1PERL5100DELTA(1) Perl Programmers Reference Guide PERL5100DELTA(1)
2
6 perl5100delta - what is new for perl 5.10.0
7
9 This document describes the differences between the 5.8.8 release and
10 the 5.10.0 release.
11 Many of the bug fixes in 5.10.0 were already seen in the 5.8.X
13 maintenance releases; they are not duplicated here and are documented
14 in the set of man pages named perl58[1-8]?delta.
15
17 The "feature" pragma
18 The "feature" pragma is used to enable new syntax that would break
19 Perl's backwards-compatibility with older releases of the language.
20 It's a lexical pragma, like "strict" or "warnings".
21 Currently the following new features are available: "switch" (adds a
23 switch statement), "say" (adds a "say" built-in function), and "state"
24 (adds a "state" keyword for declaring "static" variables). Those
25 features are described in their own sections of this document.
26 The "feature" pragma is also implicitly loaded when you require a
28 minimal perl version (with the "use VERSION" construct) greater than,
29 or equal to, 5.9.5. See feature for details.
30 New -E command-line switch
32 -E is equivalent to -e, but it implicitly enables all optional features
33 (like "use feature ":5.10"").
34 Defined-or operator
36 A new operator "//" (defined-or) has been implemented. The following
37 expression:
38 $a // $b
40 is merely equivalent to
42 defined $a ? $a : $b
44 and the statement
46 $c //= $d;
48 can now be used instead of
50 $c = $d unless defined $c;
52 The "//" operator has the same precedence and associativity as "||".
54 Special care has been taken to ensure that this operator Do What You
55 Mean while not breaking old code, but some edge cases involving the
56 empty regular expression may now parse differently. See perlop for
57 details.
58 Switch and Smart Match operator
60 Perl 5 now has a switch statement. It's available when "use feature
61 'switch'" is in effect. This feature introduces three new keywords,
62 "given", "when", and "default":
63 given ($foo) {
65 when (/^abc/) { $abc = 1; }
66 when (/^def/) { $def = 1; }
67 when (/^xyz/) { $xyz = 1; }
68 default { $nothing = 1; }
69 }
70 A more complete description of how Perl matches the switch variable
72 against the "when" conditions is given in "Switch statements" in
73 perlsyn.
74 This kind of match is called smart match, and it's also possible to use
76 it outside of switch statements, via the new "~~" operator. See "Smart
77 matching in detail" in perlsyn.
78 This feature was contributed by Robin Houston.
80 Regular expressions
82 Recursive Patterns
83 It is now possible to write recursive patterns without using the
84 "(??{})" construct. This new way is more efficient, and in many
85 cases easier to read.
86 Each capturing parenthesis can now be treated as an independent
88 pattern that can be entered by using the "(?PARNO)" syntax ("PARNO"
89 standing for "parenthesis number"). For example, the following
90 pattern will match nested balanced angle brackets:
91 /
93 ^ # start of line
94 ( # start capture buffer 1
95 < # match an opening angle bracket
96 (?: # match one of:
97 (?> # don't backtrack over the inside of this group
98 [^<>]+ # one or more non angle brackets
99 ) # end non backtracking group
100 | # ... or ...
101 (?1) # recurse to bracket 1 and try it again
102 )* # 0 or more times.
103 > # match a closing angle bracket
104 ) # end capture buffer one
105 $ # end of line
106 /x
107 PCRE users should note that Perl's recursive regex feature allows
109 backtracking into a recursed pattern, whereas in PCRE the recursion
110 is atomic or "possessive" in nature. As in the example above, you
111 can add (?>) to control this selectively. (Yves Orton)
112 Named Capture Buffers
114 It is now possible to name capturing parenthesis in a pattern and
115 refer to the captured contents by name. The naming syntax is
116 "(?<NAME>....)". It's possible to backreference to a named buffer
117 with the "\k<NAME>" syntax. In code, the new magical hashes "%+"
118 and "%-" can be used to access the contents of the capture buffers.
119 Thus, to replace all doubled chars with a single copy, one could
121 write
122 s/(?<letter>.)\k<letter>/$+{letter}/g
124 Only buffers with defined contents will be "visible" in the "%+"
126 hash, so it's possible to do something like
127 foreach my $name (keys %+) {
129 print "content of buffer '$name' is $+{$name}\n";
130 }
131 The "%-" hash is a bit more complete, since it will contain array
133 refs holding values from all capture buffers similarly named, if
134 there should be many of them.
135 "%+" and "%-" are implemented as tied hashes through the new module
137 "Tie::Hash::NamedCapture".
138 Users exposed to the .NET regex engine will find that the perl
140 implementation differs in that the numerical ordering of the
141 buffers is sequential, and not "unnamed first, then named". Thus in
142 the pattern
143 /(A)(?<B>B)(C)(?<D>D)/
145 $1 will be 'A', $2 will be 'B', $3 will be 'C' and $4 will be 'D'
147 and not $1 is 'A', $2 is 'C' and $3 is 'B' and $4 is 'D' that a
148 .NET programmer would expect. This is considered a feature. :-)
149 (Yves Orton)
150 Possessive Quantifiers
152 Perl now supports the "possessive quantifier" syntax of the "atomic
153 match" pattern. Basically a possessive quantifier matches as much
154 as it can and never gives any back. Thus it can be used to control
155 backtracking. The syntax is similar to non-greedy matching, except
156 instead of using a '?' as the modifier the '+' is used. Thus "?+",
157 "*+", "++", "{min,max}+" are now legal quantifiers. (Yves Orton)
158 Backtracking control verbs
160 The regex engine now supports a number of special-purpose backtrack
161 control verbs: (*THEN), (*PRUNE), (*MARK), (*SKIP), (*COMMIT),
162 (*FAIL) and (*ACCEPT). See perlre for their descriptions. (Yves
163 Orton)
164 Relative backreferences
166 A new syntax "\g{N}" or "\gN" where "N" is a decimal integer allows
167 a safer form of back-reference notation as well as allowing
168 relative backreferences. This should make it easier to generate and
169 embed patterns that contain backreferences. See "Capture buffers"
170 in perlre. (Yves Orton)
171 "\K" escape
173 The functionality of Jeff Pinyan's module Regexp::Keep has been
174 added to the core. In regular expressions you can now use the
175 special escape "\K" as a way to do something like floating length
176 positive lookbehind. It is also useful in substitutions like:
177 s/(foo)bar/$1/g
179 that can now be converted to
181 s/foo\Kbar//g
183 which is much more efficient. (Yves Orton)
185 Vertical and horizontal whitespace, and linebreak
187 Regular expressions now recognize the "\v" and "\h" escapes that
188 match vertical and horizontal whitespace, respectively. "\V" and
189 "\H" logically match their complements.
