| 1 | =encoding utf8 |
| 2 | |
| 3 | =head1 NAME |
| 4 | |
| 5 | perl5100delta - what is new for perl 5.10.0 |
| 6 | |
| 7 | =head1 DESCRIPTION |
| 8 | |
| 9 | This document describes the differences between the 5.8.8 release and |
| 10 | the 5.10.0 release. |
| 11 | |
| 12 | Many of the bug fixes in 5.10.0 were already seen in the 5.8.X maintenance |
| 13 | releases; they are not duplicated here and are documented in the set of |
| 14 | man pages named perl58[1-8]?delta. |
| 15 | |
| 16 | =head1 Core Enhancements |
| 17 | |
| 18 | =head2 The C<feature> pragma |
| 19 | |
| 20 | The C<feature> pragma is used to enable new syntax that would break Perl's |
| 21 | backwards-compatibility with older releases of the language. It's a lexical |
| 22 | pragma, like C<strict> or C<warnings>. |
| 23 | |
| 24 | Currently the following new features are available: C<switch> (adds a |
| 25 | switch statement), C<say> (adds a C<say> built-in function), and C<state> |
| 26 | (adds a C<state> keyword for declaring "static" variables). Those |
| 27 | features are described in their own sections of this document. |
| 28 | |
| 29 | The C<feature> pragma is also implicitly loaded when you require a minimal |
| 30 | perl version (with the C<use VERSION> construct) greater than, or equal |
| 31 | to, 5.9.5. See L<feature> for details. |
| 32 | |
| 33 | =head2 New B<-E> command-line switch |
| 34 | |
| 35 | B<-E> is equivalent to B<-e>, but it implicitly enables all |
| 36 | optional features (like C<use feature ":5.10">). |
| 37 | |
| 38 | =head2 Defined-or operator |
| 39 | |
| 40 | A new operator C<//> (defined-or) has been implemented. |
| 41 | The following expression: |
| 42 | |
| 43 | $a // $b |
| 44 | |
| 45 | is merely equivalent to |
| 46 | |
| 47 | defined $a ? $a : $b |
| 48 | |
| 49 | and the statement |
| 50 | |
| 51 | $c //= $d; |
| 52 | |
| 53 | can now be used instead of |
| 54 | |
| 55 | $c = $d unless defined $c; |
| 56 | |
| 57 | The C<//> operator has the same precedence and associativity as C<||>. |
| 58 | Special care has been taken to ensure that this operator Do What You Mean |
| 59 | while not breaking old code, but some edge cases involving the empty |
| 60 | regular expression may now parse differently. See L<perlop> for |
| 61 | details. |
| 62 | |
| 63 | =head2 Switch and Smart Match operator |
| 64 | |
| 65 | Perl 5 now has a switch statement. It's available when C<use feature |
| 66 | 'switch'> is in effect. This feature introduces three new keywords, |
| 67 | C<given>, C<when>, and C<default>: |
| 68 | |
| 69 | given ($foo) { |
| 70 | when (/^abc/) { $abc = 1; } |
| 71 | when (/^def/) { $def = 1; } |
| 72 | when (/^xyz/) { $xyz = 1; } |
| 73 | default { $nothing = 1; } |
| 74 | } |
| 75 | |
| 76 | A more complete description of how Perl matches the switch variable |
| 77 | against the C<when> conditions is given in L<perlsyn/"Switch statements">. |
| 78 | |
| 79 | This kind of match is called I<smart match>, and it's also possible to use |
| 80 | it outside of switch statements, via the new C<~~> operator. See |
| 81 | L<perlsyn/"Smart matching in detail">. |
| 82 | |
| 83 | This feature was contributed by Robin Houston. |
| 84 | |
| 85 | =head2 Regular expressions |
| 86 | |
| 87 | =over 4 |
| 88 | |
| 89 | =item Recursive Patterns |
| 90 | |
| 91 | It is now possible to write recursive patterns without using the C<(??{})> |
| 92 | construct. This new way is more efficient, and in many cases easier to |
| 93 | read. |
| 94 | |
| 95 | Each capturing parenthesis can now be treated as an independent pattern |
| 96 | that can be entered by using the C<(?PARNO)> syntax (C<PARNO> standing for |
| 97 | "parenthesis number"). For example, the following pattern will match |
| 98 | nested balanced angle brackets: |
| 99 | |
| 100 | / |
| 101 | ^ # start of line |
| 102 | ( # start capture buffer 1 |
| 103 | < # match an opening angle bracket |
| 104 | (?: # match one of: |
| 105 | (?> # don't backtrack over the inside of this group |
| 106 | [^<>]+ # one or more non angle brackets |
| 107 | ) # end non backtracking group |
| 108 | | # ... or ... |
| 109 | (?1) # recurse to bracket 1 and try it again |
| 110 | )* # 0 or more times. |
| 111 | > # match a closing angle bracket |
| 112 | ) # end capture buffer one |
| 113 | $ # end of line |
| 114 | /x |
| 115 | |
| 116 | PCRE users should note that Perl's recursive regex feature allows |
| 117 | backtracking into a recursed pattern, whereas in PCRE the recursion is |
| 118 | atomic or "possessive" in nature. As in the example above, you can |
| 119 | add (?>) to control this selectively. (Yves Orton) |
| 120 | |
| 121 | =item Named Capture Buffers |
| 122 | |
