+variables.
-=head2 Taint failures under C<-U>
+=item Fcntl
-When running in unsafe mode, taint violations could sometimes
-cause silent failures. This has been fixed.
+More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
+large file (more than 4GB) access (NOTE: the O_LARGEFILE is
+automatically added to sysopen() flags if large file support has been
+configured, as is the default), Free/Net/OpenBSD locking behaviour
+flags F_FLOCK, F_POSIX, Linux F_SHLCK, and O_ACCMODE: the combined
+mask of O_RDONLY, O_WRONLY, and O_RDWR. The seek()/sysseek()
+constants SEEK_SET, SEEK_CUR, and SEEK_END are available via the
+C<:seek> tag. The chmod()/stat() S_IF* constants and S_IS* functions
+are available via the C<:mode> tag.
-=head2 END blocks and the C<-c> switch
+=item File::Compare
-Prior versions used to run BEGIN B<and> END blocks when Perl was
-run in compile-only mode. Since this is typically not the expected
-behavior, END blocks are not executed anymore when the C<-c> switch
-is used.
+A compare_text() function has been added, which allows custom
+comparison functions. See L<File::Compare>.
-See L<CHECK blocks> for how to run things when the compile phase ends.
+=item File::Find
-=head2 Potential to leak DATA filehandles
+File::Find now works correctly when the wanted() function is either
+autoloaded or is a symbolic reference.
-Using the C<__DATA__> token creates an implicit filehandle to
-the file that contains the token. It is the program's
-responsibility to close it when it is done reading from it.
+A bug that caused File::Find to lose track of the working directory
+when pruning top-level directories has been fixed.
-This caveat is now better explained in the documentation.
-See L<perldata>.
+File::Find now also supports several other options to control its
+behavior. It can follow symbolic links if the C<follow> option is
+specified. Enabling the C<no_chdir> option will make File::Find skip
+changing the current directory when walking directories. The C<untaint>
+flag can be useful when running with taint checks enabled.
-=head2 Diagnostics follow STDERR
+See L<File::Find>.
-Diagnostic output now goes to whichever file the C<STDERR> handle
-is pointing at, instead of always going to the underlying C runtime
-library's C<stderr>.
+=item File::Glob
-=head2 Other fixes for better diagnostics
+This extension implements BSD-style file globbing. By default,
+it will also be used for the internal implementation of the glob()
+operator. See L<File::Glob>.
-Line numbers are no longer suppressed (under most likely circumstances)
-during the global destruction phase.
+=item File::Spec
-Diagnostics emitted from code running in threads other than the main
-thread are now accompanied by the thread ID.
+New methods have been added to the File::Spec module: devnull() returns
+the name of the null device (/dev/null on Unix) and tmpdir() the name of
+the temp directory (normally /tmp on Unix). There are now also methods
+to convert between absolute and relative filenames: abs2rel() and
+rel2abs(). For compatibility with operating systems that specify volume
+names in file paths, the splitpath(), splitdir(), and catdir() methods
+have been added.
-Embedded null characters in diagnostics now actually show up. They
-used to truncate the message in prior versions.
+=item File::Spec::Functions
-$foo::a and $foo::b are now exempt from "possible typo" warnings only
-if sort() is encountered in package C<foo>.
+The new File::Spec::Functions modules provides a function interface
+to the File::Spec module. Allows shorthand
-Unrecognized alphabetic escapes encountered when parsing quote
-constructs now generate a warning, since they may take on new
-semantics in later versions of Perl.
+ $fullname = catfile($dir1, $dir2, $file);
-Many diagnostics now report the internal operation in which the warning
-was provoked, like so:
+instead of
- Use of uninitialized value in concatenation (.) at (eval 1) line 1.
- Use of uninitialized value in print at (eval 1) line 1.
+ $fullname = File::Spec->catfile($dir1, $dir2, $file);
-Diagnostics that occur within eval may also report the file and line
-number where the eval is located, in addition to the eval sequence
-number and the line number within the evaluated text itself. For
-example:
+=item Getopt::Long
- Not enough arguments for scalar at (eval 4)[newlib/perl5db.pl:1411] line 2, at EOF
+Getopt::Long licensing has changed to allow the Perl Artistic License
+as well as the GPL. It used to be GPL only, which got in the way of
+non-GPL applications that wanted to use Getopt::Long.
-=head1 Performance enhancements
+Getopt::Long encourages the use of Pod::Usage to produce help
+messages. For example:
-=head2 Simple sort() using { $a <=> $b } and the like are optimized
+ use Getopt::Long;
+ use Pod::Usage;
+ my $man = 0;
+ my $help = 0;
+ GetOptions('help|?' => \$help, man => \$man) or pod2usage(2);
+ pod2usage(1) if $help;
+ pod2usage(-exitstatus => 0, -verbose => 2) if $man;
+
+ __END__
-Many common sort() operations using a simple inlined block are now
-optimized for faster performance.
+ =head1 NAME
-=head2 Optimized assignments to lexical variables
+ sample - Using GetOpt::Long and Pod::Usage
-Certain operations in the RHS of assignment statements have been
-optimized to directly set the lexical variable on the LHS,
-eliminating redundant copying overheads.
+ =head1 SYNOPSIS
-=head2 Faster subroutine calls
+ sample [options] [file ...]
-Minor changes in how subroutine calls are handled internally
-provide marginal improvements in performance.
+ Options:
+ -help brief help message
+ -man full documentation
-=head1 Platform specific changes
+ =head1 OPTIONS
-=head2 Supported platforms
+ =over 8
-=over 4
+ =item B<-help>
-=item *
+ Print a brief help message and exits.
-VM/ESA is now supported.
+ =item B<-man>
-=item *
+ Prints the manual page and exits.
-Siemens BS2000 is now supported under the POSIX Shell.
+ =back
-=item *
+ =head1 DESCRIPTION
-The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
-extension.
+ B<This program> will read the given input file(s) and do someting
+ useful with the contents thereof.
-=item *
+ =cut
-GNU/Hurd is now supported.
+See L<Pod::Usage> for details.
-=item *
+A bug that prevented the non-option call-back <> from being
+specified as the first argument has been fixed.
-Rhapsody/Darwin is now supported.
+To specify the characters < and > as option starters, use ><. Note,
+however, that changing option starters is strongly deprecated.
-=item *
+=item IO
-EPOC is is now supported (on Psion 5).
+write() and syswrite() will now accept a single-argument
+form of the call, for consistency with Perl's syswrite().
-=back
+You can now create a TCP-based IO::Socket::INET without forcing
+a connect attempt. This allows you to configure its options
+(like making it non-blocking) and then call connect() manually.
-=head2 DOS
+A bug that prevented the IO::Socket::protocol() accessor
+from ever returning the correct value has been corrected.
-=over 4
+IO::Socket::connect now uses non-blocking IO instead of alarm()
+to do connect timeouts.
-=item *
+IO::Socket::accept now uses select() instead of alarm() for doing
+timeouts.
-Perl now works with djgpp 2.02 (and 2.03 alpha).
+IO::Socket::INET->new now sets $! correctly on failure. $@ is
+still set for backwards compatability.
-=item *
+=item JPL
-Environment variable names are not converted to uppercase any more.
+Java Perl Lingo is now distributed with Perl. See jpl/README
+for more information.
-=item *
+=item lib
-Incorrect exit codes from backticks have been fixed.
+C<use lib> now weeds out any trailing duplicate entries.
+C<no lib> removes all named entries.
-=item *
+=item Math::BigInt
-This port continues to use its own builtin globbing (not File::Glob).
+The bitwise operations C<<< << >>>, C<<< >> >>>, C<&>, C<|>,
+and C<~> are now supported on bigints.
-=back
+=item Math::Complex
-=head2 OS390 (OpenEdition MVS)
+The accessor methods Re, Im, arg, abs, rho, and theta can now also
+act as mutators (accessor $z->Re(), mutator $z->Re(3)).
-Support for this EBCDIC platform has not been renewed in this release.
-There are difficulties in reconciling Perl's standardization on UTF-8
-as its internal representation for characters with the EBCDIC character
-set, because the two are incompatible.
+The class method C<display_format> and the corresponding object method
+C<display_format>, in addition to accepting just one argument, now can
+also accept a parameter hash. Recognized keys of a parameter hash are
+C<"style">, which corresponds to the old one parameter case, and two
+new parameters: C<"format">, which is a printf()-style format string
+(defaults usually to C<"%.15g">, you can revert to the default by
+setting the format string to C<undef>) used for both parts of a
+complex number, and C<"polar_pretty_print"> (defaults to true),
+which controls whether an attempt is made to try to recognize small
+multiples and rationals of pi (2pi, pi/2) at the argument (angle) of a
+polar complex number.
-It is unclear whether future versions will renew support for this
-platform, but the possibility exists.
+The potentially disruptive change is that in list context both methods
+now I<return the parameter hash>, instead of only the value of the
+C<"style"> parameter.
-=head2 VMS
+=item Math::Trig
-Numerous revisions and extensions to configuration, build, testing, and
-installation process to accomodate core changes and VMS-specific options.
+A little bit of radial trigonometry (cylindrical and spherical),
+radial coordinate conversions, and the great circle distance were added.
-Expand %ENV-handling code to allow runtime mapping to logical names,
-CLI symbols, and CRTL environ array.
+=item Pod::Parser, Pod::InputObjects
-Extension of subprocess invocation code to accept filespecs as command
-"verbs".