190 "\R" matches a generic linebreak, that is, vertical whitespace,
192 plus the multi-character sequence "\x0D\x0A".
193 Optional pre-match and post-match captures with the /p flag
195 There is a new flag "/p" for regular expressions. Using this makes
196 the engine preserve a copy of the part of the matched string before
197 the matching substring to the new special variable "${^PREMATCH}",
198 the part after the matching substring to "${^POSTMATCH}", and the
199 matched substring itself to "${^MATCH}".
200 Perl is still able to store these substrings to the special
202 variables "$`", "$'", $&, but using these variables anywhere in the
203 program adds a penalty to all regular expression matches, whereas
204 if you use the "/p" flag and the new special variables instead, you
205 pay only for the regular expressions where the flag is used.
206 For more detail on the new variables, see perlvar; for the use of
208 the regular expression flag, see perlop and perlre.
209 "say()"
211 say() is a new built-in, only available when "use feature 'say'" is in
212 effect, that is similar to print(), but that implicitly appends a
213 newline to the printed string. See "say" in perlfunc. (Robin Houston)
214 Lexical $_
216 The default variable $_ can now be lexicalized, by declaring it like
217 any other lexical variable, with a simple
218 my $_;
220 The operations that default on $_ will use the lexically-scoped version
222 of $_ when it exists, instead of the global $_.
223 In a "map" or a "grep" block, if $_ was previously my'ed, then the $_
225 inside the block is lexical as well (and scoped to the block).
226 In a scope where $_ has been lexicalized, you can still have access to
228 the global version of $_ by using $::_, or, more simply, by overriding
229 the lexical declaration with "our $_". (Rafael Garcia-Suarez)
230 The "_" prototype
232 A new prototype character has been added. "_" is equivalent to "$" but
233 defaults to $_ if the corresponding argument isn't supplied (both "$"
234 and "_" denote a scalar). Due to the optional nature of the argument,
235 you can only use it at the end of a prototype, or before a semicolon.
236 This has a small incompatible consequence: the prototype() function has
238 been adjusted to return "_" for some built-ins in appropriate cases
239 (for example, "prototype('CORE::rmdir')"). (Rafael Garcia-Suarez)
240 UNITCHECK blocks
242 "UNITCHECK", a new special code block has been introduced, in addition
243 to "BEGIN", "CHECK", "INIT" and "END".
244 "CHECK" and "INIT" blocks, while useful for some specialized purposes,
246 are always executed at the transition between the compilation and the
247 execution of the main program, and thus are useless whenever code is
248 loaded at runtime. On the other hand, "UNITCHECK" blocks are executed
249 just after the unit which defined them has been compiled. See perlmod
250 for more information. (Alex Gough)
251 New Pragma, "mro"
253 A new pragma, "mro" (for Method Resolution Order) has been added. It
254 permits to switch, on a per-class basis, the algorithm that perl uses
255 to find inherited methods in case of a multiple inheritance hierarchy.
256 The default MRO hasn't changed (DFS, for Depth First Search). Another
257 MRO is available: the C3 algorithm. See mro for more information.
258 (Brandon Black)
259 Note that, due to changes in the implementation of class hierarchy
261 search, code that used to undef the *ISA glob will most probably break.
262 Anyway, undef'ing *ISA had the side-effect of removing the magic on the
263 @ISA array and should not have been done in the first place. Also, the
264 cache *::ISA::CACHE:: no longer exists; to force reset the @ISA cache,
265 you now need to use the "mro" API, or more simply to assign to @ISA
266 (e.g. with "@ISA = @ISA").
267 readdir() may return a "short filename" on Windows
269 The readdir() function may return a "short filename" when the long
270 filename contains characters outside the ANSI codepage. Similarly
271 Cwd::cwd() may return a short directory name, and glob() may return
272 short names as well. On the NTFS file system these short names can
273 always be represented in the ANSI codepage. This will not be true for
274 all other file system drivers; e.g. the FAT filesystem stores short
275 filenames in the OEM codepage, so some files on FAT volumes remain
276 inaccessible through the ANSI APIs.
277 Similarly, $^X, @INC, and $ENV{PATH} are preprocessed at startup to
279 make sure all paths are valid in the ANSI codepage (if possible).
280 The Win32::GetLongPathName() function now returns the UTF-8 encoded
282 correct long file name instead of using replacement characters to force
283 the name into the ANSI codepage. The new Win32::GetANSIPathName()
284 function can be used to turn a long pathname into a short one only if
285 the long one cannot be represented in the ANSI codepage.
286 Many other functions in the "Win32" module have been improved to accept
288 UTF-8 encoded arguments. Please see Win32 for details.
289 readpipe() is now overridable
291 The built-in function readpipe() is now overridable. Overriding it
292 permits also to override its operator counterpart, "qx//" (a.k.a.
293 "``"). Moreover, it now defaults to $_ if no argument is provided.
294 (Rafael Garcia-Suarez)
295 Default argument for readline()
297 readline() now defaults to *ARGV if no argument is provided. (Rafael
298 Garcia-Suarez)
299 state() variables
301 A new class of variables has been introduced. State variables are
302 similar to "my" variables, but are declared with the "state" keyword in
303 place of "my". They're visible only in their lexical scope, but their
304 value is persistent: unlike "my" variables, they're not undefined at
305 scope entry, but retain their previous value. (Rafael Garcia-Suarez,
306 Nicholas Clark)
307 To use state variables, one needs to enable them by using
309 use feature 'state';
311 or by using the "-E" command-line switch in one-liners. See
313 "Persistent Private Variables" in perlsub.
314 Stacked filetest operators
316 As a new form of syntactic sugar, it's now possible to stack up
317 filetest operators. You can now write "-f -w -x $file" in a row to mean
318 "-x $file && -w _ && -f _". See "-X" in perlfunc.
319 UNIVERSAL::DOES()
321 The "UNIVERSAL" class has a new method, "DOES()". It has been added to
322 solve semantic problems with the "isa()" method. "isa()" checks for
323 inheritance, while "DOES()" has been designed to be overridden when
324 module authors use other types of relations between classes (in
325 addition to inheritance). (chromatic)
326 See "$obj->DOES( ROLE )" in UNIVERSAL.
328 Formats
330 Formats were improved in several ways. A new field, "^*", can be used
331 for variable-width, one-line-at-a-time text. Null characters are now
332 handled correctly in picture lines. Using "@#" and "~~" together will
333 now produce a compile-time error, as those format fields are
334 incompatible. perlform has been improved, and miscellaneous bugs
335 fixed.
336 Byte-order modifiers for pack() and unpack()
338 There are two new byte-order modifiers, ">" (big-endian) and "<"
339 (little-endian), that can be appended to most pack() and unpack()
340 template characters and groups to force a certain byte-order for that
341 type or group. See "pack" in perlfunc and perlpacktut for details.