| 123 | It is now possible to name capturing parenthesis in a pattern and refer to |
| 124 | the captured contents by name. The naming syntax is C<< (?<NAME>....) >>. |
| 125 | It's possible to backreference to a named buffer with the C<< \k<NAME> >> |
| 126 | syntax. In code, the new magical hashes C<%+> and C<%-> can be used to |
| 127 | access the contents of the capture buffers. |
| 128 | |
| 129 | Thus, to replace all doubled chars with a single copy, one could write |
| 130 | |
| 131 | s/(?<letter>.)\k<letter>/$+{letter}/g |
| 132 | |
| 133 | Only buffers with defined contents will be "visible" in the C<%+> hash, so |
| 134 | it's possible to do something like |
| 135 | |
| 136 | foreach my $name (keys %+) { |
| 137 | print "content of buffer '$name' is $+{$name}\n"; |
| 138 | } |
| 139 | |
| 140 | The C<%-> hash is a bit more complete, since it will contain array refs |
| 141 | holding values from all capture buffers similarly named, if there should |
| 142 | be many of them. |
| 143 | |
| 144 | C<%+> and C<%-> are implemented as tied hashes through the new module |
| 145 | C<Tie::Hash::NamedCapture>. |
| 146 | |
| 147 | Users exposed to the .NET regex engine will find that the perl |
| 148 | implementation differs in that the numerical ordering of the buffers |
| 149 | is sequential, and not "unnamed first, then named". Thus in the pattern |
| 150 | |
| 151 | /(A)(?<B>B)(C)(?<D>D)/ |
| 152 | |
| 153 | $1 will be 'A', $2 will be 'B', $3 will be 'C' and $4 will be 'D' and not |
| 154 | $1 is 'A', $2 is 'C' and $3 is 'B' and $4 is 'D' that a .NET programmer |
| 155 | would expect. This is considered a feature. :-) (Yves Orton) |
| 156 | |
| 157 | =item Possessive Quantifiers |
| 158 | |
| 159 | Perl now supports the "possessive quantifier" syntax of the "atomic match" |
| 160 | pattern. Basically a possessive quantifier matches as much as it can and never |
| 161 | gives any back. Thus it can be used to control backtracking. The syntax is |
| 162 | similar to non-greedy matching, except instead of using a '?' as the modifier |
| 163 | the '+' is used. Thus C<?+>, C<*+>, C<++>, C<{min,max}+> are now legal |
| 164 | quantifiers. (Yves Orton) |
| 165 | |
| 166 | =item Backtracking control verbs |
| 167 | |
| 168 | The regex engine now supports a number of special-purpose backtrack |
| 169 | control verbs: (*THEN), (*PRUNE), (*MARK), (*SKIP), (*COMMIT), (*FAIL) |
| 170 | and (*ACCEPT). See L<perlre> for their descriptions. (Yves Orton) |
| 171 | |
| 172 | =item Relative backreferences |
| 173 | |
| 174 | A new syntax C<\g{N}> or C<\gN> where "N" is a decimal integer allows a |
| 175 | safer form of back-reference notation as well as allowing relative |
| 176 | backreferences. This should make it easier to generate and embed patterns |
| 177 | that contain backreferences. See L<perlre/"Capture buffers">. (Yves Orton) |
| 178 | |
| 179 | =item C<\K> escape |
| 180 | |
| 181 | The functionality of Jeff Pinyan's module Regexp::Keep has been added to |
| 182 | the core. In regular expressions you can now use the special escape C<\K> |
| 183 | as a way to do something like floating length positive lookbehind. It is |
| 184 | also useful in substitutions like: |
| 185 | |
| 186 | s/(foo)bar/$1/g |
| 187 | |
| 188 | that can now be converted to |
| 189 | |
| 190 | s/foo\Kbar//g |
| 191 | |
| 192 | which is much more efficient. (Yves Orton) |
| 193 | |
| 194 | =item Vertical and horizontal whitespace, and linebreak |
| 195 | |
| 196 | Regular expressions now recognize the C<\v> and C<\h> escapes that match |
| 197 | vertical and horizontal whitespace, respectively. C<\V> and C<\H> |
| 198 | logically match their complements. |
| 199 | |
| 200 | C<\R> matches a generic linebreak, that is, vertical whitespace, plus |
| 201 | the multi-character sequence C<"\x0D\x0A">. |
| 202 | |
| 203 | =item Optional pre-match and post-match captures with the /p flag |
| 204 | |
| 205 | There is a new flag C</p> for regular expressions. Using this |
| 206 | makes the engine preserve a copy of the part of the matched string before |
| 207 | the matching substring to the new special variable C<${^PREMATCH}>, the |
| 208 | part after the matching substring to C<${^POSTMATCH}>, and the matched |
| 209 | substring itself to C<${^MATCH}>. |
| 210 | |
| 211 | Perl is still able to store these substrings to the special variables |
| 212 | C<$`>, C<$'>, C<$&>, but using these variables anywhere in the program |
| 213 | adds a penalty to all regular expression matches, whereas if you use |
| 214 | the C</p> flag and the new special variables instead, you pay only for |
| 215 | the regular expressions where the flag is used. |
| 216 | |