+Pod::Parser is a base class for parsing and selecting sections of
+pod documentation from an input stream. This module takes care of
+identifying pod paragraphs and commands in the input and hands off the
+parsed paragraphs and commands to user-defined methods which are free
+to interpret or translate them as they see fit.
-Add to Perl command line processing the ability to use default file types and
-to recognize Unix-style C<2E<gt>&1>.
+Pod::InputObjects defines some input objects needed by Pod::Parser, and
+for advanced users of Pod::Parser that need more about a command besides
+its name and text.
-Expansion of File::Spec::VMS routines, and integration into ExtUtils::MM_VMS.
+As of release 5.6.0 of Perl, Pod::Parser is now the officially sanctioned
+"base parser code" recommended for use by all pod2xxx translators.
+Pod::Text (pod2text) and Pod::Man (pod2man) have already been converted
+to use Pod::Parser and efforts to convert Pod::HTML (pod2html) are already
+underway. For any questions or comments about pod parsing and translating
+issues and utilities, please use the pod-people@perl.org mailing list.
-Extension of ExtUtils::MM_VMS to handle complex extensions more flexibly.
+For further information, please see L<Pod::Parser> and L<Pod::InputObjects>.
-Barewords at start of Unix-syntax paths may be treated as text rather than
-only as logical names.
+=item Pod::Checker, podchecker
-Optional secure translation of several logical names used internally by Perl.
+This utility checks pod files for correct syntax, according to
+L<perlpod>. Obvious errors are flagged as such, while warnings are
+printed for mistakes that can be handled gracefully. The checklist is
+not complete yet. See L<Pod::Checker>.
-Miscellaneous bugfixing and porting of new core code to VMS.
+=item Pod::ParseUtils, Pod::Find
-Thanks are gladly extended to the many people who have contributed VMS
-patches, testing, and ideas.
+These modules provide a set of gizmos that are useful mainly for pod
+translators. L<Pod::Find|Pod::Find> traverses directory structures and
+returns found pod files, along with their canonical names (like
+C<File::Spec::Unix>). L<Pod::ParseUtils|Pod::ParseUtils> contains
+B<Pod::List> (useful for storing pod list information), B<Pod::Hyperlink>
+(for parsing the contents of C<LE<lt>E<gt>> sequences) and B<Pod::Cache>
+(for caching information about pod files, e.g., link nodes).
-=head2 Win32
+=item Pod::Select, podselect
-Perl can now emulate fork() internally, using multiple interpreters running
-in different concurrent threads. This support must be enabled at build
-time. See L<perlfork> for detailed information.
+Pod::Select is a subclass of Pod::Parser which provides a function
+named "podselect()" to filter out user-specified sections of raw pod
+documentation from an input stream. podselect is a script that provides
+access to Pod::Select from other scripts to be used as a filter.
+See L<Pod::Select>.
-When given a pathname that consists only of a drivename, such as C<A:>,
-opendir() and stat() now use the current working directory for the drive
-rather than the drive root.
+=item Pod::Usage, pod2usage
-The builtin XSUB functions in the Win32:: namespace are documented. See
-L<Win32>.
+Pod::Usage provides the function "pod2usage()" to print usage messages for
+a Perl script based on its embedded pod documentation. The pod2usage()
+function is generally useful to all script authors since it lets them
+write and maintain a single source (the pods) for documentation, thus
+removing the need to create and maintain redundant usage message text
+consisting of information already in the pods.
-$^X now contains the full path name of the running executable.
+There is also a pod2usage script which can be used from other kinds of
+scripts to print usage messages from pods (even for non-Perl scripts
+with pods embedded in comments).
-A Win32::GetLongPathName() function is provided to complement
-Win32::GetFullPathName() and Win32::GetShortPathName(). See L<Win32>.
+For details and examples, please see L<Pod::Usage>.
-POSIX::uname() is supported.
+=item Pod::Text and Pod::Man
-system(1,...) now returns true process IDs rather than process
-handles. kill() accepts any real process id, rather than strictly
-return values from system(1,...).
+Pod::Text has been rewritten to use Pod::Parser. While pod2text() is
+still available for backwards compatibility, the module now has a new
+preferred interface. See L<Pod::Text> for the details. The new Pod::Text
+module is easily subclassed for tweaks to the output, and two such
+subclasses (Pod::Text::Termcap for man-page-style bold and underlining
+using termcap information, and Pod::Text::Color for markup with ANSI color
+sequences) are now standard.
-For better compatibility with Unix, C<kill(0, $pid)> can now be used to
-test whether a process exists.
+pod2man has been turned into a module, Pod::Man, which also uses
+Pod::Parser. In the process, several outstanding bugs related to quotes
+in section headers, quoting of code escapes, and nested lists have been
+fixed. pod2man is now a wrapper script around this module.
-The C<Shell> module is supported.
+=item SDBM_File
-Better support for building Perl under command.com in Windows 95
-has been added.
+An EXISTS method has been added to this module (and sdbm_exists() has
+been added to the underlying sdbm library), so one can now call exists
+on an SDBM_File tied hash and get the correct result, rather than a
+runtime error.
-Scripts are read in binary mode by default to allow ByteLoader (and
-the filter mechanism in general) to work properly. For compatibility,
-the DATA filehandle will be set to text mode if a carriage return is
-detected at the end of the line containing the __END__ or __DATA__
-token; if not, the DATA filehandle will be left open in binary mode.
-Earlier versions always opened the DATA filehandle in text mode.
+A bug that may have caused data loss when more than one disk block
+happens to be read from the database in a single FETCH() has been
+fixed.
-The glob() operator is implemented via the C<File::Glob> extension,
-which supports glob syntax of the C shell. This increases the flexibility
-of the glob() operator, but there may be compatibility issues for
-programs that relied on the older globbing syntax. If you want to
-preserve compatibility with the older syntax, you might want to run
-perl with C<-MFile::DosGlob>. For details and compatibility information,
-see L<File::Glob>.
+=item Sys::Syslog
-=head1 New tests
+Sys::Syslog now uses XSUBs to access facilities from syslog.h so it
+no longer requires syslog.ph to exist.
-=over 4
+=item Sys::Hostname
-=item lib/attrs
+Sys::Hostname now uses XSUBs to call the C library's gethostname() or
+uname() if they exist.
-Compatibility tests for C<sub : attrs> vs the older C<use attrs>.
+=item Term::ANSIColor
-=item lib/env
+Term::ANSIColor is a very simple module to provide easy and readable
+access to the ANSI color and highlighting escape sequences, supported by
+most ANSI terminal emulators. It is now included standard.
-Tests for new environment scalar capability (e.g., C<use Env qw($BAR);>).
+=item Time::Local
-=item lib/env-array
+The timelocal() and timegm() functions used to silently return bogus
+results when the date fell outside the machine's integer range. They
+now consistently croak() if the date falls in an unsupported range.
-Tests for new environment array capability (e.g., C<use Env qw(@PATH);>).
+=item Win32
-=item lib/io_const
+The error return value in list context has been changed for all functions
+that return a list of values. Previously these functions returned a list
+with a single element C<undef> if an error occurred. Now these functions
+return the empty list in these situations. This applies to the following
+functions:
-IO constants (SEEK_*, _IO*).
+ Win32::FsType
+ Win32::GetOSVersion
-=item lib/io_dir
+The remaining functions are unchanged and continue to return C<undef> on
+error even in list context.
-Directory-related IO methods (new, read, close, rewind, tied delete).
+The Win32::SetLastError(ERROR) function has been added as a complement
+to the Win32::GetLastError() function.
-=item lib/io_multihomed
+The new Win32::GetFullPathName(FILENAME) returns the full absolute
+pathname for FILENAME in scalar context. In list context it returns
+a two-element list containing the fully qualified directory name and
+the filename. See L<Win32>.
-INET sockets with multi-homed hosts.
+=item XSLoader
-=item lib/io_poll
+The XSLoader extension is a simpler alternative to DynaLoader.
+See L<XSLoader>.
-IO poll().
+=item DBM Filters
-=item lib/io_unix
+A new feature called "DBM Filters" has been added to all the
+DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
+DBM Filters add four new methods to each DBM module:
-UNIX sockets.
+ filter_store_key
+ filter_store_value
+ filter_fetch_key
+ filter_fetch_value
-=item op/attrs
+These can be used to filter key-value pairs before the pairs are
+written to the database or just after they are read from the database.
+See L<perldbmfilter> for further information.
-Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>.
+=back
-=item op/filetest
+=head2 Pragmata
-File test operators.
+C<use attrs> is now obsolete, and is only provided for
+backward-compatibility. It's been replaced by the C<sub : attributes>
+syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>.
-=item op/lex_assign
+Lexical warnings pragma, C<use warnings;>, to control optional warnings.
+See L<perllexwarn>.
-Verify operations that access pad objects (lexicals and temporaries).
+C<use filetest> to control the behaviour of filetests (C<-r> C<-w>
+...). Currently only one subpragma implemented, "use filetest
+'access';", that uses access(2) or equivalent to check permissions
+instead of using stat(2) as usual. This matters in filesystems
+where there are ACLs (access control lists): the stat(2) might lie,
+but access(2) knows better.
-=item op/exists_sub
+The C<open> pragma can be used to specify default disciplines for
+handle constructors (e.g. open()) and for qx//. The two
+pseudo-disciplines C<:raw> and C<:crlf> are currently supported on
+DOS-derivative platforms (i.e. where binmode is not a no-op).
+See also L</"binmode() can be used to set :crlf and :raw modes">.
-Verify C<exists &sub> operations.