342 "no VERSION"
344 You can now use "no" followed by a version number to specify that you
345 want to use a version of perl older than the specified one.
346 "chdir", "chmod" and "chown" on filehandles
348 "chdir", "chmod" and "chown" can now work on filehandles as well as
349 filenames, if the system supports respectively "fchdir", "fchmod" and
350 "fchown", thanks to a patch provided by Gisle Aas.
351 OS groups
353 $( and $) now return groups in the order where the OS returns them,
354 thanks to Gisle Aas. This wasn't previously the case.
355 Recursive sort subs
357 You can now use recursive subroutines with sort(), thanks to Robin
358 Houston.
359 Exceptions in constant folding
361 The constant folding routine is now wrapped in an exception handler,
362 and if folding throws an exception (such as attempting to evaluate
363 0/0), perl now retains the current optree, rather than aborting the
364 whole program. Without this change, programs would not compile if they
365 had expressions that happened to generate exceptions, even though those
366 expressions were in code that could never be reached at runtime.
367 (Nicholas Clark, Dave Mitchell)
368 Source filters in @INC
370 It's possible to enhance the mechanism of subroutine hooks in @INC by
371 adding a source filter on top of the filehandle opened and returned by
372 the hook. This feature was planned a long time ago, but wasn't quite
373 working until now. See "require" in perlfunc for details. (Nicholas
374 Clark)
375 New internal variables
377 "${^RE_DEBUG_FLAGS}"
378 This variable controls what debug flags are in effect for the
379 regular expression engine when running under "use re "debug"". See
380 re for details.
381 "${^CHILD_ERROR_NATIVE}"
383 This variable gives the native status returned by the last pipe
384 close, backtick command, successful call to wait() or waitpid(), or
385 from the system() operator. See perlvar for details. (Contributed
386 by Gisle Aas.)
387 "${^RE_TRIE_MAXBUF}"
389 See "Trie optimisation of literal string alternations".
390 "${^WIN32_SLOPPY_STAT}"
392 See "Sloppy stat on Windows".
393 Miscellaneous
395 "unpack()" now defaults to unpacking the $_ variable.
396 "mkdir()" without arguments now defaults to $_.
398 The internal dump output has been improved, so that non-printable
400 characters such as newline and backspace are output in "\x" notation,
401 rather than octal.
402 The -C option can no longer be used on the "#!" line. It wasn't working
404 there anyway, since the standard streams are already set up at this
405 point in the execution of the perl interpreter. You can use binmode()
406 instead to get the desired behaviour.
407 UCD 5.0.0
409 The copy of the Unicode Character Database included in Perl 5 has been
410 updated to version 5.0.0.
411 MAD
413 MAD, which stands for Miscellaneous Attribute Decoration, is a still-
414 in-development work leading to a Perl 5 to Perl 6 converter. To enable
415 it, it's necessary to pass the argument "-Dmad" to Configure. The
416 obtained perl isn't binary compatible with a regular perl 5.10, and has
417 space and speed penalties; moreover not all regression tests still pass
418 with it. (Larry Wall, Nicholas Clark)
419 kill() on Windows
421 On Windows platforms, "kill(-9, $pid)" now kills a process tree. (On
422 Unix, this delivers the signal to all processes in the same process
423 group.)
424
426 Packing and UTF-8 strings
427 The semantics of pack() and unpack() regarding UTF-8-encoded data has
428 been changed. Processing is now by default character per character
429 instead of byte per byte on the underlying encoding. Notably, code that
430 used things like "pack("a*", $string)" to see through the encoding of
431 string will now simply get back the original $string. Packed strings
432 can also get upgraded during processing when you store upgraded
433 characters. You can get the old behaviour by using "use bytes".
434 To be consistent with pack(), the "C0" in unpack() templates indicates
436 that the data is to be processed in character mode, i.e. character by
437 character; on the contrary, "U0" in unpack() indicates UTF-8 mode,
438 where the packed string is processed in its UTF-8-encoded Unicode form
439 on a byte by byte basis. This is reversed with regard to perl 5.8.X,
440 but now consistent between pack() and unpack().
441 Moreover, "C0" and "U0" can also be used in pack() templates to specify
443 respectively character and byte modes.
444 "C0" and "U0" in the middle of a pack or unpack format now switch to
446 the specified encoding mode, honoring parens grouping. Previously,
447 parens were ignored.
448 Also, there is a new pack() character format, "W", which is intended to
450 replace the old "C". "C" is kept for unsigned chars coded as bytes in
451 the strings internal representation. "W" represents unsigned (logical)
452 character values, which can be greater than 255. It is therefore more
453 robust when dealing with potentially UTF-8-encoded data (as "C" will
454 wrap values outside the range 0..255, and not respect the string
455 encoding).
456 In practice, that means that pack formats are now encoding-neutral,
458 except "C".
459 For consistency, "A" in unpack() format now trims all Unicode
461 whitespace from the end of the string. Before perl 5.9.2, it used to
462 strip only the classical ASCII space characters.
463 Byte/character count feature in unpack()
465 A new unpack() template character, ".", returns the number of bytes or
466 characters (depending on the selected encoding mode, see above) read so
467 far.
468 The $* and $# variables have been removed
470 $*, which was deprecated in favor of the "/s" and "/m" regexp
471 modifiers, has been removed.
472 The deprecated $# variable (output format for numbers) has been
474 removed.
475 Two new severe warnings, "$#/$* is no longer supported", have been
477 added.
478 substr() lvalues are no longer fixed-length
480 The lvalues returned by the three argument form of substr() used to be
481 a "fixed length window" on the original string. In some cases this
482 could cause surprising action at distance or other undefined behaviour.
483 Now the length of the window adjusts itself to the length of the string
484 assigned to it.
485 Parsing of "-f _"
487 The identifier "_" is now forced to be a bareword after a filetest
488 operator. This solves a number of misparsing issues when a global "_"
489 subroutine is defined.
490 ":unique"
492 The ":unique" attribute has been made a no-op, since its current
493 implementation was fundamentally flawed and not threadsafe.
494 Effect of pragmas in eval
496 The compile-time value of the "%^H" hint variable can now propagate
497 into eval("")uated code. This makes it more useful to implement lexical
498 pragmas.
499 As a side-effect of this, the overloaded-ness of constants now
501 propagates into eval("").
502 chdir FOO
504 A bareword argument to chdir() is now recognized as a file handle.
505 Earlier releases interpreted the bareword as a directory name. (Gisle
506 Aas)
507 Handling of .pmc files
509 An old feature of perl was that before "require" or "use" look for a
510 file with a .pm extension, they will first look for a similar filename
511 with a .pmc extension. If this file is found, it will be loaded in
512 place of any potentially existing file ending in a .pm extension.