| 217 | For more detail on the new variables, see L<perlvar>; for the use of |
| 218 | the regular expression flag, see L<perlop> and L<perlre>. |
| 219 | |
| 220 | =back |
| 221 | |
| 222 | =head2 C<say()> |
| 223 | |
| 224 | say() is a new built-in, only available when C<use feature 'say'> is in |
| 225 | effect, that is similar to print(), but that implicitly appends a newline |
| 226 | to the printed string. See L<perlfunc/say>. (Robin Houston) |
| 227 | |
| 228 | =head2 Lexical C<$_> |
| 229 | |
| 230 | The default variable C<$_> can now be lexicalized, by declaring it like |
| 231 | any other lexical variable, with a simple |
| 232 | |
| 233 | my $_; |
| 234 | |
| 235 | The operations that default on C<$_> will use the lexically-scoped |
| 236 | version of C<$_> when it exists, instead of the global C<$_>. |
| 237 | |
| 238 | In a C<map> or a C<grep> block, if C<$_> was previously my'ed, then the |
| 239 | C<$_> inside the block is lexical as well (and scoped to the block). |
| 240 | |
| 241 | In a scope where C<$_> has been lexicalized, you can still have access to |
| 242 | the global version of C<$_> by using C<$::_>, or, more simply, by |
| 243 | overriding the lexical declaration with C<our $_>. (Rafael Garcia-Suarez) |
| 244 | |
| 245 | =head2 The C<_> prototype |
| 246 | |
| 247 | A new prototype character has been added. C<_> is equivalent to C<$> but |
| 248 | defaults to C<$_> if the corresponding argument isn't supplied (both C<$> |
| 249 | and C<_> denote a scalar). Due to the optional nature of the argument, |
| 250 | you can only use it at the end of a prototype, or before a semicolon. |
| 251 | |
| 252 | This has a small incompatible consequence: the prototype() function has |
| 253 | been adjusted to return C<_> for some built-ins in appropriate cases (for |
| 254 | example, C<prototype('CORE::rmdir')>). (Rafael Garcia-Suarez) |
| 255 | |
| 256 | =head2 UNITCHECK blocks |
| 257 | |
| 258 | C<UNITCHECK>, a new special code block has been introduced, in addition to |
| 259 | C<BEGIN>, C<CHECK>, C<INIT> and C<END>. |
| 260 | |
| 261 | C<CHECK> and C<INIT> blocks, while useful for some specialized purposes, |
| 262 | are always executed at the transition between the compilation and the |
| 263 | execution of the main program, and thus are useless whenever code is |
| 264 | loaded at runtime. On the other hand, C<UNITCHECK> blocks are executed |
| 265 | just after the unit which defined them has been compiled. See L<perlmod> |
| 266 | for more information. (Alex Gough) |
| 267 | |
| 268 | =head2 New Pragma, C<mro> |
| 269 | |
| 270 | A new pragma, C<mro> (for Method Resolution Order) has been added. It |
| 271 | permits to switch, on a per-class basis, the algorithm that perl uses to |
| 272 | find inherited methods in case of a multiple inheritance hierarchy. The |
| 273 | default MRO hasn't changed (DFS, for Depth First Search). Another MRO is |
| 274 | available: the C3 algorithm. See L<mro> for more information. |
| 275 | (Brandon Black) |
| 276 | |
| 277 | Note that, due to changes in the implementation of class hierarchy search, |
| 278 | code that used to undef the C<*ISA> glob will most probably break. Anyway, |
| 279 | undef'ing C<*ISA> had the side-effect of removing the magic on the @ISA |
| 280 | array and should not have been done in the first place. Also, the |
| 281 | cache C<*::ISA::CACHE::> no longer exists; to force reset the @ISA cache, |
| 282 | you now need to use the C<mro> API, or more simply to assign to @ISA |
| 283 | (e.g. with C<@ISA = @ISA>). |
| 284 | |
| 285 | =head2 readdir() may return a "short filename" on Windows |
| 286 | |
| 287 | The readdir() function may return a "short filename" when the long |
| 288 | filename contains characters outside the ANSI codepage. Similarly |
| 289 | Cwd::cwd() may return a short directory name, and glob() may return short |
| 290 | names as well. On the NTFS file system these short names can always be |
| 291 | represented in the ANSI codepage. This will not be true for all other file |
| 292 | system drivers; e.g. the FAT filesystem stores short filenames in the OEM |
| 293 | codepage, so some files on FAT volumes remain unaccessible through the |
| 294 | ANSI APIs. |
| 295 | |
| 296 | Similarly, $^X, @INC, and $ENV{PATH} are preprocessed at startup to make |
| 297 | sure all paths are valid in the ANSI codepage (if possible). |
| 298 | |
| 299 | The Win32::GetLongPathName() function now returns the UTF-8 encoded |
| 300 | correct long file name instead of using replacement characters to force |
| 301 | the name into the ANSI codepage. The new Win32::GetANSIPathName() |
| 302 | function can be used to turn a long pathname into a short one only if the |