+=head1 Utility Changes
-=back
+=head2 dprofpp
-=head1 Modules and Pragmata
+C<dprofpp> is used to display profile data generated using C<Devel::DProf>.
+See L<dprofpp>.
-=head2 Modules
+=head2 find2perl
-=over 4
+The C<find2perl> utility now uses the enhanced features of the File::Find
+module. The -depth and -follow options are supported. Pod documentation
+is also included in the script.
-=item attributes
+=head2 h2xs
-While used internally by Perl as a pragma, this module also
-provides a way to fetch subroutine and variable attributes.
-See L<attributes>.
+The C<h2xs> tool can now work in conjunction with C<C::Scan> (available
+from CPAN) to automatically parse real-life header files. The C<-M>,
+C<-a>, C<-k>, and C<-o> options are new.
-=item B
+=head2 perlcc
- WARNING: The Compiler suite remains highly experimental. The
- generated code may not be correct, even it manages to execute
- without errors.
+C<perlcc> now supports the C and Bytecode backends. By default,
+it generates output from the simple C backend rather than the
+optimized C backend.
-The Perl Compiler suite has been extensively reworked for this
-release. More of the standard Perl testsuite passes when run
-under the Compiler, but there is still a significant way to
-go to achieve production quality compiled executables.
+Support for non-Unix platforms has been improved.
-=item ByteLoader
+=head2 perldoc
-The ByteLoader is a dedicated extension to generate and run
-Perl bytecode. See L<ByteLoader>.
+C<perldoc> has been reworked to avoid possible security holes.
+It will not by default let itself be run as the superuser, but you
+may still use the B<-U> switch to try to make it drop privileges
+first.
-=item constant
+=head2 The Perl Debugger
-References can now be used.
+Many bug fixes and enhancements were added to F<perl5db.pl>, the
+Perl debugger. The help documentation was rearranged. New commands
+include C<< < ? >>, C<< > ? >>, and C<< { ? >> to list out current
+actions, C<man I<docpage>> to run your doc viewer on some perl
+docset, and support for quoted options. The help information was
+rearranged, and should be viewable once again if you're using B<less>
+as your pager. A serious security hole was plugged--you should
+immediately remove all older versions of the Perl debugger as
+installed in previous releases, all the way back to perl3, from
+your system to avoid being bitten by this.
-The new version also allows a leading underscore in constant names, but
-disallows a double leading underscore (as in "__LINE__"). Some other names
-are disallowed or warned against, including BEGIN, END, etc. Some names
-which were forced into main:: used to fail silently in some cases; now they're
-fatal (outside of main::) and an optional warning (inside of main::).
-The ability to detect whether a constant had been set with a given name has
-been added.
+=head1 Improved Documentation
-See L<constant>.
+Many of the platform-specific README files are now part of the perl
+installation. See L<perl> for the complete list.
-=item charnames
+=over 4
-This pragma implements the C<\N> string escape. See L<charnames>.
+=item perlapi.pod
-=item Data::Dumper
+The official list of public Perl API functions.
-A C<Maxdepth> setting can be specified to avoid venturing
-too deeply into deep data structures. See L<Data::Dumper>.
+=item perlboot.pod
-The XSUB implementation of Dump() is now automatically called if the
-C<Useqq> setting is not in use.
+A tutorial for beginners on object-oriented Perl.
-Dumping C<qr//> objects works correctly.
+=item perlcompile.pod
-=item DB
+An introduction to using the Perl Compiler suite.
-C<DB> is an experimental module that exposes a clean abstraction
-to Perl's debugging API.
+=item perldbmfilter.pod
-=item DB_File
+A howto document on using the DBM filter facility.
-DB_File can now be built with Berkeley DB versions 1, 2 or 3.
-See C<ext/DB_File/Changes>.
+=item perldebug.pod
-=item Devel::DProf
+All material unrelated to running the Perl debugger, plus all
+low-level guts-like details that risked crushing the casual user
+of the debugger, have been relocated from the old manpage to the
+next entry below.
-Devel::DProf, a Perl source code profiler has been added. See
-L<Devel::DProf> and L<dprofpp>.
+=item perldebguts.pod
-=item Dumpvalue
+This new manpage contains excessively low-level material not related
+to the Perl debugger, but slightly related to debugging Perl itself.
+It also contains some arcane internal details of how the debugging
+process works that may only be of interest to developers of Perl
+debuggers.
-The Dumpvalue module provides screen dumps of Perl data.
+=item perlfork.pod
-=item DynaLoader
+Notes on the fork() emulation currently available for the Windows platform.
-DynaLoader now supports a dl_unload_file() function on platforms that
-support unloading shared objects using dlclose().
+=item perlfilter.pod
-Perl can also optionally arrange to unload all extension shared objects
-loaded by Perl. To enable this, build Perl with the Configure option
-C<-Accflags=-DDL_UNLOAD_ALL_AT_EXIT>. (This maybe useful if you are
-using Apache with mod_perl.)
+An introduction to writing Perl source filters.
-=item Benchmark
+=item perlhack.pod
-Overall, Benchmark results exhibit lower average error and better timing
-accuracy.
+Some guidelines for hacking the Perl source code.
-You can now run tests for I<n> seconds instead of guessing the right
-number of tests to run: e.g., timethese(-5, ...) will run each
-code for at least 5 CPU seconds. Zero as the "number of repetitions"
-means "for at least 3 CPU seconds". The output format has also
-changed. For example:
+=item perlintern.pod
- use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}})
+A list of internal functions in the Perl source code.
+(List is currently empty.)
-will now output something like this:
+=item perllexwarn.pod
- Benchmark: running a, b, each for at least 5 CPU seconds...
- a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516)
- b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686)
+Introduction and reference information about lexically scoped
+warning categories.
-New features: "each for at least N CPU seconds...", "wallclock secs",
-and the "@ operations/CPU second (n=operations)".
+=item perlnumber.pod
-timethese() now returns a reference to a hash of Benchmark objects containing
-the test results, keyed on the names of the tests.
+Detailed information about numbers as they are represented in Perl.
-timethis() now returns the iterations field in the Benchmark result object
-instead of 0.
+=item perlopentut.pod
-timethese(), timethis(), and the new cmpthese() (see below) can also take
-a format specifier of 'none' to suppress output.
+A tutorial on using open() effectively.
-A new function countit() is just like timeit() except that it takes a
-TIME instead of a COUNT.
+=item perlreftut.pod
-A new function cmpthese() prints a chart comparing the results of each test
-returned from a timethese() call. For each possible pair of tests, the
-percentage speed difference (iters/sec or seconds/iter) is shown.
+A tutorial that introduces the essentials of references.
-For other details, see L<Benchmark>.
+=item perltootc.pod
-=item Devel::Peek
+A tutorial on managing class data for object modules.
-The Devel::Peek module provides access to the internal representation
-of Perl variables and data. It is a data debugging tool for the XS programmer.
+=item perltodo.pod
-=item English
+Discussion of the most often wanted features that may someday be
+supported in Perl.
-$PERL_VERSION now stands for C<$^V> (a string value) rather than for C<$]>
-(a numeric value).
+=item perlunicode.pod
-=item Env
+An introduction to Unicode support features in Perl.
-Env now supports accessing environment variables like PATH as array
-variables.
+=back
-=item Fcntl
+=head1 Performance enhancements
-More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for
-large file (more than 4GB) access (NOTE: the O_LARGEFILE is
-automatically added to sysopen() flags if large file support has been
-configured, as is the default), Free/Net/OpenBSD locking behaviour
-flags F_FLOCK, F_POSIX, Linux F_SHLCK, and O_ACCMODE: the combined
-mask of O_RDONLY, O_WRONLY, and O_RDWR. The seek()/sysseek()
-constants SEEK_SET, SEEK_CUR, and SEEK_END are available via the
-C<:seek> tag. The chmod()/stat() S_IF* constants and S_IS* functions
-are available via the C<:mode> tag.
+=head2 Simple sort() using { $a <=> $b } and the like are optimized
-=item File::Compare
+Many common sort() operations using a simple inlined block are now
+optimized for faster performance.
-A compare_text() function has been added, which allows custom
-comparison functions. See L<File::Compare>.
+=head2 Optimized assignments to lexical variables
-=item File::Find
+Certain operations in the RHS of assignment statements have been
+optimized to directly set the lexical variable on the LHS,
+eliminating redundant copying overheads.
-File::Find now works correctly when the wanted() function is either
-autoloaded or is a symbolic reference.
+=head2 Faster subroutine calls
-A bug that caused File::Find to lose track of the working directory
-when pruning top-level directories has been fixed.
+Minor changes in how subroutine calls are handled internally
+provide marginal improvements in performance.
-File::Find now also supports several other options to control its
-behavior. It can follow symbolic links if the C<follow> option is
-specified. Enabling the C<no_chdir> option will make File::Find skip
-changing the current directory when walking directories. The C<untaint>
-flag can be useful when running with taint checks enabled.
+=item delete(), each(), values() and hash iteration are faster
-See L<File::Find>.
+The hash values returned by delete(), each(), values() and hashes in a
+list context are the actual values in the hash, instead of copies.
+This results in significantly better performance, because it eliminates
+needless copying in most situations.
-=item File::Glob
+=head1 Installation and Configuration Improvements
-This extension implements BSD-style file globbing. By default,
-it will also be used for the internal implementation of the glob()
-operator. See L<File::Glob>.
+=head2 -Dusethreads means something different
-=item File::Spec
+The -Dusethreads flag now enables the experimental interpreter-based thread
+support by default. To get the flavor of experimental threads that was in
+5.005 instead, you need to run Configure with "-Dusethreads -Duse5005threads".