513 Previously, .pmc files were loaded only if more recent than the
515 matching .pm file. Starting with 5.9.4, they'll be always loaded if
516 they exist.
517 $^V is now a "version" object instead of a v-string
519 $^V can still be used with the %vd format in printf, but any character-
520 level operations will now access the string representation of the
521 "version" object and not the ordinals of a v-string. Expressions like
522 "substr($^V, 0, 2)" or "split //, $^V" no longer work and must be
523 rewritten.
524 @- and @+ in patterns
526 The special arrays "@-" and "@+" are no longer interpolated in regular
527 expressions. (Sadahiro Tomoyuki)
528 $AUTOLOAD can now be tainted
530 If you call a subroutine by a tainted name, and if it defers to an
531 AUTOLOAD function, then $AUTOLOAD will be (correctly) tainted. (Rick
532 Delaney)
533 Tainting and printf
535 When perl is run under taint mode, "printf()" and "sprintf()" will now
536 reject any tainted format argument. (Rafael Garcia-Suarez)
537 undef and signal handlers
539 Undefining or deleting a signal handler via "undef $SIG{FOO}" is now
540 equivalent to setting it to 'DEFAULT'. (Rafael Garcia-Suarez)
541 strictures and dereferencing in defined()
543 "use strict 'refs'" was ignoring taking a hard reference in an argument
544 to defined(), as in :
545 use strict 'refs';
547 my $x = 'foo';
548 if (defined $$x) {...}
549 This now correctly produces the run-time error "Can't use string as a
551 SCALAR ref while "strict refs" in use".
552 "defined @$foo" and "defined %$bar" are now also subject to "strict
554 'refs'" (that is, $foo and $bar shall be proper references there.)
555 ("defined(@foo)" and "defined(%bar)" are discouraged constructs
556 anyway.) (Nicholas Clark)
557 "(?p{})" has been removed
559 The regular expression construct "(?p{})", which was deprecated in perl
560 5.8, has been removed. Use "(??{})" instead. (Rafael Garcia-Suarez)
561 Pseudo-hashes have been removed
563 Support for pseudo-hashes has been removed from Perl 5.9. (The "fields"
564 pragma remains here, but uses an alternate implementation.)
565 Removal of the bytecode compiler and of perlcc
567 "perlcc", the byteloader and the supporting modules (B::C, B::CC,
568 B::Bytecode, etc.) are no longer distributed with the perl sources.
569 Those experimental tools have never worked reliably, and, due to the
570 lack of volunteers to keep them in line with the perl interpreter
571 developments, it was decided to remove them instead of shipping a
572 broken version of those. The last version of those modules can be
573 found with perl 5.9.4.
574 However the B compiler framework stays supported in the perl core, as
576 with the more useful modules it has permitted (among others, B::Deparse
577 and B::Concise).
578 Removal of the JPL
580 The JPL (Java-Perl Lingo) has been removed from the perl sources
581 tarball.
582 Recursive inheritance detected earlier
584 Perl will now immediately throw an exception if you modify any
585 package's @ISA in such a way that it would cause recursive inheritance.
586 Previously, the exception would not occur until Perl attempted to make
588 use of the recursive inheritance while resolving a method or doing a
589 "$foo->isa($bar)" lookup.
590 warnings::enabled and warnings::warnif changed to favor users of modules
592 The behaviour in 5.10.x favors the person using the module; The
593 behaviour in 5.8.x favors the module writer;
594 Assume the following code:
596 main calls Foo::Bar::baz()
598 Foo::Bar inherits from Foo::Base
599 Foo::Bar::baz() calls Foo::Base::_bazbaz()
600 Foo::Base::_bazbaz() calls: warnings::warnif('substr', 'some warning
601 message');
602 On 5.8.x, the code warns when Foo::Bar contains "use warnings;" It does
604 not matter if Foo::Base or main have warnings enabled to disable the
605 warning one has to modify Foo::Bar.
606 On 5.10.0 and newer, the code warns when main contains "use warnings;"
608 It does not matter if Foo::Base or Foo::Bar have warnings enabled to
609 disable the warning one has to modify main.
610
612 Upgrading individual core modules
613 Even more core modules are now also available separately through the
614 CPAN. If you wish to update one of these modules, you don't need to
615 wait for a new perl release. From within the cpan shell, running the
616 'r' command will report on modules with upgrades available. See
617 "perldoc CPAN" for more information.
618 Pragmata Changes
620 "feature"
621 The new pragma "feature" is used to enable new features that might
622 break old code. See "The "feature" pragma" above.
623 "mro"
625 This new pragma enables to change the algorithm used to resolve
626 inherited methods. See "New Pragma, "mro"" above.
627 Scoping of the "sort" pragma
629 The "sort" pragma is now lexically scoped. Its effect used to be
630 global.
631 Scoping of "bignum", "bigint", "bigrat"
633 The three numeric pragmas "bignum", "bigint" and "bigrat" are now
634 lexically scoped. (Tels)
635 "base"
637 The "base" pragma now warns if a class tries to inherit from
638 itself. (Curtis "Ovid" Poe)
639 "strict" and "warnings"
641 "strict" and "warnings" will now complain loudly if they are loaded
642 via incorrect casing (as in "use Strict;"). (Johan Vromans)
643 "version"
645 The "version" module provides support for version objects.
646 "warnings"
648 The "warnings" pragma doesn't load "Carp" anymore. That means that
649 code that used "Carp" routines without having loaded it at compile
650 time might need to be adjusted; typically, the following (faulty)
651 code won't work anymore, and will require parentheses to be added
652 after the function name:
653 use warnings;
655 require Carp;
656 Carp::confess 'argh';
657 "less"
659 "less" now does something useful (or at least it tries to). In
660 fact, it has been turned into a lexical pragma. So, in your
661 modules, you can now test whether your users have requested to use
662 less CPU, or less memory, less magic, or maybe even less fat. See
663 less for more. (Joshua ben Jore)
664 New modules
666 • "encoding::warnings", by Audrey Tang, is a module to emit warnings
667 whenever an ASCII character string containing high-bit bytes is
668 implicitly converted into UTF-8. It's a lexical pragma since Perl
669 5.9.4; on older perls, its effect is global.
670 • "Module::CoreList", by Richard Clamp, is a small handy module that
672 tells you what versions of core modules ship with any versions of
673 Perl 5. It comes with a command-line frontend, "corelist".
674 • "Math::BigInt::FastCalc" is an XS-enabled, and thus faster, version
676 of "Math::BigInt::Calc".
677 • "Compress::Zlib" is an interface to the zlib compression library.
679 It comes with a bundled version of zlib, so having a working zlib
680 is not a prerequisite to install it. It's used by "Archive::Tar"
681 (see below).
