Commit | Line | Data |
f6eae373 |
1 | =head1 NAME |
2 | |
3 | perldelta - what is new for perl v5.9.5 |
4 | |
5 | =head1 DESCRIPTION |
6 | |
7 | This document describes differences between the 5.9.4 and the 5.9.5 |
8 | development releases. See L<perl590delta>, L<perl591delta>, |
9 | L<perl592delta>, L<perl593delta> and L<perl594delta> for the differences |
10 | between 5.8.0 and 5.9.4. |
11 | |
12 | =head1 Incompatible Changes |
13 | |
20ee07fb |
14 | =head2 Tainting and printf |
15 | |
16 | When perl is run under taint mode, C<printf()> and C<sprintf()> will now |
3f10c77a |
17 | reject any tainted format argument. (Rafael Garcia-Suarez) |
20ee07fb |
18 | |
54a37cc6 |
19 | =head2 undef and signal handlers |
20 | |
21 | Undefining or deleting a signal handler via C<undef $SIG{FOO}> is now |
97f820fb |
22 | equivalent to setting it to C<'DEFAULT'>. (Rafael) |
23 | |
24 | =head2 strictures and array/hash dereferencing in defined() |
25 | |
26 | C<defined @$foo> and C<defined %$bar> are now subject to C<strict 'refs'> |
27 | (that is, C<$foo> and C<$bar> shall be proper references there.) |
28 | (Nicholas Clark) |
29 | |
30 | (However, C<defined(@foo)> and C<defined(%bar)> are discouraged constructs |
31 | anyway.) |
54a37cc6 |
32 | |
74bb26f2 |
33 | =head2 C<(?p{})> has been removed |
34 | |
35 | The regular expression construct C<(?p{})>, which was deprecated in perl |
36 | 5.8, has been removed. Use C<(??{})> instead. (Rafael) |
37 | |
00880d60 |
38 | =head2 Pseudo-hashes have been removed |
39 | |
40 | Support for pseudo-hashes has been removed from Perl 5.9. (The C<fields> |
41 | pragma remains here, but uses an alternate implementation.) |
42 | |
73966613 |
43 | =head2 Removal of the bytecode compiler and of perlcc |
44 | |
45 | C<perlcc>, the byteloader and the supporting modules (B::C, B::CC, |
46 | B::Bytecode, etc.) are no longer distributed with the perl sources. Those |
47 | experimental tools have never worked reliably, and, due to the lack of |
48 | volunteers to keep them in line with the perl interpreter developments, it |
49 | was decided to remove them instead of shipping a broken version of those. |
50 | The last version of those modules can be found with perl 5.9.4. |
51 | |
52 | However the B compiler framework stays supported in the perl core, as with |
53 | the more useful modules it has permitted (among others, B::Deparse and |
54 | B::Concise). |
55 | |
56 | =head2 Removal of the JPL |
57 | |
58 | The JPL (Java-Perl Linguo) has been removed from the perl sources tarball. |
59 | |
afa2ea4a |
60 | =head2 Recursive inheritance detected earlier |
61 | |
62 | Perl will now immediately throw an exception if you modify any package's |
63 | C<@ISA> in such a way that it would cause recursive inheritance. |
64 | |
65 | Previously, the exception would not occur until Perl attempted to make |
66 | use of the recursive inheritance while resolving a method or doing a |
67 | C<$foo-E<gt>isa($bar)> lookup. |
68 | |
f6eae373 |
69 | =head1 Core Enhancements |
70 | |
072f65b4 |
71 | =head2 Regular expressions |
72 | |
73 | =over 4 |
74 | |
75 | =item Recursive Patterns |
76 | |
77 | It is now possible to write recursive patterns without using the C<(??{})> |
78 | construct. This new way is more efficient, and in many cases easier to |
79 | read. |
80 | |
81 | Each capturing parenthesis can now be treated as an independent pattern |
82 | that can be entered by using the C<(?PARNO)> syntax (C<PARNO> standing for |
83 | "parenthesis number"). For example, the following pattern will match |
84 | nested balanced angle brackets: |
85 | |
86 | / |
87 | ^ # start of line |
88 | ( # start capture buffer 1 |