-New methods have been added to the File::Spec module: devnull() returns
-the name of the null device (/dev/null on Unix) and tmpdir() the name of
-the temp directory (normally /tmp on Unix). There are now also methods
-to convert between absolute and relative filenames: abs2rel() and
-rel2abs(). For compatibility with operating systems that specify volume
-names in file paths, the splitpath(), splitdir(), and catdir() methods
-have been added.
+As of v5.6.0, interpreter-threads support is still lacking a way to
+create new threads from Perl (i.e., C<use Thread;> will not work with
+interpreter threads). C<use Thread;> continues to be available when you
+specify the -Duse5005threads option to Configure, bugs and all.
-=item File::Spec::Functions
+ NOTE: Support for threads continues to be an experimental feature.
+ Interfaces and implementation are subject to sudden and drastic changes.
-The new File::Spec::Functions modules provides a function interface
-to the File::Spec module. Allows shorthand
+=head2 New Configure flags
- $fullname = catfile($dir1, $dir2, $file);
+The following new flags may be enabled on the Configure command line
+by running Configure with C<-Dflag>.
-instead of
+ usemultiplicity
+ usethreads useithreads (new interpreter threads: no Perl API yet)
+ usethreads use5005threads (threads as they were in 5.005)
- $fullname = File::Spec->catfile($dir1, $dir2, $file);
+ use64bitint (equal to now deprecated 'use64bits')
+ use64bitall
-=item Getopt::Long
+ uselongdouble
+ usemorebits
+ uselargefiles
+ usesocks (only SOCKS v5 supported)
-Getopt::Long licensing has changed to allow the Perl Artistic License
-as well as the GPL. It used to be GPL only, which got in the way of
-non-GPL applications that wanted to use Getopt::Long.
+=head2 Threadedness and 64-bitness now more daring
-Getopt::Long encourages the use of Pod::Usage to produce help
-messages. For example:
+The Configure options enabling the use of threads and the use of
+64-bitness are now more daring in the sense that they no more have an
+explicit list of operating systems of known threads/64-bit
+capabilities. In other words: if your operating system has the
+necessary APIs and datatypes, you should be able just to go ahead and
+use them, for threads by Configure -Dusethreads, and for 64 bits
+either explicitly by Configure -Duse64bitint or implicitly if your
+system has 64-bit wide datatypes. See also L<"64-bit support">.
- use Getopt::Long;
- use Pod::Usage;
- my $man = 0;
- my $help = 0;
- GetOptions('help|?' => \$help, man => \$man) or pod2usage(2);
- pod2usage(1) if $help;
- pod2usage(-exitstatus => 0, -verbose => 2) if $man;
+=head2 Long Doubles
- __END__
+Some platforms have "long doubles", floating point numbers of even
+larger range than ordinary "doubles". To enable using long doubles for
+Perl's scalars, use -Duselongdouble.
- =head1 NAME
+=head2 -Dusemorebits
- sample - Using GetOpt::Long and Pod::Usage
+You can enable both -Duse64bitint and -Duselongdouble with -Dusemorebits.
+See also L<"64-bit support">.
- =head1 SYNOPSIS
+=head2 -Duselargefiles
- sample [options] [file ...]
+Some platforms support system APIs that are capable of handling large files
+(typically, files larger than two gigabytes). Perl will try to use these
+APIs if you ask for -Duselargefiles.
- Options:
- -help brief help message
- -man full documentation
+See L<"Large file support"> for more information.
- =head1 OPTIONS
+=head2 installusrbinperl
- =over 8
+You can use "Configure -Uinstallusrbinperl" which causes installperl
+to skip installing perl also as /usr/bin/perl. This is useful if you
+prefer not to modify /usr/bin for some reason or another but harmful
+because many scripts assume to find Perl in /usr/bin/perl.
- =item B<-help>
+=head2 SOCKS support
- Print a brief help message and exits.
+You can use "Configure -Dusesocks" which causes Perl to probe
+for the SOCKS proxy protocol library (v5, not v4). For more information
+on SOCKS, see:
- =item B<-man>
+ http://www.socks.nec.com/
- Prints the manual page and exits.
+=head2 C<-A> flag
- =back
+You can "post-edit" the Configure variables using the Configure C<-A>
+switch. The editing happens immediately after the platform specific
+hints files have been processed but before the actual configuration
+process starts. Run C<Configure -h> to find out the full C<-A> syntax.
- =head1 DESCRIPTION
+=head2 Enhanced Installation Directories
- B<This program> will read the given input file(s) and do someting
- useful with the contents thereof.
+The installation structure has been enriched to improve the support
+for maintaining multiple versions of perl, to provide locations for
+vendor-supplied modules, scripts, and manpages, and to ease maintenance
+of locally-added modules, scripts, and manpages. See the section on
+Installation Directories in the INSTALL file for complete details.
+For most users building and installing from source, the defaults should
+be fine.
- =cut
+If you previously used C<Configure -Dsitelib> or C<-Dsitearch> to set
+special values for library directories, you might wish to consider using
+the new C<-Dsiteprefix> setting instead. Also, if you wish to re-use a
+config.sh file from an earlier version of perl, you should be sure to
+check that Configure makes sensible choices for the new directories.
+See INSTALL for complete details.
-See L<Pod::Usage> for details.
+=head1 Platform specific changes
-A bug that prevented the non-option call-back <> from being
-specified as the first argument has been fixed.
+=head2 Supported platforms
-To specify the characters < and > as option starters, use ><. Note,
-however, that changing option starters is strongly deprecated.
+=over 4
-=item IO
+=item *
-write() and syswrite() will now accept a single-argument
-form of the call, for consistency with Perl's syswrite().
+VM/ESA is now supported.
-You can now create a TCP-based IO::Socket::INET without forcing
-a connect attempt. This allows you to configure its options
-(like making it non-blocking) and then call connect() manually.
+=item *
-A bug that prevented the IO::Socket::protocol() accessor
-from ever returning the correct value has been corrected.
+Siemens BS2000 is now supported under the POSIX Shell.
-IO::Socket::connect now uses non-blocking IO instead of alarm()
-to do connect timeouts.
+=item *
-IO::Socket::accept now uses select() instead of alarm() for doing
-timeouts.
+The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread
+extension.
-IO::Socket::INET->new now sets $! correctly on failure. $@ is
-still set for backwards compatability.
+=item *
-=item JPL
+GNU/Hurd is now supported.
-Java Perl Lingo is now distributed with Perl. See jpl/README
-for more information.
+=item *
-=item lib
+Rhapsody/Darwin is now supported.
-C<use lib> now weeds out any trailing duplicate entries.
-C<no lib> removes all named entries.
+=item *
-=item Math::BigInt
+EPOC is is now supported (on Psion 5).
-The bitwise operations C<<< << >>>, C<<< >> >>>, C<&>, C<|>,
-and C<~> are now supported on bigints.
+=item *
-=item Math::Complex
+The cygwin port (formerly cygwin32) has been greatly improved.
-The accessor methods Re, Im, arg, abs, rho, and theta can now also
-act as mutators (accessor $z->Re(), mutator $z->Re(3)).
+=back
-The class method C<display_format> and the corresponding object method
-C<display_format>, in addition to accepting just one argument, now can
-also accept a parameter hash. Recognized keys of a parameter hash are
-C<"style">, which corresponds to the old one parameter case, and two
-new parameters: C<"format">, which is a printf()-style format string
-(defaults usually to C<"%.15g">, you can revert to the default by
-setting the format string to C<undef>) used for both parts of a
-complex number, and C<"polar_pretty_print"> (defaults to true),
-which controls whether an attempt is made to try to recognize small
-multiples and rationals of pi (2pi, pi/2) at the argument (angle) of a
-polar complex number.
+=head2 DOS
-The potentially disruptive change is that in list context both methods
-now I<return the parameter hash>, instead of only the value of the
-C<"style"> parameter.
+=over 4
-=item Math::Trig
+=item *
-A little bit of radial trigonometry (cylindrical and spherical),
-radial coordinate conversions, and the great circle distance were added.
+Perl now works with djgpp 2.02 (and 2.03 alpha).
-=item Pod::Parser, Pod::InputObjects
+=item *
-Pod::Parser is a base class for parsing and selecting sections of
-pod documentation from an input stream. This module takes care of
-identifying pod paragraphs and commands in the input and hands off the
-parsed paragraphs and commands to user-defined methods which are free
-to interpret or translate them as they see fit.
+Environment variable names are not converted to uppercase any more.
-Pod::InputObjects defines some input objects needed by Pod::Parser, and
-for advanced users of Pod::Parser that need more about a command besides
-its name and text.
+=item *
-As of release 5.6.0 of Perl, Pod::Parser is now the officially sanctioned
-"base parser code" recommended for use by all pod2xxx translators.
-Pod::Text (pod2text) and Pod::Man (pod2man) have already been converted
-to use Pod::Parser and efforts to convert Pod::HTML (pod2html) are already
-underway. For any questions or comments about pod parsing and translating
-issues and utilities, please use the pod-people@perl.org mailing list.
+Incorrect exit codes from backticks have been fixed.
-For further information, please see L<Pod::Parser> and L<Pod::InputObjects>.
+=item *
-=item Pod::Checker, podchecker
+This port continues to use its own builtin globbing (not File::Glob).
-This utility checks pod files for correct syntax, according to
-L<perlpod>. Obvious errors are flagged as such, while warnings are
-printed for mistakes that can be handled gracefully. The checklist is
-not complete yet. See L<Pod::Checker>.