682 • "IO::Zlib" is an "IO::"-style interface to "Compress::Zlib".
684 • "Archive::Tar" is a module to manipulate "tar" archives.
686 • "Digest::SHA" is a module used to calculate many types of SHA
688 digests, has been included for SHA support in the CPAN module.
689 • "ExtUtils::CBuilder" and "ExtUtils::ParseXS" have been added.
691 • "Hash::Util::FieldHash", by Anno Siegel, has been added. This
693 module provides support for field hashes: hashes that maintain an
694 association of a reference with a value, in a thread-safe garbage-
695 collected way. Such hashes are useful to implement inside-out
696 objects.
697 • "Module::Build", by Ken Williams, has been added. It's an
699 alternative to "ExtUtils::MakeMaker" to build and install perl
700 modules.
701 • "Module::Load", by Jos Boumans, has been added. It provides a
703 single interface to load Perl modules and .pl files.
704 • "Module::Loaded", by Jos Boumans, has been added. It's used to mark
706 modules as loaded or unloaded.
707 • "Package::Constants", by Jos Boumans, has been added. It's a simple
709 helper to list all constants declared in a given package.
710 • "Win32API::File", by Tye McQueen, has been added (for Windows
712 builds). This module provides low-level access to Win32 system API
713 calls for files/dirs.
714 • "Locale::Maketext::Simple", needed by CPANPLUS, is a simple wrapper
716 around "Locale::Maketext::Lexicon". Note that
717 "Locale::Maketext::Lexicon" isn't included in the perl core; the
718 behaviour of "Locale::Maketext::Simple" gracefully degrades when
719 the later isn't present.
720 • "Params::Check" implements a generic input parsing/checking
722 mechanism. It is used by CPANPLUS.
723 • "Term::UI" simplifies the task to ask questions at a terminal
725 prompt.
726 • "Object::Accessor" provides an interface to create per-object
728 accessors.
729 • "Module::Pluggable" is a simple framework to create modules that
731 accept pluggable sub-modules.
732 • "Module::Load::Conditional" provides simple ways to query and
734 possibly load installed modules.
735 • "Time::Piece" provides an object oriented interface to time
737 functions, overriding the built-ins localtime() and gmtime().
738 • "IPC::Cmd" helps to find and run external commands, possibly
740 interactively.
741 • "File::Fetch" provide a simple generic file fetching mechanism.
743 • "Log::Message" and "Log::Message::Simple" are used by the log
745 facility of "CPANPLUS".
746 • "Archive::Extract" is a generic archive extraction mechanism for
748 .tar (plain, gzipped or bzipped) or .zip files.
749 • "CPANPLUS" provides an API and a command-line tool to access the
751 CPAN mirrors.
752 • "Pod::Escapes" provides utilities that are useful in decoding Pod
754 E<...> sequences.
755 • "Pod::Simple" is now the backend for several of the Pod-related
757 modules included with Perl.
758 Selected Changes to Core Modules
760 "Attribute::Handlers"
761 "Attribute::Handlers" can now report the caller's file and line
762 number. (David Feldman)
763 All interpreted attributes are now passed as array references.
765 (Damian Conway)
766 "B::Lint"
768 "B::Lint" is now based on "Module::Pluggable", and so can be
769 extended with plugins. (Joshua ben Jore)
770 "B" It's now possible to access the lexical pragma hints ("%^H") by
772 using the method B::COP::hints_hash(). It returns a "B::RHE"
773 object, which in turn can be used to get a hash reference via the
774 method B::RHE::HASH(). (Joshua ben Jore)
775 "Thread"
777 As the old 5005thread threading model has been removed, in favor of
778 the ithreads scheme, the "Thread" module is now a compatibility
779 wrapper, to be used in old code only. It has been removed from the
780 default list of dynamic extensions.
781
783 perl -d
784 The Perl debugger can now save all debugger commands for sourcing
785 later; notably, it can now emulate stepping backwards, by
786 restarting and rerunning all bar the last command from a saved
787 command history.
788 It can also display the parent inheritance tree of a given class,
790 with the "i" command.
791 ptar
793 "ptar" is a pure perl implementation of "tar" that comes with
794 "Archive::Tar".
795 ptardiff
797 "ptardiff" is a small utility used to generate a diff between the
798 contents of a tar archive and a directory tree. Like "ptar", it
799 comes with "Archive::Tar".
800 shasum
802 "shasum" is a command-line utility, used to print or to check SHA
803 digests. It comes with the new "Digest::SHA" module.
804 corelist
806 The "corelist" utility is now installed with perl (see "New
807 modules" above).
808 h2ph and h2xs
810 "h2ph" and "h2xs" have been made more robust with regard to
811 "modern" C code.
812 "h2xs" implements a new option "--use-xsloader" to force use of
814 "XSLoader" even in backwards compatible modules.
815 The handling of authors' names that had apostrophes has been fixed.
817 Any enums with negative values are now skipped.
819 perlivp
821 "perlivp" no longer checks for *.ph files by default. Use the new
822 "-a" option to run all tests.
823 find2perl
825 "find2perl" now assumes "-print" as a default action. Previously,
826 it needed to be specified explicitly.
827 Several bugs have been fixed in "find2perl", regarding "-exec" and
829 "-eval". Also the options "-path", "-ipath" and "-iname" have been
830 added.
831 config_data
833 "config_data" is a new utility that comes with "Module::Build". It
834 provides a command-line interface to the configuration of Perl
835 modules that use Module::Build's framework of configurability (that
836 is, *::ConfigData modules that contain local configuration
837 information for their parent modules.)
838 cpanp
840 "cpanp", the CPANPLUS shell, has been added. ("cpanp-run-perl", a
841 helper for CPANPLUS operation, has been added too, but isn't
842 intended for direct use).
843 cpan2dist
845 "cpan2dist" is a new utility that comes with CPANPLUS. It's a tool
846 to create distributions (or packages) from CPAN modules.
847 pod2html
849 The output of "pod2html" has been enhanced to be more customizable
850 via CSS. Some formatting problems were also corrected. (Jari Aalto)
851
853 The perlpragma manpage documents how to write one's own lexical pragmas
854 in pure Perl (something that is possible starting with 5.9.4).
855 The new perlglossary manpage is a glossary of terms used in the Perl
857 documentation, technical and otherwise, kindly provided by O'Reilly
858 Media, Inc.
859 The perlreguts manpage, courtesy of Yves Orton, describes internals of
861 the Perl regular expression engine.
862 The perlreapi manpage describes the interface to the perl interpreter
864 used to write pluggable regular expression engines (by AEvar Arnfjoerd`
865 Bjarmason).
866 The perlunitut manpage is a tutorial for programming with Unicode and
868 string encodings in Perl, courtesy of Juerd Waalboer.