| 303 | long one cannot be represented in the ANSI codepage. |
| 304 | |
| 305 | Many other functions in the C<Win32> module have been improved to accept |
| 306 | UTF-8 encoded arguments. Please see L<Win32> for details. |
| 307 | |
| 308 | =head2 readpipe() is now overridable |
| 309 | |
| 310 | The built-in function readpipe() is now overridable. Overriding it permits |
| 311 | also to override its operator counterpart, C<qx//> (a.k.a. C<``>). |
| 312 | Moreover, it now defaults to C<$_> if no argument is provided. (Rafael |
| 313 | Garcia-Suarez) |
| 314 | |
| 315 | =head2 Default argument for readline() |
| 316 | |
| 317 | readline() now defaults to C<*ARGV> if no argument is provided. (Rafael |
| 318 | Garcia-Suarez) |
| 319 | |
| 320 | =head2 state() variables |
| 321 | |
| 322 | A new class of variables has been introduced. State variables are similar |
| 323 | to C<my> variables, but are declared with the C<state> keyword in place of |
| 324 | C<my>. They're visible only in their lexical scope, but their value is |
| 325 | persistent: unlike C<my> variables, they're not undefined at scope entry, |
| 326 | but retain their previous value. (Rafael Garcia-Suarez, Nicholas Clark) |
| 327 | |
| 328 | To use state variables, one needs to enable them by using |
| 329 | |
| 330 | use feature 'state'; |
| 331 | |
| 332 | or by using the C<-E> command-line switch in one-liners. |
| 333 | See L<perlsub/"Persistent Private Variables">. |
| 334 | |
| 335 | =head2 Stacked filetest operators |
| 336 | |
| 337 | As a new form of syntactic sugar, it's now possible to stack up filetest |
| 338 | operators. You can now write C<-f -w -x $file> in a row to mean |
| 339 | C<-x $file && -w _ && -f _>. See L<perlfunc/-X>. |
| 340 | |
| 341 | =head2 UNIVERSAL::DOES() |
| 342 | |
| 343 | The C<UNIVERSAL> class has a new method, C<DOES()>. It has been added to |
| 344 | solve semantic problems with the C<isa()> method. C<isa()> checks for |
| 345 | inheritance, while C<DOES()> has been designed to be overridden when |
| 346 | module authors use other types of relations between classes (in addition |
| 347 | to inheritance). (chromatic) |
| 348 | |
| 349 | See L<< UNIVERSAL/"$obj->DOES( ROLE )" >>. |
| 350 | |
| 351 | =head2 Formats |
| 352 | |
| 353 | Formats were improved in several ways. A new field, C<^*>, can be used for |
| 354 | variable-width, one-line-at-a-time text. Null characters are now handled |
| 355 | correctly in picture lines. Using C<@#> and C<~~> together will now |
| 356 | produce a compile-time error, as those format fields are incompatible. |
| 357 | L<perlform> has been improved, and miscellaneous bugs fixed. |
| 358 | |
| 359 | =head2 Byte-order modifiers for pack() and unpack() |
| 360 | |
| 361 | There are two new byte-order modifiers, C<E<gt>> (big-endian) and C<E<lt>> |
| 362 | (little-endian), that can be appended to most pack() and unpack() template |
| 363 | characters and groups to force a certain byte-order for that type or group. |
| 364 | See L<perlfunc/pack> and L<perlpacktut> for details. |
| 365 | |
| 366 | =head2 C<no VERSION> |
| 367 | |
| 368 | You can now use C<no> followed by a version number to specify that you |
| 369 | want to use a version of perl older than the specified one. |
| 370 | |
| 371 | =head2 C<chdir>, C<chmod> and C<chown> on filehandles |
| 372 | |
| 373 | C<chdir>, C<chmod> and C<chown> can now work on filehandles as well as |
| 374 | filenames, if the system supports respectively C<fchdir>, C<fchmod> and |
| 375 | C<fchown>, thanks to a patch provided by Gisle Aas. |
| 376 | |
| 377 | =head2 OS groups |
| 378 | |
| 379 | C<$(> and C<$)> now return groups in the order where the OS returns them, |
| 380 | thanks to Gisle Aas. This wasn't previously the case. |
| 381 | |
| 382 | =head2 Recursive sort subs |
| 383 | |
| 384 | You can now use recursive subroutines with sort(), thanks to Robin Houston. |
| 385 | |
| 386 | =head2 Exceptions in constant folding |
| 387 | |
| 388 | The constant folding routine is now wrapped in an exception handler, and |
| 389 | if folding throws an exception (such as attempting to evaluate 0/0), perl |
| 390 | now retains the current optree, rather than aborting the whole program. |
| 391 | Without this change, programs would not compile if they had expressions that |
| 392 | happened to generate exceptions, even though those expressions were in code |
| 393 | that could never be reached at runtime. (Nicholas Clark, Dave Mitchell) |
| 394 | |
| 395 | =head2 Source filters in @INC |
| 396 | |
| 397 | It's possible to enhance the mechanism of subroutine hooks in @INC by |
| 398 | adding a source filter on top of the filehandle opened and returned by the |