89 | < # match an opening angle bracket |
90 | (?: # match one of: |
91 | (?> # don't backtrack over the inside of this group |
92 | [^<>]+ # one or more non angle brackets |
93 | ) # end non backtracking group |
94 | | # ... or ... |
95 | (?1) # recurse to bracket 1 and try it again |
96 | )* # 0 or more times. |
97 | > # match a closing angle bracket |
98 | ) # end capture buffer one |
99 | $ # end of line |
100 | /x |
101 | |
102 | Note, users experienced with PCRE will find that the Perl implementation |
103 | of this feature differs from the PCRE one in that it is possible to |
104 | backtrack into a recursed pattern, whereas in PCRE the recursion is |
73966613 |
105 | atomic or "possessive" in nature. (Yves Orton) |
072f65b4 |
106 | |
107 | =item Named Capture Buffers |
108 | |
109 | It is now possible to name capturing parenthesis in a pattern and refer to |
110 | the captured contents by name. The naming syntax is C<< (?<NAME>....) >>. |
111 | It's possible to backreference to a named buffer with the C<< \k<NAME> >> |
97f820fb |
112 | syntax. In code, the new magical hashes C<%+> and C<%-> can be used to |
113 | access the contents of the capture buffers. |
072f65b4 |
114 | |
115 | Thus, to replace all doubled chars, one could write |
116 | |
117 | s/(?<letter>.)\k<letter>/$+{letter}/g |
118 | |
97f820fb |
119 | Only buffers with defined contents will be "visible" in the C<%+> hash, so |
072f65b4 |
120 | it's possible to do something like |
121 | |
122 | foreach my $name (keys %+) { |
123 | print "content of buffer '$name' is $+{$name}\n"; |
124 | } |
125 | |
97f820fb |
126 | The C<%-> hash is a bit more complete, since it will contain array refs |
127 | holding values from all capture buffers similarly named, if there should |
128 | be many of them. |
129 | |
130 | C<%+> and C<%-> are implemented as tied hashes through the new module |
80305961 |
131 | C<Tie::Hash::NamedCapture>. |
97f820fb |
132 | |
072f65b4 |
133 | Users exposed to the .NET regex engine will find that the perl |
134 | implementation differs in that the numerical ordering of the buffers |
135 | is sequential, and not "unnamed first, then named". Thus in the pattern |
136 | |
137 | /(A)(?<B>B)(C)(?<D>D)/ |
138 | |
139 | $1 will be 'A', $2 will be 'B', $3 will be 'C' and $4 will be 'D' and not |
140 | $1 is 'A', $2 is 'C' and $3 is 'B' and $4 is 'D' that a .NET programmer |
73966613 |
141 | would expect. This is considered a feature. :-) (Yves Orton) |
072f65b4 |
142 | |
b9b4dddf |
143 | =item Possessive Quantifiers |
144 | |
ee9b8eae |
145 | Perl now supports the "possessive quantifier" syntax of the "atomic match" |
b9b4dddf |
146 | pattern. Basically a possessive quantifier matches as much as it can and never |
ee9b8eae |
147 | gives any back. Thus it can be used to control backtracking. The syntax is |
b9b4dddf |
148 | similar to non-greedy matching, except instead of using a '?' as the modifier |
149 | the '+' is used. Thus C<?+>, C<*+>, C<++>, C<{min,max}+> are now legal |
73966613 |
150 | quantifiers. (Yves Orton) |
b9b4dddf |
151 | |
24b23f37 |
152 | =item Backtracking control verbs |
153 | |
3f10c77a |
154 | The regex engine now supports a number of special-purpose backtrack |
5d458dd8 |
155 | control verbs: (*THEN), (*PRUNE), (*MARK), (*SKIP), (*COMMIT), (*FAIL) |
c74340f9 |
156 | and (*ACCEPT). See L<perlre> for their descriptions. (Yves Orton) |
157 | |
158 | =item Relative backreferences |
159 | |
2bf803e2 |
160 | A new syntax C<\g{N}> or C<\gN> where "N" is a decimal integer allows a |
161 | safer form of back-reference notation as well as allowing relative |
162 | backreferences. This should make it easier to generate and embed patterns |