+=back
-=item Pod::ParseUtils, Pod::Find
+=head2 OS390 (OpenEdition MVS)
-These modules provide a set of gizmos that are useful mainly for pod
-translators. L<Pod::Find|Pod::Find> traverses directory structures and
-returns found pod files, along with their canonical names (like
-C<File::Spec::Unix>). L<Pod::ParseUtils|Pod::ParseUtils> contains
-B<Pod::List> (useful for storing pod list information), B<Pod::Hyperlink>
-(for parsing the contents of C<LE<lt>E<gt>> sequences) and B<Pod::Cache>
-(for caching information about pod files, e.g., link nodes).
+Support for this EBCDIC platform has not been renewed in this release.
+There are difficulties in reconciling Perl's standardization on UTF-8
+as its internal representation for characters with the EBCDIC character
+set, because the two are incompatible.
-=item Pod::Select, podselect
+It is unclear whether future versions will renew support for this
+platform, but the possibility exists.
-Pod::Select is a subclass of Pod::Parser which provides a function
-named "podselect()" to filter out user-specified sections of raw pod
-documentation from an input stream. podselect is a script that provides
-access to Pod::Select from other scripts to be used as a filter.
-See L<Pod::Select>.
+=head2 VMS
-=item Pod::Usage, pod2usage
+Numerous revisions and extensions to configuration, build, testing, and
+installation process to accomodate core changes and VMS-specific options.
-Pod::Usage provides the function "pod2usage()" to print usage messages for
-a Perl script based on its embedded pod documentation. The pod2usage()
-function is generally useful to all script authors since it lets them
-write and maintain a single source (the pods) for documentation, thus
-removing the need to create and maintain redundant usage message text
-consisting of information already in the pods.
+Expand %ENV-handling code to allow runtime mapping to logical names,
+CLI symbols, and CRTL environ array.
-There is also a pod2usage script which can be used from other kinds of
-scripts to print usage messages from pods (even for non-Perl scripts
-with pods embedded in comments).
+Extension of subprocess invocation code to accept filespecs as command
+"verbs".
-For details and examples, please see L<Pod::Usage>.
+Add to Perl command line processing the ability to use default file types and
+to recognize Unix-style C<2E<gt>&1>.
-=item Pod::Text and Pod::Man
+Expansion of File::Spec::VMS routines, and integration into ExtUtils::MM_VMS.
-Pod::Text has been rewritten to use Pod::Parser. While pod2text() is
-still available for backwards compatibility, the module now has a new
-preferred interface. See L<Pod::Text> for the details. The new Pod::Text
-module is easily subclassed for tweaks to the output, and two such
-subclasses (Pod::Text::Termcap for man-page-style bold and underlining
-using termcap information, and Pod::Text::Color for markup with ANSI color
-sequences) are now standard.
+Extension of ExtUtils::MM_VMS to handle complex extensions more flexibly.
+
+Barewords at start of Unix-syntax paths may be treated as text rather than
+only as logical names.
+
+Optional secure translation of several logical names used internally by Perl.
+
+Miscellaneous bugfixing and porting of new core code to VMS.
-pod2man has been turned into a module, Pod::Man, which also uses
-Pod::Parser. In the process, several outstanding bugs related to quotes
-in section headers, quoting of code escapes, and nested lists have been
-fixed. pod2man is now a wrapper script around this module.
+Thanks are gladly extended to the many people who have contributed VMS
+patches, testing, and ideas.
-=item SDBM_File
+=head2 Win32
-An EXISTS method has been added to this module (and sdbm_exists() has
-been added to the underlying sdbm library), so one can now call exists
-on an SDBM_File tied hash and get the correct result, rather than a
-runtime error.
+Perl can now emulate fork() internally, using multiple interpreters running
+in different concurrent threads. This support must be enabled at build
+time. See L<perlfork> for detailed information.
-A bug that may have caused data loss when more than one disk block
-happens to be read from the database in a single FETCH() has been
-fixed.
+When given a pathname that consists only of a drivename, such as C<A:>,
+opendir() and stat() now use the current working directory for the drive
+rather than the drive root.
-=item Sys::Syslog
+The builtin XSUB functions in the Win32:: namespace are documented. See
+L<Win32>.
-Sys::Syslog now uses XSUBs to access facilities from syslog.h so it
-no longer requires syslog.ph to exist.
+$^X now contains the full path name of the running executable.
-=item Sys::Hostname
+A Win32::GetLongPathName() function is provided to complement
+Win32::GetFullPathName() and Win32::GetShortPathName(). See L<Win32>.
-Sys::Hostname now uses XSUBs to call the C library's gethostname() or
-uname() if they exist.
+POSIX::uname() is supported.
-=item Term::ANSIColor
+system(1,...) now returns true process IDs rather than process
+handles. kill() accepts any real process id, rather than strictly
+return values from system(1,...).
-Term::ANSIColor is a very simple module to provide easy and readable
-access to the ANSI color and highlighting escape sequences, supported by
-most ANSI terminal emulators. It is now included standard.
+For better compatibility with Unix, C<kill(0, $pid)> can now be used to
+test whether a process exists.
-=item Time::Local
+The C<Shell> module is supported.
-The timelocal() and timegm() functions used to silently return bogus
-results when the date fell outside the machine's integer range. They
-now consistently croak() if the date falls in an unsupported range.
+Better support for building Perl under command.com in Windows 95
+has been added.
-=item Win32
+Scripts are read in binary mode by default to allow ByteLoader (and
+the filter mechanism in general) to work properly. For compatibility,
+the DATA filehandle will be set to text mode if a carriage return is
+detected at the end of the line containing the __END__ or __DATA__
+token; if not, the DATA filehandle will be left open in binary mode.
+Earlier versions always opened the DATA filehandle in text mode.
-The error return value in list context has been changed for all functions
-that return a list of values. Previously these functions returned a list
-with a single element C<undef> if an error occurred. Now these functions
-return the empty list in these situations. This applies to the following
-functions:
+The glob() operator is implemented via the C<File::Glob> extension,
+which supports glob syntax of the C shell. This increases the flexibility
+of the glob() operator, but there may be compatibility issues for
+programs that relied on the older globbing syntax. If you want to
+preserve compatibility with the older syntax, you might want to run
+perl with C<-MFile::DosGlob>. For details and compatibility information,
+see L<File::Glob>.
- Win32::FsType
- Win32::GetOSVersion
+=head1 Significant bug fixes
-The remaining functions are unchanged and continue to return C<undef> on
-error even in list context.
+=head2 <HANDLE> on empty files
-The Win32::SetLastError(ERROR) function has been added as a complement
-to the Win32::GetLastError() function.
+With C<$/> set to C<undef>, "slurping" an empty file returns a string of
+zero length (instead of C<undef>, as it used to) the first time the
+HANDLE is read after C<$/> is set to C<undef>. Further reads yield
+C<undef>.
-The new Win32::GetFullPathName(FILENAME) returns the full absolute
-pathname for FILENAME in scalar context. In list context it returns
-a two-element list containing the fully qualified directory name and
-the filename. See L<Win32>.
+This means that the following will append "foo" to an empty file (it used
+to do nothing):
-=item XSLoader
+ perl -0777 -pi -e 's/^/foo/' empty_file
-The XSLoader extension is a simpler alternative to DynaLoader.
-See L<XSLoader>.
+The behaviour of:
-=item DBM Filters
+ perl -pi -e 's/^/foo/' empty_file
-A new feature called "DBM Filters" has been added to all the
-DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File.
-DBM Filters add four new methods to each DBM module:
+is unchanged (it continues to leave the file empty).
- filter_store_key
- filter_store_value
- filter_fetch_key
- filter_fetch_value
+=head2 C<eval '...'> improvements
-These can be used to filter key-value pairs before the pairs are
-written to the database or just after they are read from the database.
-See L<perldbmfilter> for further information.
+Line numbers (as reflected by caller() and most diagnostics) within
+C<eval '...'> were often incorrect where here documents were involved.
+This has been corrected.
-=back
+Lexical lookups for variables appearing in C<eval '...'> within
+functions that were themselves called within an C<eval '...'> were
+searching the wrong place for lexicals. The lexical search now
+correctly ends at the subroutine's block boundary.
-=head2 Pragmata
+The use of C<return> within C<eval {...}> caused $@ not to be reset
+correctly when no exception occurred within the eval. This has
+been fixed.
-C<use attrs> is now obsolete, and is only provided for
-backward-compatibility. It's been replaced by the C<sub : attributes>
-syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>.
+Parsing of here documents used to be flawed when they appeared as
+the replacement expression in C<eval 's/.../.../e'>. This has
+been fixed.
-Lexical warnings pragma, C<use warnings;>, to control optional warnings.
-See L<perllexwarn>.
+=head2 All compilation errors are true errors
-C<use filetest> to control the behaviour of filetests (C<-r> C<-w>
-...). Currently only one subpragma implemented, "use filetest
-'access';", that uses access(2) or equivalent to check permissions
-instead of using stat(2) as usual. This matters in filesystems
-where there are ACLs (access control lists): the stat(2) might lie,
-but access(2) knows better.
+Some "errors" encountered at compile time were by neccessity
+generated as warnings followed by eventual termination of the
+program. This enabled more such errors to be reported in a
+single run, rather than causing a hard stop at the first error
+that was encountered.