869 A new manual page, perlunifaq (the Perl Unicode FAQ), has been added
871 (Juerd Waalboer).
872 The perlcommunity manpage gives a description of the Perl community on
874 the Internet and in real life. (Edgar "Trizor" Bering)
875 The CORE manual page documents the "CORE::" namespace. (Tels)
877 The long-existing feature of "/(?{...})/" regexps setting $_ and pos()
879 is now documented.
880
882 In-place sorting
883 Sorting arrays in place ("@a = sort @a") is now optimized to avoid
884 making a temporary copy of the array.
885 Likewise, "reverse sort ..." is now optimized to sort in reverse,
887 avoiding the generation of a temporary intermediate list.
888 Lexical array access
890 Access to elements of lexical arrays via a numeric constant between 0
891 and 255 is now faster. (This used to be only the case for global
892 arrays.)
893 XS-assisted SWASHGET
895 Some pure-perl code that perl was using to retrieve Unicode properties
896 and transliteration mappings has been reimplemented in XS.
897 Constant subroutines
899 The interpreter internals now support a far more memory efficient form
900 of inlineable constants. Storing a reference to a constant value in a
901 symbol table is equivalent to a full typeglob referencing a constant
902 subroutine, but using about 400 bytes less memory. This proxy constant
903 subroutine is automatically upgraded to a real typeglob with subroutine
904 if necessary. The approach taken is analogous to the existing space
905 optimisation for subroutine stub declarations, which are stored as
906 plain scalars in place of the full typeglob.
907 Several of the core modules have been converted to use this feature for
909 their system dependent constants - as a result "use POSIX;" now takes
910 about 200K less memory.
911 "PERL_DONT_CREATE_GVSV"
913 The new compilation flag "PERL_DONT_CREATE_GVSV", introduced as an
914 option in perl 5.8.8, is turned on by default in perl 5.9.3. It
915 prevents perl from creating an empty scalar with every new typeglob.
916 See perl589delta for details.
917 Weak references are cheaper
919 Weak reference creation is now O(1) rather than O(n), courtesy of
920 Nicholas Clark. Weak reference deletion remains O(n), but if deletion
921 only happens at program exit, it may be skipped completely.
922 sort() enhancements
924 Salvador Fandin~o provided improvements to reduce the memory usage of
925 "sort" and to speed up some cases.
926 Memory optimisations
928 Several internal data structures (typeglobs, GVs, CVs, formats) have
929 been restructured to use less memory. (Nicholas Clark)
930 UTF-8 cache optimisation
932 The UTF-8 caching code is now more efficient, and used more often.
933 (Nicholas Clark)
934 Sloppy stat on Windows
936 On Windows, perl's stat() function normally opens the file to determine
937 the link count and update attributes that may have been changed through
938 hard links. Setting ${^WIN32_SLOPPY_STAT} to a true value speeds up
939 stat() by not performing this operation. (Jan Dubois)
940 Regular expressions optimisations
942 Engine de-recursivised
943 The regular expression engine is no longer recursive, meaning that
944 patterns that used to overflow the stack will either die with
945 useful explanations, or run to completion, which, since they were
946 able to blow the stack before, will likely take a very long time to
947 happen. If you were experiencing the occasional stack overflow (or
948 segfault) and upgrade to discover that now perl apparently hangs
949 instead, look for a degenerate regex. (Dave Mitchell)
950 Single char char-classes treated as literals
952 Classes of a single character are now treated the same as if the
953 character had been used as a literal, meaning that code that uses
954 char-classes as an escaping mechanism will see a speedup. (Yves
955 Orton)
956 Trie optimisation of literal string alternations
958 Alternations, where possible, are optimised into more efficient
959 matching structures. String literal alternations are merged into a
960 trie and are matched simultaneously. This means that instead of
961 O(N) time for matching N alternations at a given point, the new
962 code performs in O(1) time. A new special variable,
963 ${^RE_TRIE_MAXBUF}, has been added to fine-tune this optimization.
964 (Yves Orton)
965 Note: Much code exists that works around perl's historic poor
967 performance on alternations. Often the tricks used to do so will
968 disable the new optimisations. Hopefully the utility modules used
969 for this purpose will be educated about these new optimisations.
970 Aho-Corasick start-point optimisation
972 When a pattern starts with a trie-able alternation and there aren't
973 better optimisations available, the regex engine will use Aho-
974 Corasick matching to find the start point. (Yves Orton)
975
977 Configuration improvements
978 "-Dusesitecustomize"
979 Run-time customization of @INC can be enabled by passing the
980 "-Dusesitecustomize" flag to Configure. When enabled, this will
981 make perl run $sitelibexp/sitecustomize.pl before anything else.
982 This script can then be set up to add additional entries to @INC.
983 Relocatable installations
985 There is now Configure support for creating a relocatable perl
986 tree. If you Configure with "-Duserelocatableinc", then the paths
987 in @INC (and everything else in %Config) can be optionally located
988 via the path of the perl executable.
989 That means that, if the string ".../" is found at the start of any
991 path, it's substituted with the directory of $^X. So, the
992 relocation can be configured on a per-directory basis, although the
993 default with "-Duserelocatableinc" is that everything is relocated.
994 The initial install is done to the original configured prefix.
995 strlcat() and strlcpy()
997 The configuration process now detects whether strlcat() and
998 strlcpy() are available. When they are not available, perl's own
999 version is used (from Russ Allbery's public domain implementation).
1000 Various places in the perl interpreter now use them. (Steve Peters)
1001 "d_pseudofork" and "d_printf_format_null"
1003 A new configuration variable, available as $Config{d_pseudofork} in
1004 the Config module, has been added, to distinguish real fork()
1005 support from fake pseudofork used on Windows platforms.
1006 A new configuration variable, "d_printf_format_null", has been
1008 added, to see if printf-like formats are allowed to be NULL.
1009 Configure help
1011 "Configure -h" has been extended with the most commonly used
1012 options.
1013 Compilation improvements
1015 Parallel build
1016 Parallel makes should work properly now, although there may still
1017 be problems if "make test" is instructed to run in parallel.
1018 Borland's compilers support
1020 Building with Borland's compilers on Win32 should work more
1021 smoothly. In particular Steve Hay has worked to side step many
1022 warnings emitted by their compilers and at least one C compiler
1023 internal error.
1024 Static build on Windows
1026 Perl extensions on Windows now can be statically built into the
1027 Perl DLL.
1028 Also, it's now possible to build a "perl-static.exe" that doesn't
1030 depend on the Perl DLL on Win32. See the Win32 makefiles for
1031 details. (Vadim Konovalov)
1032 ppport.h files
1034 All ppport.h files in the XS modules bundled with perl are now
1035 autogenerated at build time. (Marcus Holland-Moritz)
1036 C++ compatibility
1038 Efforts have been made to make perl and the core XS modules
1039 compilable with various C++ compilers (although the situation is
1040 not perfect with some of the compilers on some of the platforms
1041 tested.)