3f10c77a |
163 | that contain backreferences. See L<perlre/"Capture buffers">. (Yves Orton) |
24b23f37 |
164 | |
97f820fb |
165 | =item C<\K> escape |
ee9b8eae |
166 | |
167 | The functionality of Jeff Pinyan's module Regexp::Keep has been added to |
168 | the core. You can now use in regular expressions the special escape C<\K> |
169 | as a way to do something like floating length positive lookbehind. It is |
170 | also useful in substitutions like: |
171 | |
172 | s/(foo)bar/$1/g |
173 | |
174 | that can now be converted to |
175 | |
176 | s/foo\Kbar//g |
177 | |
97f820fb |
178 | which is much more efficient. (Yves Orton) |
ee9b8eae |
179 | |
41b9272f |
180 | =item Vertical and horizontal whitespace, and linebreak |
181 | |
182 | Regular expressions now recognize the C<\v> and C<\h> escapes, that match |
183 | vertical and horizontal whitespace, respectively. C<\V> and C<\H> |
184 | logically match their complements. |
185 | |
329d35d1 |
186 | C<\R> matches a generic linebreak, that is, vertical whitespace, plus |
41b9272f |
187 | the multi-character sequence C<"\x0D\x0A">. |
188 | |
75c442e4 |
189 | =back |
190 | |
d5494b07 |
191 | =head2 The C<_> prototype |
192 | |
193 | A new prototype character has been added. C<_> is equivalent to C<$> (it |
194 | denotes a scalar), but defaults to C<$_> if the corresponding argument |
195 | isn't supplied. Due to the optional nature of the argument, you can only |
196 | use it at the end of a prototype, or before a semicolon. |
197 | |
73966613 |
198 | This has a small incompatible consequence: the prototype() function has |
199 | been adjusted to return C<_> for some built-ins in appropriate cases (for |
97f820fb |
200 | example, C<prototype('CORE::rmdir')>). (Rafael) |
73966613 |
201 | |
49f595a6 |
202 | =head2 UNITCHECK blocks |
203 | |
204 | C<UNITCHECK>, a new special code block has been introduced, in addition to |
205 | C<BEGIN>, C<CHECK>, C<INIT> and C<END>. |
206 | |
207 | C<CHECK> and C<INIT> blocks, while useful for some specialized purposes, |
208 | are always executed at the transition between the compilation and the |
209 | execution of the main program, and thus are useless whenever code is |
210 | loaded at runtime. On the other hand, C<UNITCHECK> blocks are executed |
211 | just after the unit which defined them has been compiled. See L<perlmod> |
212 | for more information. (Alex Gough) |
213 | |
5a093634 |
214 | =head2 readpipe() is now overridable |
215 | |
216 | The built-in function readpipe() is now overridable. Overriding it permits |
74bb26f2 |
217 | also to override its operator counterpart, C<qx//> (a.k.a. C<``>). |
218 | Moreover, it now defaults to C<$_> if no argument is provided. (Rafael) |
219 | |
220 | =head2 default argument for readline() |
221 | |
222 | readline() now defaults to C<*ARGV> if no argument is provided. (Rafael) |
5a093634 |
223 | |
73966613 |
224 | =head2 UCD 5.0.0 |
225 | |
226 | The copy of the Unicode Character Database included in Perl 5.9 has |
227 | been updated to version 5.0.0. |
228 | |
97f820fb |
229 | =head2 Smart match |
230 | |
231 | The smart match operator (C<~~>) is now available by default (you don't |
232 | need to enable it with C<use feature> any longer). (Michael G Schwern) |
233 | |
74bb26f2 |
234 | =head2 Implicit loading of C<feature> |
235 | |
236 | The C<feature> pragma is now implicitly loaded when you require a minimal |
237 | perl version (with the C<use VERSION> construct) greater than, or equal |
238 | to, 5.9.5. |
239 | |
f6eae373 |
240 | =head1 Modules and Pragmas |
241 | |
74bb26f2 |
242 | =head2 New Pragma, C<mro> |
243 | |
244 | A new pragma, C<mro> (for Method Resolution Order) has been added. It |