-=head1 Utility Changes
+The mechanism for reporting such errors has been reimplemented
+to queue compile-time errors and report them at the end of the
+compilation as true errors rather than as warnings. This fixes
+cases where error messages leaked through in the form of warnings
+when code was compiled at run time using C<eval STRING>, and
+also allows such errors to be reliably trapped using C<eval "...">.
-=head2 perlcc
+=head2 Implicitly closed filehandles are safer
-C<perlcc> now supports the C and Bytecode backends. By default,
-it generates output from the simple C backend rather than the
-optimized C backend.
+Sometimes implicitly closed filehandles (as when they are localized,
+and Perl automatically closes them on exiting the scope) could
+inadvertently set $? or $!. This has been corrected.
-Support for non-Unix platforms has been improved.
-=head2 perldoc
+=head2 Behavior of list slices is more consistent
-C<perldoc> has been reworked to avoid possible security holes.
-It will not by default let itself be run as the superuser, but you
-may still use the B<-U> switch to try to make it drop privileges
-first.
+When taking a slice of a literal list (as opposed to a slice of
+an array or hash), Perl used to return an empty list if the
+result happened to be composed of all undef values.
-=head2 The Perl Debugger
+The new behavior is to produce an empty list if (and only if)
+the original list was empty. Consider the following example:
-Many bug fixes and enhancements were added to F<perl5db.pl>, the
-Perl debugger. The help documentation was rearranged. New commands
-include C<< < ? >>, C<< > ? >>, and C<< { ? >> to list out current
-actions, C<man I<docpage>> to run your doc viewer on some perl
-docset, and support for quoted options. The help information was
-rearranged, and should be viewable once again if you're using B<less>
-as your pager. A serious security hole was plugged--you should
-immediately remove all older versions of the Perl debugger as
-installed in previous releases, all the way back to perl3, from
-your system to avoid being bitten by this.
+ @a = (1,undef,undef,2)[2,1,2];
-=head1 Documentation Changes
+The old behavior would have resulted in @a having no elements.
+The new behavior ensures it has three undefined elements.
-=over 4
+Note in particular that the behavior of slices of the following
+cases remains unchanged:
-=item perlapi.pod
+ @a = ()[1,2];
+ @a = (getpwent)[7,0];
+ @a = (anything_returning_empty_list())[2,1,2];
+ @a = @b[2,1,2];
+ @a = @c{'a','b','c'};
-The official list of public Perl API functions.
+See L<perldata>.
-=item perlcompile.pod
+=head2 C<(\$)> prototype and C<$foo{a}>
-An introduction to using the Perl Compiler suite.
+A scalar reference prototype now correctly allows a hash or
+array element in that slot.
-=item perldebug.pod
+=head2 C<goto &sub> and AUTOLOAD
-All material unrelated to running the Perl debugger, plus all
-low-level guts-like details that risked crushing the casual user
-of the debugger, have been relocated from the old manpage to the
-next entry below.
+The C<goto &sub> construct works correctly when C<&sub> happens
+to be autoloaded.
-=item perldebguts.pod
+=head2 C<-bareword> allowed under C<use integer>
-This new manpage contains excessively low-level material not related
-to the Perl debugger, but slightly related to debugging Perl itself.
-It also contains some arcane internal details of how the debugging
-process works that may only be of interest to developers of Perl
-debuggers.
+The autoquoting of barewords preceded by C<-> did not work
+in prior versions when the C<integer> pragma was enabled.
+This has been fixed.
-=item perlfilter.pod
+=head2 Failures in DESTROY()
-An introduction to writing Perl source filters.
+When code in a destructor threw an exception, it went unnoticed
+in earlier versions of Perl, unless someone happened to be
+looking in $@ just after the point the destructor happened to
+run. Such failures are now visible as warnings when warnings are
+enabled.
-=item perlhack.pod
+=head2 Locale bugs fixed
-Some guidelines for hacking the Perl source code.
+printf() and sprintf() previously reset the numeric locale
+back to the default "C" locale. This has been fixed.
-=item perlintern.pod
+Numbers formatted according to the local numeric locale
+(such as using a decimal comma instead of a decimal dot) caused
+"isn't numeric" warnings, even while the operations accessing
+those numbers produced correct results. These warnings have been
+discontinued.
-A list of internal functions in the Perl source code.
-(List is currently empty.)
+=head2 Memory leaks
-=item perlopentut.pod
+The C<eval 'return sub {...}'> construct could sometimes leak
+memory. This has been fixed.
-A tutorial on using open() effectively.
+Operations that aren't filehandle constructors used to leak memory
+when used on invalid filehandles. This has been fixed.
-=item perlreftut.pod
+Constructs that modified C<@_> could fail to deallocate values
+in C<@_> and thus leak memory. This has been corrected.
-A tutorial that introduces the essentials of references.
+=head2 Spurious subroutine stubs after failed subroutine calls
-=item perlboot.pod
+Perl could sometimes create empty subroutine stubs when a
+subroutine was not found in the package. Such cases stopped
+later method lookups from progressing into base packages.
+This has been corrected.
-A tutorial for beginners on object-oriented Perl.
+=head2 Taint failures under C<-U>
-=item perltootc.pod
+When running in unsafe mode, taint violations could sometimes
+cause silent failures. This has been fixed.
+
+=head2 END blocks and the C<-c> switch
+
+Prior versions used to run BEGIN B<and> END blocks when Perl was
+run in compile-only mode. Since this is typically not the expected
+behavior, END blocks are not executed anymore when the C<-c> switch
+is used.
-A tutorial on managing class data for object modules.
+See L<CHECK blocks> for how to run things when the compile phase ends.
-=item perlunicode.pod
+=head2 Potential to leak DATA filehandles
-An introduction to Unicode support features in Perl.
+Using the C<__DATA__> token creates an implicit filehandle to
+the file that contains the token. It is the program's
+responsibility to close it when it is done reading from it.
-=back
+This caveat is now better explained in the documentation.
+See L<perldata>.
=head1 New or Changed Diagnostics
(F) The argument to exists() must be a hash or array element, such as:
$foo{$bar}
- $ref->[12]->["susie"]
+ $ref->{"susie"}[12]
=item %s argument is not a HASH or ARRAY element or slice
(F) The argument to delete() must be either a hash or array element, such as:
$foo{$bar}
- $ref->[12]->["susie"]
+ $ref->{"susie"}[12]
or a hash or array slice, such as:
=item Reference is already weak
-(W misc) You have attempted to weaken a reference that is already weak.
-Doing so has no effect.
+(W misc) You have attempted to weaken a reference that is already weak.
+Doing so has no effect.
+
+=item setpgrp can't take arguments
+
+(F) Your system has the setpgrp() from BSD 4.2, which takes no arguments,
+unlike POSIX setpgid(), which takes a process ID and process group ID.
+
+=item Strange *+?{} on zero-length expression
+
+(W regexp) You applied a regular expression quantifier in a place where it
+makes no sense, such as on a zero-width assertion.
+Try putting the quantifier inside the assertion instead. For example,
+the way to match "abc" provided that it is followed by three
+repetitions of "xyz" is C</abc(?=(?:xyz){3})/>, not C</abc(?=xyz){3}/>.
+
+=item switching effective %s is not implemented
+
+(F) While under the C<use filetest> pragma, we cannot switch the
+real and effective uids or gids.
+
+=item This Perl can't reset CRTL environ elements (%s)
+
+=item This Perl can't set CRTL environ elements (%s=%s)
+
+(W internal) Warnings peculiar to VMS. You tried to change or delete an element
+of the CRTL's internal environ array, but your copy of Perl wasn't
+built with a CRTL that contained the setenv() function. You'll need to
+rebuild Perl with a CRTL that does, or redefine F<PERL_ENV_TABLES> (see
+L<perlvms>) so that the environ array isn't the target of the change to
+%ENV which produced the warning.
+
+=item Too late to run %s block
+
+(W void) A CHECK or INIT block is being defined during run time proper,
+when the opportunity to run them has already passed. Perhaps you are
+loading a file with C<require> or C<do> when you should be using
+C<use> instead. Or perhaps you should put the C<require> or C<do>
+inside a BEGIN block.
+
+=item Unknown open() mode '%s'
+
+(F) The second argument of 3-argument open() is not among the list
+of valid modes: C<< < >>, C<< > >>, C<<< >> >>>, C<< +< >>,
+C<< +> >>, C<<< +>> >>>, C<-|>, C<|->.
+
+=item Unknown process %x sent message to prime_env_iter: %s
+
+(P) An error peculiar to VMS. Perl was reading values for %ENV before
+iterating over it, and someone else stuck a message in the stream of
+data Perl expected. Someone's very confused, or perhaps trying to
+subvert Perl's population of %ENV for nefarious purposes.
+
+=item Unrecognized escape \\%c passed through
+
+(W misc) You used a backslash-character combination which is not recognized
+by Perl. The character was understood literally.
+
+=item Unterminated attribute parameter in attribute list
+
+(F) The lexer saw an opening (left) parenthesis character while parsing an
+attribute list, but the matching closing (right) parenthesis
+character was not found. You may need to add (or remove) a backslash
+character to get your parentheses to balance. See L<attributes>.
+
+=item Unterminated attribute list
+
+(F) The lexer found something other than a simple identifier at the start
+of an attribute, and it wasn't a semicolon or the start of a
+block. Perhaps you terminated the parameter list of the previous attribute
+too soon. See L<attributes>.
+
+=item Unterminated attribute parameter in subroutine attribute list
+
+(F) The lexer saw an opening (left) parenthesis character while parsing a
+subroutine attribute list, but the matching closing (right) parenthesis
+character was not found. You may need to add (or remove) a backslash
+character to get your parentheses to balance.