1042 Support for Microsoft 64-bit compiler
1044 Support for building perl with Microsoft's 64-bit compiler has been
1045 improved. (ActiveState)
1046 Visual C++
1048 Perl can now be compiled with Microsoft Visual C++ 2005 (and 2008
1049 Beta 2).
1050 Win32 builds
1052 All win32 builds (MS-Win, WinCE) have been merged and cleaned up.
1053 Installation improvements
1055 Module auxiliary files
1056 README files and changelogs for CPAN modules bundled with perl are
1057 no longer installed.
1058 New Or Improved Platforms
1060 Perl has been reported to work on Symbian OS. See perlsymbian for more
1061 information.
1062 Many improvements have been made towards making Perl work correctly on
1064 z/OS.
1065 Perl has been reported to work on DragonFlyBSD and MidnightBSD.
1067 Perl has also been reported to work on NexentaOS (
1069 http://www.gnusolaris.org/ ).
1070 The VMS port has been improved. See perlvms.
1072 Support for Cray XT4 Catamount/Qk has been added. See
1074 hints/catamount.sh in the source code distribution for more
1075 information.
1076 Vendor patches have been merged for RedHat and Gentoo.
1078 DynaLoader::dl_unload_file() now works on Windows.
1080
1082 strictures in regexp-eval blocks
1083 "strict" wasn't in effect in regexp-eval blocks ("/(?{...})/").
1084 Calling CORE::require()
1086 CORE::require() and CORE::do() were always parsed as require() and
1087 do() when they were overridden. This is now fixed.
1088 Subscripts of slices
1090 You can now use a non-arrowed form for chained subscripts after a
1091 list slice, like in:
1092 ({foo => "bar"})[0]{foo}
1094 This used to be a syntax error; a "->" was required.
1096 "no warnings 'category'" works correctly with -w
1098 Previously when running with warnings enabled globally via "-w",
1099 selective disabling of specific warning categories would actually
1100 turn off all warnings. This is now fixed; now "no warnings 'io';"
1101 will only turn off warnings in the "io" class. Previously it would
1102 erroneously turn off all warnings.
1103 threads improvements
1105 Several memory leaks in ithreads were closed. Also, ithreads were
1106 made less memory-intensive.
1107 "threads" is now a dual-life module, also available on CPAN. It has
1109 been expanded in many ways. A kill() method is available for thread
1110 signalling. One can get thread status, or the list of running or
1111 joinable threads.
1112 A new "threads->exit()" method is used to exit from the application
1114 (this is the default for the main thread) or from the current
1115 thread only (this is the default for all other threads). On the
1116 other hand, the exit() built-in now always causes the whole
1117 application to terminate. (Jerry D. Hedden)
1118 chr() and negative values
1120 chr() on a negative value now gives "\x{FFFD}", the Unicode
1121 replacement character, unless when the "bytes" pragma is in effect,
1122 where the low eight bits of the value are used.
1123 PERL5SHELL and tainting
1125 On Windows, the PERL5SHELL environment variable is now checked for
1126 taintedness. (Rafael Garcia-Suarez)
1127 Using *FILE{IO}
1129 "stat()" and "-X" filetests now treat *FILE{IO} filehandles like
1130 *FILE filehandles. (Steve Peters)
1131 Overloading and reblessing
1133 Overloading now works when references are reblessed into another
1134 class. Internally, this has been implemented by moving the flag
1135 for "overloading" from the reference to the referent, which
1136 logically is where it should always have been. (Nicholas Clark)
1137 Overloading and UTF-8
1139 A few bugs related to UTF-8 handling with objects that have
1140 stringification overloaded have been fixed. (Nicholas Clark)
1141 eval memory leaks fixed
1143 Traditionally, "eval 'syntax error'" has leaked badly. Many (but
1144 not all) of these leaks have now been eliminated or reduced. (Dave
1145 Mitchell)
1146 Random device on Windows
1148 In previous versions, perl would read the file /dev/urandom if it
1149 existed when seeding its random number generator. That file is
1150 unlikely to exist on Windows, and if it did would probably not
1151 contain appropriate data, so perl no longer tries to read it on
1152 Windows. (Alex Davies)
1153 PERLIO_DEBUG
1155 The "PERLIO_DEBUG" environment variable no longer has any effect
1156 for setuid scripts and for scripts run with -T.
1157 Moreover, with a thread-enabled perl, using "PERLIO_DEBUG" could
1159 lead to an internal buffer overflow. This has been fixed.
1160 PerlIO::scalar and read-only scalars
1162 PerlIO::scalar will now prevent writing to read-only scalars.
1163 Moreover, seek() is now supported with PerlIO::scalar-based
1164 filehandles, the underlying string being zero-filled as needed.
1165 (Rafael, Jarkko Hietaniemi)
1166 study() and UTF-8
1168 study() never worked for UTF-8 strings, but could lead to false
1169 results. It's now a no-op on UTF-8 data. (Yves Orton)
1170 Critical signals
1172 The signals SIGILL, SIGBUS and SIGSEGV are now always delivered in
1173 an "unsafe" manner (contrary to other signals, that are deferred
1174 until the perl interpreter reaches a reasonably stable state; see
1175 "Deferred Signals (Safe Signals)" in perlipc). (Rafael)
1176 @INC-hook fix
1178 When a module or a file is loaded through an @INC-hook, and when
1179 this hook has set a filename entry in %INC, __FILE__ is now set for
1180 this module accordingly to the contents of that %INC entry.
1181 (Rafael)
1182 "-t" switch fix
1184 The "-w" and "-t" switches can now be used together without messing
1185 up which categories of warnings are activated. (Rafael)
1186 Duping UTF-8 filehandles
1188 Duping a filehandle which has the ":utf8" PerlIO layer set will now
1189 properly carry that layer on the duped filehandle. (Rafael)
1190 Localisation of hash elements
1192 Localizing a hash element whose key was given as a variable didn't
1193 work correctly if the variable was changed while the local() was in
1194 effect (as in "local $h{$x}; ++$x"). (Bo Lindbergh)
1195
1197 Use of uninitialized value
1198 Perl will now try to tell you the name of the variable (if any)
1199 that was undefined.
1200 Deprecated use of my() in false conditional
1202 A new deprecation warning, Deprecated use of my() in false
1203 conditional, has been added, to warn against the use of the dubious
1204 and deprecated construct
1205 my $x if 0;
1207 See perldiag. Use "state" variables instead.
1209 !=~ should be !~
1211 A new warning, "!=~ should be !~", is emitted to prevent this
1212 misspelling of the non-matching operator.