245 | permits to switch, on a per-class basis, the algorithm that perl uses to |
246 | find inherited methods in case of a mutiple inheritance hierachy. The |
247 | default MRO hasn't changed (DFS, for Depth First Search). Another MRO is |
248 | available: the C3 algorithm. See L<mro> for more information. |
249 | (Brandon Black) |
250 | |
91ddf7c8 |
251 | Note that, due to changes in the implentation of class hierarchy search, |
252 | code that used to undef the C<*ISA> glob will most probably break. Anyway, |
253 | undef'ing C<*ISA> had the side-effect of removing the magic on the @ISA |
254 | array and should not have been done in the first place. |
255 | |
3284ac36 |
256 | =head2 bignum, bigint, bigrat |
257 | |
258 | The three numeric pragmas C<bignum>, C<bigint> and C<bigrat> are now |
259 | lexically scoped. (Tels) |
260 | |
f6eae373 |
261 | =head2 New Core Modules |
262 | |
73966613 |
263 | =over 4 |
264 | |
265 | =item * |
266 | |
267 | C<Locale::Maketext::Simple>, needed by CPANPLUS, is a simple wrapper around |
268 | C<Locale::Maketext::Lexicon>. Note that C<Locale::Maketext::Lexicon> isn't |
269 | included in the perl core; the behaviour of C<Locale::Maketext::Simple> |
270 | gracefully degrades when the later isn't present. |
271 | |
272 | =item * |
273 | |
274 | C<Params::Check> implements a generic input parsing/checking mechanism. It |
275 | is used by CPANPLUS. |
276 | |
5a093634 |
277 | =item * |
278 | |
279 | C<Term::UI> simplifies the task to ask questions at a terminal prompt. |
280 | |
281 | =item * |
282 | |
283 | C<Object::Accessor> provides an interface to create per-object accessors. |
284 | |
97f820fb |
285 | =item * |
286 | |
287 | C<Module::Pluggable> is a simple framework to create modules that accept |
288 | pluggable sub-modules. |
289 | |
290 | =item * |
291 | |
292 | C<Module::Load::Conditional> provides simple ways to query and possibly |
293 | load installed modules. |
294 | |
295 | =item * |
296 | |
297 | C<Time::Piece> provides an object oriented interface to time functions, |
298 | overriding the built-ins localtime() and gmtime(). |
299 | |
300 | =item * |
301 | |
302 | C<IPC::Cmd> helps to find and run external commands, possibly |
303 | interactively. |
304 | |
305 | =item * |
306 | |
307 | C<File::Fetch> provide a simple generic file fetching mechanism. |
308 | |
309 | =item * |
310 | |
201a0ee1 |
311 | C<Log::Message> and C<Log::Message::Simple> are used by the log facility |
312 | of C<CPANPLUS>. |
313 | |
314 | =item * |
315 | |
97f820fb |
316 | C<Archive::Extract> is a generic archive extraction mechanism |
317 | for F<.tar> (plain, gziped or bzipped) or F<.zip> files. |
318 | |
74bb26f2 |
319 | =item * |
320 | |
321 | C<CPANPLUS> provides an API and a command-line tool to access the CPAN |
322 | mirrors. |
323 | |
73966613 |
324 | =back |
325 | |
d5494b07 |
326 | =head2 Module changes |
327 | |
328 | =over 4 |
329 | |
ddf4dafe |
330 | =item C<assertions> |
331 | |
332 | The C<assertions> pragma, its submodules C<assertions::activate> and |
333 | C<assertions::compat> and the B<-A> command-line switch have been removed. |
334 | The interface was not judged mature enough for inclusion in a stable |
335 | release. |
336 | |
d5494b07 |
337 | =item C<base> |
338 | |
339 | The C<base> pragma now warns if a class tries to inherit from itself. |
97f820fb |
340 | (Curtis "Ovid" Poe) |
d5494b07 |
341 | |
74bb26f2 |
342 | =item C<strict> and C<warnings> |
343 | |
344 | C<strict> and C<warnings> will now complain loudly if they are loaded via |
345 | incorrect casing (as in C<use Strict;>). (Johan Vromans) |
346 | |
18857c0b |
347 | =item C<warnings> |
348 | |