| 399 | hook. This feature was planned a long time ago, but wasn't quite working |
| 400 | until now. See L<perlfunc/require> for details. (Nicholas Clark) |
| 401 | |
| 402 | =head2 New internal variables |
| 403 | |
| 404 | =over 4 |
| 405 | |
| 406 | =item C<${^RE_DEBUG_FLAGS}> |
| 407 | |
| 408 | This variable controls what debug flags are in effect for the regular |
| 409 | expression engine when running under C<use re "debug">. See L<re> for |
| 410 | details. |
| 411 | |
| 412 | =item C<${^CHILD_ERROR_NATIVE}> |
| 413 | |
| 414 | This variable gives the native status returned by the last pipe close, |
| 415 | backtick command, successful call to wait() or waitpid(), or from the |
| 416 | system() operator. See L<perlvar> for details. (Contributed by Gisle Aas.) |
| 417 | |
| 418 | =item C<${^RE_TRIE_MAXBUF}> |
| 419 | |
| 420 | See L</"Trie optimisation of literal string alternations">. |
| 421 | |
| 422 | =item C<${^WIN32_SLOPPY_STAT}> |
| 423 | |
| 424 | See L</"Sloppy stat on Windows">. |
| 425 | |
| 426 | =back |
| 427 | |
| 428 | =head2 Miscellaneous |
| 429 | |
| 430 | C<unpack()> now defaults to unpacking the C<$_> variable. |
| 431 | |
| 432 | C<mkdir()> without arguments now defaults to C<$_>. |
| 433 | |
| 434 | The internal dump output has been improved, so that non-printable characters |
| 435 | such as newline and backspace are output in C<\x> notation, rather than |
| 436 | octal. |
| 437 | |
| 438 | The B<-C> option can no longer be used on the C<#!> line. It wasn't |
| 439 | working there anyway, since the standard streams are already set up |
| 440 | at this point in the execution of the perl interpreter. You can use |
| 441 | binmode() instead to get the desired behaviour. |
| 442 | |
| 443 | =head2 UCD 5.0.0 |
| 444 | |
| 445 | The copy of the Unicode Character Database included in Perl 5 has |
| 446 | been updated to version 5.0.0. |
| 447 | |
| 448 | =head2 MAD |
| 449 | |
| 450 | MAD, which stands for I<Miscellaneous Attribute Decoration>, is a |
| 451 | still-in-development work leading to a Perl 5 to Perl 6 converter. To |
| 452 | enable it, it's necessary to pass the argument C<-Dmad> to Configure. The |
| 453 | obtained perl isn't binary compatible with a regular perl 5.10, and has |
| 454 | space and speed penalties; moreover not all regression tests still pass |
| 455 | with it. (Larry Wall, Nicholas Clark) |
| 456 | |
| 457 | =head2 kill() on Windows |
| 458 | |
| 459 | On Windows platforms, C<kill(-9, $pid)> now kills a process tree. |
| 460 | (On Unix, this delivers the signal to all processes in the same process |
| 461 | group.) |
| 462 | |
| 463 | =head1 Incompatible Changes |
| 464 | |
| 465 | =head2 Packing and UTF-8 strings |
| 466 | |
| 467 | The semantics of pack() and unpack() regarding UTF-8-encoded data has been |
| 468 | changed. Processing is now by default character per character instead of |
| 469 | byte per byte on the underlying encoding. Notably, code that used things |
| 470 | like C<pack("a*", $string)> to see through the encoding of string will now |
| 471 | simply get back the original $string. Packed strings can also get upgraded |
| 472 | during processing when you store upgraded characters. You can get the old |
| 473 | behaviour by using C<use bytes>. |
| 474 | |
| 475 | To be consistent with pack(), the C<C0> in unpack() templates indicates |
| 476 | that the data is to be processed in character mode, i.e. character by |
| 477 | character; on the contrary, C<U0> in unpack() indicates UTF-8 mode, where |
| 478 | the packed string is processed in its UTF-8-encoded Unicode form on a byte |
| 479 | by byte basis. This is reversed with regard to perl 5.8.X, but now consistent |
| 480 | between pack() and unpack(). |
| 481 | |
| 482 | Moreover, C<C0> and C<U0> can also be used in pack() templates to specify |
| 483 | respectively character and byte modes. |
| 484 | |
| 485 | C<C0> and C<U0> in the middle of a pack or unpack format now switch to the |
| 486 | specified encoding mode, honoring parens grouping. Previously, parens were |
| 487 | ignored. |
| 488 | |
| 489 | Also, there is a new pack() character format, C<W>, which is intended to |
| 490 | replace the old C<C>. C<C> is kept for unsigned chars coded as bytes in |
| 491 | the strings internal representation. C<W> represents unsigned (logical) |
| 492 | character values, which can be greater than 255. It is therefore more |
| 493 | robust when dealing with potentially UTF-8-encoded data (as C<C> will wrap |
| 494 | values outside the range 0..255, and not respect the string encoding). |
| 495 | |
| 496 | In practice, that means that pack formats are now encoding-neutral, except |
| 497 | C<C>. |
| 498 | |