+
+=item Unterminated subroutine attribute list
+
+(F) The lexer found something other than a simple identifier at the start
+of a subroutine attribute, and it wasn't a semicolon or the start of a
+block. Perhaps you terminated the parameter list of the previous attribute
+too soon.
+
+=item Value of CLI symbol "%s" too long
+
+(W misc) A warning peculiar to VMS. Perl tried to read the value of an %ENV
+element from a CLI symbol table, and found a resultant string longer
+than 1024 characters. The return value has been truncated to 1024
+characters.
+
+=item Version number must be a constant number
+
+(P) The attempt to translate a C<use Module n.n LIST> statement into
+its equivalent C<BEGIN> block found an internal inconsistency with
+the version number.
+
+=back
+
+=head1 New tests
+
+=over 4
+
+=item lib/attrs
+
+Compatibility tests for C<sub : attrs> vs the older C<use attrs>.
+
+=item lib/env
+
+Tests for new environment scalar capability (e.g., C<use Env qw($BAR);>).
+
+=item lib/env-array
+
+Tests for new environment array capability (e.g., C<use Env qw(@PATH);>).
+
+=item lib/io_const
+
+IO constants (SEEK_*, _IO*).
+
+=item lib/io_dir
+
+Directory-related IO methods (new, read, close, rewind, tied delete).
+
+=item lib/io_multihomed
+
+INET sockets with multi-homed hosts.
+
+=item lib/io_poll
+
+IO poll().
+
+=item lib/io_unix
+
+UNIX sockets.
+
+=item op/attrs
+
+Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>.
+
+=item op/filetest
+
+File test operators.
+
+=item op/lex_assign
+
+Verify operations that access pad objects (lexicals and temporaries).
+
+=item op/exists_sub
+
+Verify C<exists &sub> operations.
+
+=back
+
+=head1 Incompatible Changes
+
+=head2 Perl Source Incompatibilities
+
+Beware that any new warnings that have been added or old ones
+that have been enhanced are B<not> considered incompatible changes.
+
+Since all new warnings must be explicitly requested via the C<-w>
+switch or the C<warnings> pragma, it is ultimately the programmer's
+responsibility to ensure that warnings are enabled judiciously.
+
+=over 4
+
+=item CHECK is a new keyword
+
+All subroutine definitions named CHECK are now special. See
+C</"Support for CHECK blocks"> for more information.
+
+=item Treatment of list slices of undef has changed
+
+There is a potential incompatibility in the behavior of list slices
+that are comprised entirely of undefined values.
+See L</"Behavior of list slices is more consistent">.
+
+=head2 Format of $English::PERL_VERSION is different
+
+The English module now sets $PERL_VERSION to $^V (a string value) rather
+than C<$]> (a numeric value). This is a potential incompatibility.
+Send us a report via perlbug if you are affected by this.
+
+See L</"Improved Perl version numbering system"> for the reasons for
+this change.
+
+=item Literals of the form C<1.2.3> parse differently
+
+Previously, numeric literals with more than one dot in them were
+interpreted as a floating point number concatenated with one or more
+numbers. Such "numbers" are now parsed as strings composed of the
+specified ordinals.
+
+For example, C<print 97.98.99> used to output C<97.9899> in earlier
+versions, but now prints C<abc>.
+
+See L</"Support for strings represented as a vector of ordinals">.
+
+=item Possibly changed pseudo-random number generator
+
+Perl programs that depend on reproducing a specific set of pseudo-random
+numbers may now produce different output due to improvements made to the
+rand() builtin. You can use C<sh Configure -Drandfunc=rand> to obtain
+the old behavior.
+
+See L</"Better pseudo-random number generator">.
+
+=item Hashing function for hash keys has changed
+
+Even though Perl hashes are not order preserving, the apparently
+random order encountered when iterating on the contents of a hash
+is actually determined by the hashing algorithm used. Improvements
+in the algorithm may yield a random order that is B<different> from
+that of previous versions, especially when iterating on hashes.
+
+See L</"Better worst-case behavior of hashes"> for additional
+information.
+
+=item C<undef> fails on read only values
+
+Using the C<undef> operator on a readonly value (such as $1) has
+the same effect as assigning C<undef> to the readonly value--it
+throws an exception.
+
+=item Close-on-exec bit may be set on pipe and socket handles
+
+Pipe and socket handles are also now subject to the close-on-exec
+behavior determined by the special variable $^F.
+
+See L</"More consistent close-on-exec behavior">.
+
+=item Writing C<"$$1"> to mean C<"${$}1"> is unsupported
+
+Perl 5.004 deprecated the interpretation of C<$$1> and
+similar within interpolated strings to mean C<$$ . "1">,
+but still allowed it.
+
+In Perl 5.6.0 and later, C<"$$1"> always means C<"${$1}">.
+
+=item delete(), values() and C<\(%h)> operate on aliases to values, not copies
+
+delete(), each(), values() and hashes in a list context return the actual
+values in the hash, instead of copies (as they used to in earlier
+versions). Typical idioms for using these constructs copy the
+returned values, but this can make a significant difference when
+creating references to the returned values. Keys in the hash are still
+returned as copies when iterating on a hash.
+
+See also L</"delete(), each(), values() and hash iteration are faster">.
+
+=item vec(EXPR,OFFSET,BITS) enforces powers-of-two BITS
+
+vec() generates a run-time error if the BITS argument is not
+a valid power-of-two integer.
+
+=item Text of some diagnostic output has changed
+
+Most references to internal Perl operations in diagnostics
+have been changed to be more descriptive. This may be an
+issue for programs that may incorrectly rely on the exact
+text of diagnostics for proper functioning.
+
+=item C<%@> has been removed
+
+The undocumented special variable C<%@> that used to accumulate
+"background" errors (such as those that happen in DESTROY())
+has been removed, because it could potentially result in memory
+leaks.
+
+=item Parenthesized not() behaves like a list operator
-=item setpgrp can't take arguments
+The C<not> operator now falls under the "if it looks like a function,
+it behaves like a function" rule.
-(F) Your system has the setpgrp() from BSD 4.2, which takes no arguments,
-unlike POSIX setpgid(), which takes a process ID and process group ID.
+As a result, the parenthesized form can be used with C<grep> and C<map>.
+The following construct used to be a syntax error before, but it works
+as expected now:
-=item Strange *+?{} on zero-length expression
+ grep not($_), @things;
-(W regexp) You applied a regular expression quantifier in a place where it
-makes no sense, such as on a zero-width assertion.
-Try putting the quantifier inside the assertion instead. For example,
-the way to match "abc" provided that it is followed by three
-repetitions of "xyz" is C</abc(?=(?:xyz){3})/>, not C</abc(?=xyz){3}/>.
+On the other hand, using C<not> with a literal list slice may not
+work. The following previously allowed construct:
-=item switching effective %s is not implemented
+ print not (1,2,3)[0];
-(F) While under the C<use filetest> pragma, we cannot switch the
-real and effective uids or gids.
+needs to be written with additional parentheses now:
-=item This Perl can't reset CRTL environ elements (%s)
+ print not((1,2,3)[0]);
-=item This Perl can't set CRTL environ elements (%s=%s)
+The behavior remains unaffected when C<not> is not followed by parentheses.
-(W internal) Warnings peculiar to VMS. You tried to change or delete an element
-of the CRTL's internal environ array, but your copy of Perl wasn't
-built with a CRTL that contained the setenv() function. You'll need to
-rebuild Perl with a CRTL that does, or redefine F<PERL_ENV_TABLES> (see
-L<perlvms>) so that the environ array isn't the target of the change to
-%ENV which produced the warning.
+=item Semantics of bareword prototype C<(*)> have changed
-=item Too late to run %s block
+The semantics of the bareword prototype C<*> have changed. Perl 5.005
+always coerced simple scalar arguments to a typeglob, which wasn't useful
+in situations where the subroutine must distinguish between a simple
+scalar and a typeglob. The new behavior is to not coerce bareword
+arguments to a typeglob. The value will always be visible as either
+a simple scalar or as a reference to a typeglob.
-(W void) A CHECK or INIT block is being defined during run time proper,
-when the opportunity to run them has already passed. Perhaps you are
-loading a file with C<require> or C<do> when you should be using
-C<use> instead. Or perhaps you should put the C<require> or C<do>
-inside a BEGIN block.
+See L</"More functional bareword prototype (*)">.
-=item Unknown open() mode '%s'
+=head2 Semantics of bit operators may have changed on 64-bit platforms
-(F) The second argument of 3-argument open() is not among the list
-of valid modes: C<< < >>, C<< > >>, C<<< >> >>>, C<< +< >>,
-C<< +> >>, C<<< +>> >>>, C<-|>, C<|->.
+If your platform is either natively 64-bit or if Perl has been
+configured to used 64-bit integers, i.e., $Config{ivsize} is 8,
+there may be a potential incompatibility in the behavior of bitwise
+numeric operators (& | ^ ~ << >>). These operators used to strictly
+operate on the lower 32 bits of integers in previous versions, but now
+operate over the entire native integral width. In particular, note
+that unary C<~> will produce different results on platforms that have
+different $Config{ivsize}. For portability, be sure to mask off
+the excess bits in the result of unary C<~>, e.g., C<~$x & 0xffffffff>.
-=item Unknown process %x sent message to prime_env_iter: %s
+See L</"Bit operators support full native integer width">.