349 | The C<warnings> pragma doesn't load C<Carp> anymore. That means that code |
350 | that used C<Carp> routines without having loaded it at compile time might |
351 | need to be adjusted; typically, the following (faulty) code won't work |
352 | anymore, and will require parentheses to be added after the function name: |
353 | |
354 | use warnings; |
355 | require Carp; |
356 | Carp::confess "argh"; |
357 | |
97f820fb |
358 | =item C<less> |
359 | |
360 | C<less> now does something useful (or at least it tries to). In fact, it |
361 | has been turned into a lexical pragma. So, in your modules, you can now |
362 | test whether your users have requested to use less CPU, or less memory, |
363 | less magic, or maybe even less fat. See L<less> for more. (Joshua ben |
364 | Jore) |
365 | |
3f10c77a |
366 | =item C<Attribute::Handlers> |
367 | |
368 | C<Attribute::Handlers> can now report the caller's file and line number. |
369 | (David Feldman) |
370 | |
97f820fb |
371 | =item C<B::Lint> |
372 | |
373 | C<B::Lint> is now based on C<Module::Pluggable>, and so can be extended |
374 | with plugins. (Joshua ben Jore) |
375 | |
376 | =item C<B> |
377 | |
378 | It's now possible to access the lexical pragma hints (C<%^H>) by using the |
379 | method B::COP::hints_hash(). It returns a C<B::RHE> object, which in turn |
380 | can be used to get a hash reference via the method B::RHE::HASH(). (Joshua |
381 | ben Jore) |
382 | |
383 | =for p5p XXX document this in B.pm too |
384 | |
ab4e6221 |
385 | =item C<Thread> |
386 | |
387 | As the old 5005thread threading model has been removed, in favor of the |
388 | ithreads scheme, the C<Thread> module is now a compatibility wrapper, to |
389 | be used in old code only. |
390 | |
d5494b07 |
391 | =back |
392 | |
f6eae373 |
393 | =head1 Utility Changes |
394 | |
74bb26f2 |
395 | =head2 C<cpanp> |
396 | |
397 | C<cpanp>, the CPANPLUS shell, has been added. (C<cpanp-run-perl>, an |
398 | helper for CPANPLUS operation, has been added too, but isn't intended for |
399 | direct use). |
400 | |
8a499140 |
401 | =head2 C<cpan2dist> |
402 | |
403 | C<cpan2dist> is a new utility, that comes with CPANPLUS. It's a tool to |
404 | create distributions (or packages) from CPAN modules. |
405 | |
74bb26f2 |
406 | =head2 C<pod2html> |
407 | |
408 | The output of C<pod2html> has been enhanced to be more customizable via |
409 | CSS. Some formatting problems were also corrected. (Jari Aalto) |
410 | |
f6eae373 |
411 | =head1 Documentation |
412 | |
74bb26f2 |
413 | =head2 New manpage, perlunifaq |
414 | |
415 | A new manual page, L<perlunifaq> (the Perl Unicode FAQ), has been added |
416 | (Juerd Waalboer). |
417 | |
f6eae373 |
418 | =head1 Performance Enhancements |
419 | |
420 | =head1 Installation and Configuration Improvements |
421 | |
73966613 |
422 | =head2 C++ compatibility |
423 | |
424 | Efforts have been made to make perl and the core XS modules compilable |
425 | with various C++ compilers (although the situation is not perfect with |
426 | some of the compilers on some of the platforms tested.) |
427 | |
ab4e6221 |
428 | =head2 Visual C++ |
429 | |
430 | Perl now can be compiled with Microsoft Visual C++ 2005. |
431 | |
3f10c77a |
432 | =head2 Static build on Win32 |
433 | |
434 | It's now possible to build a C<perl-static.exe> that doesn't depend |
435 | on C<perl59.dll> on Win32. See the Win32 makefiles for details. |
e3c82801 |
436 | (Vadim Konovalov) |
3f10c77a |
437 | |
ab4e6221 |
438 | =head2 C<d_pseudofork> |
439 | |
440 | A new configuration variable, available as C<$Config{d_pseudofork}> in |
441 | the L<Config> module, has been added, to distinguish real fork() support |