| 499 | For consistency, C<A> in unpack() format now trims all Unicode whitespace |
| 500 | from the end of the string. Before perl 5.9.2, it used to strip only the |
| 501 | classical ASCII space characters. |
| 502 | |
| 503 | =head2 Byte/character count feature in unpack() |
| 504 | |
| 505 | A new unpack() template character, C<".">, returns the number of bytes or |
| 506 | characters (depending on the selected encoding mode, see above) read so far. |
| 507 | |
| 508 | =head2 The C<$*> and C<$#> variables have been removed |
| 509 | |
| 510 | C<$*>, which was deprecated in favor of the C</s> and C</m> regexp |
| 511 | modifiers, has been removed. |
| 512 | |
| 513 | The deprecated C<$#> variable (output format for numbers) has been |
| 514 | removed. |
| 515 | |
| 516 | Two new severe warnings, C<$#/$* is no longer supported>, have been added. |
| 517 | |
| 518 | =head2 substr() lvalues are no longer fixed-length |
| 519 | |
| 520 | The lvalues returned by the three argument form of substr() used to be a |
| 521 | "fixed length window" on the original string. In some cases this could |
| 522 | cause surprising action at distance or other undefined behaviour. Now the |
| 523 | length of the window adjusts itself to the length of the string assigned to |
| 524 | it. |
| 525 | |
| 526 | =head2 Parsing of C<-f _> |
| 527 | |
| 528 | The identifier C<_> is now forced to be a bareword after a filetest |
| 529 | operator. This solves a number of misparsing issues when a global C<_> |
| 530 | subroutine is defined. |
| 531 | |
| 532 | =head2 C<:unique> |
| 533 | |
| 534 | The C<:unique> attribute has been made a no-op, since its current |
| 535 | implementation was fundamentally flawed and not threadsafe. |
| 536 | |
| 537 | =head2 Effect of pragmas in eval |
| 538 | |
| 539 | The compile-time value of the C<%^H> hint variable can now propagate into |
| 540 | eval("")uated code. This makes it more useful to implement lexical |
| 541 | pragmas. |
| 542 | |
| 543 | As a side-effect of this, the overloaded-ness of constants now propagates |
| 544 | into eval(""). |
| 545 | |
| 546 | =head2 chdir FOO |
| 547 | |
| 548 | A bareword argument to chdir() is now recognized as a file handle. |
| 549 | Earlier releases interpreted the bareword as a directory name. |
| 550 | (Gisle Aas) |
| 551 | |
| 552 | =head2 Handling of .pmc files |
| 553 | |
| 554 | An old feature of perl was that before C<require> or C<use> look for a |
| 555 | file with a F<.pm> extension, they will first look for a similar filename |
| 556 | with a F<.pmc> extension. If this file is found, it will be loaded in |
| 557 | place of any potentially existing file ending in a F<.pm> extension. |
| 558 | |
| 559 | Previously, F<.pmc> files were loaded only if more recent than the |
| 560 | matching F<.pm> file. Starting with 5.9.4, they'll be always loaded if |
| 561 | they exist. |
| 562 | |
| 563 | =head2 $^V is now a C<version> object instead of a v-string |
| 564 | |
| 565 | $^V can still be used with the C<%vd> format in printf, but any |
| 566 | character-level operations will now access the string representation |
| 567 | of the C<version> object and not the ordinals of a v-string. |
| 568 | Expressions like C<< substr($^V, 0, 2) >> or C<< split //, $^V >> |
| 569 | no longer work and must be rewritten. |
| 570 | |
| 571 | =head2 @- and @+ in patterns |
| 572 | |
| 573 | The special arrays C<@-> and C<@+> are no longer interpolated in regular |
| 574 | expressions. (Sadahiro Tomoyuki) |
| 575 | |
| 576 | =head2 $AUTOLOAD can now be tainted |
| 577 | |
| 578 | If you call a subroutine by a tainted name, and if it defers to an |
| 579 | AUTOLOAD function, then $AUTOLOAD will be (correctly) tainted. |
| 580 | (Rick Delaney) |
| 581 | |
| 582 | =head2 Tainting and printf |
| 583 | |
| 584 | When perl is run under taint mode, C<printf()> and C<sprintf()> will now |
| 585 | reject any tainted format argument. (Rafael Garcia-Suarez) |
| 586 | |
| 587 | =head2 undef and signal handlers |
| 588 | |
| 589 | Undefining or deleting a signal handler via C<undef $SIG{FOO}> is now |
| 590 | equivalent to setting it to C<'DEFAULT'>. (Rafael Garcia-Suarez) |
| 591 | |
| 592 | =head2 strictures and dereferencing in defined() |
| 593 | |
| 594 | C<use strict 'refs'> was ignoring taking a hard reference in an argument |
| 595 | to defined(), as in : |
| 596 | |
| 597 | use strict 'refs'; |
| 598 | my $x = 'foo'; |
| 599 | if (defined $$x) {...} |
| 600 | |
| 601 | This now correctly produces the run-time error C<Can't use string as a |
| 602 | SCALAR ref while "strict refs" in use>. |
| 603 | |
| 604 | C<defined @$foo> and C<defined %$bar> are now also subject to C<strict |