1213 Newline in left-justified string
1215 The warning Newline in left-justified string has been removed.
1216 Too late for "-T" option
1218 The error Too late for "-T" option has been reformulated to be more
1219 descriptive.
1220 "%s" variable %s masks earlier declaration
1222 This warning is now emitted in more consistent cases; in short,
1223 when one of the declarations involved is a "my" variable:
1224 my $x; my $x; # warns
1226 my $x; our $x; # warns
1227 our $x; my $x; # warns
1228 On the other hand, the following:
1230 our $x; our $x;
1232 now gives a ""our" variable %s redeclared" warning.
1234 readdir()/closedir()/etc. attempted on invalid dirhandle
1236 These new warnings are now emitted when a dirhandle is used but is
1237 either closed or not really a dirhandle.
1238 Opening dirhandle/filehandle %s also as a file/directory
1240 Two deprecation warnings have been added: (Rafael)
1241 Opening dirhandle %s also as a file
1243 Opening filehandle %s also as a directory
1244 Use of -P is deprecated
1246 Perl's command-line switch "-P" is now deprecated.
1247 v-string in use/require is non-portable
1249 Perl will warn you against potential backwards compatibility
1250 problems with the "use VERSION" syntax.
1251 perl -V
1253 "perl -V" has several improvements, making it more useable from
1254 shell scripts to get the value of configuration variables. See
1255 perlrun for details.
1256
1258 In general, the source code of perl has been refactored, tidied up, and
1259 optimized in many places. Also, memory management and allocation has
1260 been improved in several points.
1261 When compiling the perl core with gcc, as many gcc warning flags are
1263 turned on as is possible on the platform. (This quest for cleanliness
1264 doesn't extend to XS code because we cannot guarantee the tidiness of
1265 code we didn't write.) Similar strictness flags have been added or
1266 tightened for various other C compilers.
1267 Reordering of SVt_* constants
1269 The relative ordering of constants that define the various types of
1270 "SV" have changed; in particular, "SVt_PVGV" has been moved before
1271 "SVt_PVLV", "SVt_PVAV", "SVt_PVHV" and "SVt_PVCV". This is unlikely to
1272 make any difference unless you have code that explicitly makes
1273 assumptions about that ordering. (The inheritance hierarchy of "B::*"
1274 objects has been changed to reflect this.)
1275 Elimination of SVt_PVBM
1277 Related to this, the internal type "SVt_PVBM" has been removed. This
1278 dedicated type of "SV" was used by the "index" operator and parts of
1279 the regexp engine to facilitate fast Boyer-Moore matches. Its use
1280 internally has been replaced by "SV"s of type "SVt_PVGV".
1281 New type SVt_BIND
1283 A new type "SVt_BIND" has been added, in readiness for the project to
1284 implement Perl 6 on 5. There deliberately is no implementation yet, and
1285 they cannot yet be created or destroyed.
1286 Removal of CPP symbols
1288 The C preprocessor symbols "PERL_PM_APIVERSION" and
1289 "PERL_XS_APIVERSION", which were supposed to give the version number of
1290 the oldest perl binary-compatible (resp. source-compatible) with the
1291 present one, were not used, and sometimes had misleading values. They
1292 have been removed.
1293 Less space is used by ops
1295 The "BASEOP" structure now uses less space. The "op_seq" field has been
1296 removed and replaced by a single bit bit-field "op_opt". "op_type" is
1297 now 9 bits long. (Consequently, the "B::OP" class doesn't provide an
1298 "seq" method anymore.)
1299 New parser
1301 perl's parser is now generated by bison (it used to be generated by
1302 byacc.) As a result, it seems to be a bit more robust.
1303 Also, Dave Mitchell improved the lexer debugging output under "-DT".
1305 Use of "const"
1307 Andy Lester supplied many improvements to determine which function
1308 parameters and local variables could actually be declared "const" to
1309 the C compiler. Steve Peters provided new *_set macros and reworked the
1310 core to use these rather than assigning to macros in LVALUE context.
1311 Mathoms
1313 A new file, mathoms.c, has been added. It contains functions that are
1314 no longer used in the perl core, but that remain available for binary
1315 or source compatibility reasons. However, those functions will not be
1316 compiled in if you add "-DNO_MATHOMS" in the compiler flags.
1317 "AvFLAGS" has been removed
1319 The "AvFLAGS" macro has been removed.
1320 "av_*" changes
1322 The "av_*()" functions, used to manipulate arrays, no longer accept
1323 null "AV*" parameters.
1324 $^H and %^H
1326 The implementation of the special variables $^H and %^H has changed, to
1327 allow implementing lexical pragmas in pure Perl.
1328 B:: modules inheritance changed
1330 The inheritance hierarchy of "B::" modules has changed; "B::NV" now
1331 inherits from "B::SV" (it used to inherit from "B::IV").
1332 Anonymous hash and array constructors
1334 The anonymous hash and array constructors now take 1 op in the optree
1335 instead of 3, now that pp_anonhash and pp_anonlist return a reference
1336 to a hash/array when the op is flagged with OPf_SPECIAL. (Nicholas
1337 Clark)
1338
1340 There's still a remaining problem in the implementation of the lexical
1341 $_: it doesn't work inside "/(?{...})/" blocks. (See the TODO test in
1342 t/op/mydef.t.)
1343 Stacked filetest operators won't work when the "filetest" pragma is in
1345 effect, because they rely on the stat() buffer "_" being populated, and
1346 filetest bypasses stat().
1347 UTF-8 problems
1349 The handling of Unicode still is unclean in several places, where it's
1350 dependent on whether a string is internally flagged as UTF-8. This will
1351 be made more consistent in perl 5.12, but that won't be possible
1352 without a certain amount of backwards incompatibility.
1353
1355 When compiled with g++ and thread support on Linux, it's reported that
1356 the $! stops working correctly. This is related to the fact that the
1357 glibc provides two strerror_r(3) implementation, and perl selects the
1358 wrong one.
1359
1361 If you find what you think is a bug, you might check the articles
1362 recently posted to the comp.lang.perl.misc newsgroup and the perl bug
1363 database at http://rt.perl.org/rt3/ . There may also be information at
1364 http://www.perl.org/ , the Perl Home Page.
1365 If you believe you have an unreported bug, please run the perlbug
1367 program included with your release. Be sure to trim your bug down to a
1368 tiny but sufficient test case. Your bug report, along with the output
1369 of "perl -V", will be sent off to perlbug@perl.org to be analysed by
1370 the Perl porting team.
1371
1373 The Changes file and the perl590delta to perl595delta man pages for
1374 exhaustive details on what changed.
1375 The INSTALL file for how to build Perl.
1377 The README file for general stuff.
1379 The Artistic and Copying files for copyright information.
1381perl v5.36.3 2023-11-30 PERL5100DELTA(1)