442 | from fake pseudofork used on Windows platforms. |
443 | |
73966613 |
444 | =head2 Ports |
445 | |
446 | Perl has been reported to work on MidnightBSD. |
447 | |
f6eae373 |
448 | =head1 Selected Bug Fixes |
449 | |
49f595a6 |
450 | PerlIO::scalar will now prevent writing to read-only scalars. Moreover, |
451 | seek() is now supported with PerlIO::scalar-based filehandles, the |
97f820fb |
452 | underlying string being zero-filled as needed. (Rafael, Jarkko Hietaniemi) |
73966613 |
453 | |
454 | study() never worked for UTF-8 strings, but could lead to false results. |
455 | It's now a no-op on UTF-8 data. (Yves Orton) |
456 | |
49f595a6 |
457 | The signals SIGILL, SIGBUS and SIGSEGV are now always delivered in an |
458 | "unsafe" manner (contrary to other signals, that are deferred until the |
459 | perl interpreter reaches a reasonably stable state; see |
97f820fb |
460 | L<perlipc/"Deferred Signals (Safe Signals)">). (Rafael) |
49f595a6 |
461 | |
5a093634 |
462 | When a module or a file is loaded through an @INC-hook, and when this hook |
463 | has set a filename entry in %INC, __FILE__ is now set for this module |
97f820fb |
464 | accordingly to the contents of that %INC entry. (Rafael) |
465 | |
466 | The C<-w> and C<-t> switches can now be used together without messing |
467 | up what categories of warnings are activated or not. (Rafael) |
5a093634 |
468 | |
74bb26f2 |
469 | Duping a filehandle which has the C<:utf8> PerlIO layer set will now |
470 | properly carry that layer on the duped filehandle. (Rafael) |
471 | |
37a7450d |
472 | Localizing an hash element whose key was given as a variable didn't work |
21e0a455 |
473 | correctly if the variable was changed while the local() was in effect (as |
474 | in C<local $h{$x}; ++$x>). (Bo Lindbergh) |
37a7450d |
475 | |
f6eae373 |
476 | =head1 New or Changed Diagnostics |
477 | |
74bb26f2 |
478 | =head2 Deprecations |
479 | |
480 | Two deprecation warnings have been added: (Rafael) |
481 | |
482 | Opening dirhandle %s also as a file |
483 | Opening filehandle %s also as a directory |
484 | |
f6eae373 |
485 | =head1 Changed Internals |
486 | |
73966613 |
487 | The anonymous hash and array constructors now take 1 op in the optree |
488 | instead of 3, now that pp_anonhash and pp_anonlist return a reference to |
489 | an hash/array when the op is flagged with OPf_SPECIAL (Nicholas Clark). |
490 | |
97f820fb |
491 | =for p5p XXX have we some docs on how to create regexp engine plugins, since that's now possible ? (perlreguts) |
492 | |
493 | =for p5p XXX new BIND SV type, #29544, #29642 |
494 | |
f6eae373 |
495 | =head1 Known Problems |
496 | |
497 | =head2 Platform Specific Problems |
498 | |
499 | =head1 Reporting Bugs |
500 | |
501 | If you find what you think is a bug, you might check the articles |
502 | recently posted to the comp.lang.perl.misc newsgroup and the perl |
503 | bug database at http://rt.perl.org/rt3/ . There may also be |
504 | information at http://www.perl.org/ , the Perl Home Page. |
505 | |
506 | If you believe you have an unreported bug, please run the B<perlbug> |
507 | program included with your release. Be sure to trim your bug down |
508 | to a tiny but sufficient test case. Your bug report, along with the |
509 | output of C<perl -V>, will be sent off to perlbug@perl.org to be |
510 | analysed by the Perl porting team. |
511 | |
512 | =head1 SEE ALSO |
513 | |
514 | The F<Changes> file for exhaustive details on what changed. |
515 | |
516 | The F<INSTALL> file for how to build Perl. |
517 | |
518 | The F<README> file for general stuff. |
519 | |
520 | The F<Artistic> and F<Copying> files for copyright information. |
521 | |
522 | =cut |