| 605 | 'refs'> (that is, C<$foo> and C<$bar> shall be proper references there.) |
| 606 | (C<defined(@foo)> and C<defined(%bar)> are discouraged constructs anyway.) |
| 607 | (Nicholas Clark) |
| 608 | |
| 609 | =head2 C<(?p{})> has been removed |
| 610 | |
| 611 | The regular expression construct C<(?p{})>, which was deprecated in perl |
| 612 | 5.8, has been removed. Use C<(??{})> instead. (Rafael Garcia-Suarez) |
| 613 | |
| 614 | =head2 Pseudo-hashes have been removed |
| 615 | |
| 616 | Support for pseudo-hashes has been removed from Perl 5.9. (The C<fields> |
| 617 | pragma remains here, but uses an alternate implementation.) |
| 618 | |
| 619 | =head2 Removal of the bytecode compiler and of perlcc |
| 620 | |
| 621 | C<perlcc>, the byteloader and the supporting modules (B::C, B::CC, |
| 622 | B::Bytecode, etc.) are no longer distributed with the perl sources. Those |
| 623 | experimental tools have never worked reliably, and, due to the lack of |
| 624 | volunteers to keep them in line with the perl interpreter developments, it |
| 625 | was decided to remove them instead of shipping a broken version of those. |
| 626 | The last version of those modules can be found with perl 5.9.4. |
| 627 | |
| 628 | However the B compiler framework stays supported in the perl core, as with |
| 629 | the more useful modules it has permitted (among others, B::Deparse and |
| 630 | B::Concise). |
| 631 | |
| 632 | =head2 Removal of the JPL |
| 633 | |
| 634 | The JPL (Java-Perl Lingo) has been removed from the perl sources tarball. |
| 635 | |
| 636 | =head2 Recursive inheritance detected earlier |
| 637 | |
| 638 | Perl will now immediately throw an exception if you modify any package's |
| 639 | C<@ISA> in such a way that it would cause recursive inheritance. |
| 640 | |
| 641 | Previously, the exception would not occur until Perl attempted to make |
| 642 | use of the recursive inheritance while resolving a method or doing a |
| 643 | C<$foo-E<gt>isa($bar)> lookup. |
| 644 | |
| 645 | =head2 warnings::enabled and warnings::warnif changed to favor users of modules |
| 646 | |
| 647 | The behaviour in 5.10.x favors the person using the module; |
| 648 | The behaviour in 5.8.x favors the module writer; |
| 649 | |
| 650 | Assume the following code: |
| 651 | |
| 652 | main calls Foo::Bar::baz() |
| 653 | Foo::Bar inherits from Foo::Base |
| 654 | Foo::Bar::baz() calls Foo::Base::_bazbaz() |
| 655 | Foo::Base::_bazbaz() calls: warnings::warnif('substr', 'some warning |
| 656 | message'); |
| 657 | |
| 658 | On 5.8.x, the code warns when Foo::Bar contains C<use warnings;> |
| 659 | It does not matter if Foo::Base or main have warnings enabled |
| 660 | to disable the warning one has to modify Foo::Bar. |
| 661 | |
| 662 | On 5.10.0 and newer, the code warns when main contains C<use warnings;> |
| 663 | It does not matter if Foo::Base or Foo::Bar have warnings enabled |
| 664 | to disable the warning one has to modify main. |
| 665 | |
| 666 | =head1 Modules and Pragmata |
| 667 | |
| 668 | =head2 Upgrading individual core modules |
| 669 | |
| 670 | Even more core modules are now also available separately through the |
| 671 | CPAN. If you wish to update one of these modules, you don't need to |
| 672 | wait for a new perl release. From within the cpan shell, running the |
| 673 | 'r' command will report on modules with upgrades available. See |
| 674 | C<perldoc CPAN> for more information. |
| 675 | |
| 676 | =head2 Pragmata Changes |
| 677 | |
| 678 | =over 4 |
| 679 | |
| 680 | =item C<feature> |
| 681 | |
| 682 | The new pragma C<feature> is used to enable new features that might break |
| 683 | old code. See L</"The C<feature> pragma"> above. |
| 684 | |
| 685 | =item C<mro> |
| 686 | |
| 687 | This new pragma enables to change the algorithm used to resolve inherited |
| 688 | methods. See L</"New Pragma, C<mro>"> above. |
| 689 | |
| 690 | =item Scoping of the C<sort> pragma |
| 691 | |
| 692 | The C<sort> pragma is now lexically scoped. Its effect used to be global. |
| 693 | |
| 694 | =item Scoping of C<bignum>, C<bigint>, C<bigrat> |
| 695 | |
| 696 | The three numeric pragmas C<bignum>, C<bigint> and C<bigrat> are now |
| 697 | lexically scoped. (Tels) |
| 698 | |
| 699 | =item C<base> |
| 700 | |
| 701 | The C<base> pragma now warns if a class tries to inherit from itself. |
| 702 | (Curtis "Ovid" Poe) |
| 703 | |
| 704 | =item C<strict> and C<warnings> |
| 705 | |
| 706 | C<strict> and C<warnings> will now complain loudly if they are loaded via |
| 707 | incorrect casing (as in C<use Strict;>). (Johan Vromans) |
| 708 | |
| 709 | =item C<version> |
| 710 | |
| 711 | The C<version> module provides support for version objects. |
| 712 | |
| 713 | =item C<warnings> |