-(P) An error peculiar to VMS. Perl was reading values for %ENV before
-iterating over it, and someone else stuck a message in the stream of
-data Perl expected. Someone's very confused, or perhaps trying to
-subvert Perl's population of %ENV for nefarious purposes.
+=head2 More builtins taint their results
-=item Unrecognized escape \\%c passed through
+As described in L</"Improved security features">, there may be more
+sources of taint in a Perl program.
-(W misc) You used a backslash-character combination which is not recognized
-by Perl. The character was understood literally.
+To avoid these new tainting behaviors, you can build Perl with the
+Configure option C<-Accflags=-DINCOMPLETE_TAINTS>. Beware that the
+ensuing perl binary may be insecure.
-=item Unterminated attribute parameter in attribute list
+=back
-(F) The lexer saw an opening (left) parenthesis character while parsing an
-attribute list, but the matching closing (right) parenthesis
-character was not found. You may need to add (or remove) a backslash
-character to get your parentheses to balance. See L<attributes>.
+=head2 C Source Incompatibilities
-=item Unterminated attribute list
+=over 4
-(F) The lexer found something other than a simple identifier at the start
-of an attribute, and it wasn't a semicolon or the start of a
-block. Perhaps you terminated the parameter list of the previous attribute
-too soon. See L<attributes>.
+=item C<PERL_POLLUTE>
-=item Unterminated attribute parameter in subroutine attribute list
+Release 5.005 grandfathered old global symbol names by providing preprocessor
+macros for extension source compatibility. As of release 5.6.0, these
+preprocessor definitions are not available by default. You need to explicitly
+compile perl with C<-DPERL_POLLUTE> to get these definitions. For
+extensions still using the old symbols, this option can be
+specified via MakeMaker:
-(F) The lexer saw an opening (left) parenthesis character while parsing a
-subroutine attribute list, but the matching closing (right) parenthesis
-character was not found. You may need to add (or remove) a backslash
-character to get your parentheses to balance.
+ perl Makefile.PL POLLUTE=1
-=item Unterminated subroutine attribute list
+=item C<PERL_IMPLICIT_CONTEXT>
-(F) The lexer found something other than a simple identifier at the start
-of a subroutine attribute, and it wasn't a semicolon or the start of a
-block. Perhaps you terminated the parameter list of the previous attribute
-too soon.
+This new build option provides a set of macros for all API functions
+such that an implicit interpreter/thread context argument is passed to
+every API function. As a result of this, something like C<sv_setsv(foo,bar)>
+amounts to a macro invocation that actually translates to something like
+C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected
+to not have any significant source compatibility issues, the difference
+between a macro and a real function call will need to be considered.
-=item Value of CLI symbol "%s" too long
+This means that there B<is> a source compatibility issue as a result of
+this if your extensions attempt to use pointers to any of the Perl API
+functions.
-(W misc) A warning peculiar to VMS. Perl tried to read the value of an %ENV
-element from a CLI symbol table, and found a resultant string longer
-than 1024 characters. The return value has been truncated to 1024
-characters.
+Note that the above issue is not relevant to the default build of
+Perl, whose interfaces continue to match those of prior versions
+(but subject to the other options described here).
-=item Version number must be a constant number
+See L<perlguts/"The Perl API"> for detailed information on the
+ramifications of building Perl with this option.
-(P) The attempt to translate a C<use Module n.n LIST> statement into
-its equivalent C<BEGIN> block found an internal inconsistency with
-the version number.
+ NOTE: PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built
+ with one of -Dusethreads, -Dusemultiplicity, or both. It is not
+ intended to be enabled by users at this time.
-=back
+=item C<PERL_POLLUTE_MALLOC>
-=head1 Obsolete Diagnostics
+Enabling Perl's malloc in release 5.005 and earlier caused the namespace of
+the system's malloc family of functions to be usurped by the Perl versions,
+since by default they used the same names. Besides causing problems on
+platforms that do not allow these functions to be cleanly replaced, this
+also meant that the system versions could not be called in programs that
+used Perl's malloc. Previous versions of Perl have allowed this behaviour
+to be suppressed with the HIDEMYMALLOC and EMBEDMYMALLOC preprocessor
+definitions.
-=over 4
+As of release 5.6.0, Perl's malloc family of functions have default names
+distinct from the system versions. You need to explicitly compile perl with
+C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC
+and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now
+the default.
-=item Character class syntax [: :] is reserved for future extensions
+Note that these functions do B<not> constitute Perl's memory allocation API.
+See L<perlguts/"Memory Allocation"> for further information about that.
-(W) Within regular expression character classes ([]) the syntax beginning
-with "[:" and ending with ":]" is reserved for future extensions.
-If you need to represent those character sequences inside a regular
-expression character class, just quote the square brackets with the
-backslash: "\[:" and ":\]".
+=back
-=item Ill-formed logical name |%s| in prime_env_iter
+=head2 Compatible C Source API Changes
-(W) A warning peculiar to VMS. A logical name was encountered when preparing
-to iterate over %ENV which violates the syntactic rules governing logical
-names. Because it cannot be translated normally, it is skipped, and will not
-appear in %ENV. This may be a benign occurrence, as some software packages
-might directly modify logical name tables and introduce nonstandard names,
-or it may indicate that a logical name table has been corrupted.
+=over
-=item Probable precedence problem on %s
+=item C<PATCHLEVEL> is now C<PERL_VERSION>
-(W) The compiler found a bareword where it expected a conditional,
-which often indicates that an || or && was parsed as part of the
-last argument of the previous construct, for example:
+The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION>
+are now available by default from perl.h, and reflect the base revision,
+patchlevel, and subversion respectively. C<PERL_REVISION> had no
+prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were
+previously available as C<PATCHLEVEL> and C<SUBVERSION>.
- open FOO || die;
+The new names cause less pollution of the B<cpp> namespace and reflect what
+the numbers have come to stand for in common practice. For compatibility,
+the old names are still supported when F<patchlevel.h> is explicitly
+included (as required before), so there is no source incompatibility
+from the change.
-=item regexp too big
+=back
-(F) The current implementation of regular expressions uses shorts as
-address offsets within a string. Unfortunately this means that if
-the regular expression compiles to longer than 32767, it'll blow up.
-Usually when you want a regular expression this big, there is a better
-way to do it with multiple statements. See L<perlre>.
+=head2 Binary Incompatibilities
-=item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
+In general, the default build of this release is expected to be binary
+compatible for extensions built with the 5.005 release or its maintenance
+versions. However, specific platforms may have broken binary compatibility
+due to changes in the defaults used in hints files. Therefore, please be
+sure to always check the platform-specific README files for any notes to
+the contrary.
-(D) Perl versions before 5.004 misinterpreted any type marker followed
-by "$" and a digit. For example, "$$0" was incorrectly taken to mean
-"${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
+The usethreads or usemultiplicity builds are B<not> binary compatible
+with the corresponding builds in 5.005.
-However, the developers of Perl 5.004 could not fix this bug completely,
-because at least two widely-used modules depend on the old meaning of
-"$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
-old (broken) way inside strings; but it generates this message as a
-warning. And in Perl 5.005, this special treatment will cease.
+On platforms that require an explicit list of exports (AIX, OS/2 and Windows,
+among others), purely internal symbols such as parser functions and the
+run time opcodes are not exported by default. Perl 5.005 used to export
+all functions irrespective of whether they were considered part of the
+public API or not.
-=back
+For the full list of public API functions, see L<perlapi>.
=head1 Known Problems
=back
-=head1 BUGS
+=head1 Obsolete Diagnostics
+
+=over 4
+
+=item Character class syntax [: :] is reserved for future extensions
+
+(W) Within regular expression character classes ([]) the syntax beginning
+with "[:" and ending with ":]" is reserved for future extensions.
+If you need to represent those character sequences inside a regular
+expression character class, just quote the square brackets with the
+backslash: "\[:" and ":\]".
+
+=item Ill-formed logical name |%s| in prime_env_iter
+
+(W) A warning peculiar to VMS. A logical name was encountered when preparing
+to iterate over %ENV which violates the syntactic rules governing logical
+names. Because it cannot be translated normally, it is skipped, and will not
+appear in %ENV. This may be a benign occurrence, as some software packages
+might directly modify logical name tables and introduce nonstandard names,
+or it may indicate that a logical name table has been corrupted.
+
+=item Probable precedence problem on %s
+
+(W) The compiler found a bareword where it expected a conditional,
+which often indicates that an || or && was parsed as part of the
+last argument of the previous construct, for example:
+
+ open FOO || die;
+
+=item regexp too big
+
+(F) The current implementation of regular expressions uses shorts as
+address offsets within a string. Unfortunately this means that if
+the regular expression compiles to longer than 32767, it'll blow up.
+Usually when you want a regular expression this big, there is a better
+way to do it with multiple statements. See L<perlre>.
+
+=item Use of "$$<digit>" to mean "${$}<digit>" is deprecated
+
+(D) Perl versions before 5.004 misinterpreted any type marker followed
+by "$" and a digit. For example, "$$0" was incorrectly taken to mean
+"${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004.
+
+However, the developers of Perl 5.004 could not fix this bug completely,
+because at least two widely-used modules depend on the old meaning of
+"$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the
+old (broken) way inside strings; but it generates this message as a
+warning. And in Perl 5.005, this special treatment will cease.
+
+=back
+
+=head1 Reporting Bugs
If you find what you think is a bug, you might check the
articles recently posted to the comp.lang.perl.misc newsgroup.
projects / p5sagit/p5-mst-13.2.git / commitdiff