| 714 | |
| 715 | The C<warnings> pragma doesn't load C<Carp> anymore. That means that code |
| 716 | that used C<Carp> routines without having loaded it at compile time might |
| 717 | need to be adjusted; typically, the following (faulty) code won't work |
| 718 | anymore, and will require parentheses to be added after the function name: |
| 719 | |
| 720 | use warnings; |
| 721 | require Carp; |
| 722 | Carp::confess 'argh'; |
| 723 | |
| 724 | =item C<less> |
| 725 | |
| 726 | C<less> now does something useful (or at least it tries to). In fact, it |
| 727 | has been turned into a lexical pragma. So, in your modules, you can now |
| 728 | test whether your users have requested to use less CPU, or less memory, |
| 729 | less magic, or maybe even less fat. See L<less> for more. (Joshua ben |
| 730 | Jore) |
| 731 | |
| 732 | =back |
| 733 | |
| 734 | =head2 New modules |
| 735 | |
| 736 | =over 4 |
| 737 | |
| 738 | =item * |
| 739 | |
| 740 | C<encoding::warnings>, by Audrey Tang, is a module to emit warnings |
| 741 | whenever an ASCII character string containing high-bit bytes is implicitly |
| 742 | converted into UTF-8. It's a lexical pragma since Perl 5.9.4; on older |
| 743 | perls, its effect is global. |
| 744 | |
| 745 | =item * |
| 746 | |
| 747 | C<Module::CoreList>, by Richard Clamp, is a small handy module that tells |
| 748 | you what versions of core modules ship with any versions of Perl 5. It |
| 749 | comes with a command-line frontend, C<corelist>. |
| 750 | |
| 751 | =item * |
| 752 | |
| 753 | C<Math::BigInt::FastCalc> is an XS-enabled, and thus faster, version of |
| 754 | C<Math::BigInt::Calc>. |
| 755 | |
| 756 | =item * |
| 757 | |
| 758 | C<Compress::Zlib> is an interface to the zlib compression library. It |
| 759 | comes with a bundled version of zlib, so having a working zlib is not a |
| 760 | prerequisite to install it. It's used by C<Archive::Tar> (see below). |
| 761 | |
| 762 | =item * |
| 763 | |
| 764 | C<IO::Zlib> is an C<IO::>-style interface to C<Compress::Zlib>. |
| 765 | |
| 766 | =item * |
| 767 | |
| 768 | C<Archive::Tar> is a module to manipulate C<tar> archives. |
| 769 | |
| 770 | =item * |
| 771 | |
| 772 | C<Digest::SHA> is a module used to calculate many types of SHA digests, |
| 773 | has been included for SHA support in the CPAN module. |
| 774 | |
| 775 | =item * |
| 776 | |
| 777 | C<ExtUtils::CBuilder> and C<ExtUtils::ParseXS> have been added. |
| 778 | |
| 779 | =item * |
| 780 | |
| 781 | C<Hash::Util::FieldHash>, by Anno Siegel, has been added. This module |
| 782 | provides support for I<field hashes>: hashes that maintain an association |
| 783 | of a reference with a value, in a thread-safe garbage-collected way. |
| 784 | Such hashes are useful to implement inside-out objects. |
| 785 | |
| 786 | =item * |
| 787 | |
| 788 | C<Module::Build>, by Ken Williams, has been added. It's an alternative to |
| 789 | C<ExtUtils::MakeMaker> to build and install perl modules. |
| 790 | |
| 791 | =item * |
| 792 | |
| 793 | C<Module::Load>, by Jos Boumans, has been added. It provides a single |
| 794 | interface to load Perl modules and F<.pl> files. |
| 795 | |
| 796 | =item * |
| 797 | |
| 798 | C<Module::Loaded>, by Jos Boumans, has been added. It's used to mark |
| 799 | modules as loaded or unloaded. |
| 800 | |
| 801 | =item * |
| 802 | |
| 803 | C<Package::Constants>, by Jos Boumans, has been added. It's a simple |
| 804 | helper to list all constants declared in a given package. |
| 805 | |
| 806 | =item * |
| 807 | |
| 808 | C<Win32API::File>, by Tye McQueen, has been added (for Windows builds). |
| 809 | This module provides low-level access to Win32 system API calls for |
| 810 | files/dirs. |
| 811 | |
| 812 | =item * |
| 813 | |
| 814 | C<Locale::Maketext::Simple>, needed by CPANPLUS, is a simple wrapper around |
| 815 | C<Locale::Maketext::Lexicon>. Note that C<Locale::Maketext::Lexicon> isn't |
| 816 | included in the perl core; the behaviour of C<Locale::Maketext::Simple> |
| 817 | gracefully degrades when the later isn't present. |
| 818 | |
| 819 | =item * |
| 820 | |
| 821 | C<Params::Check> implements a generic input parsing/checking mechanism. It |
| 822 | is used by CPANPLUS. |
| 823 | |
| 824 | =item * |
| 825 | |
| 826 | C<Term::UI> simplifies the task to ask questions at a terminal prompt. |
| 827 | |
| 828 | =item * |
| 829 | |
| 830 | C<Object::Accessor> provides an interface to create per-object accessors. |
| 831 | |
| 832 | =item * |
| 833 | |
| 834 | C<Module::Pluggable> is a simple framework to create modules that accept |
| 835 | pluggable sub-modules. |
| 836 | |
| 837 | =item * |
| 838 | |
| 839 | C<Module::Load::Conditional> provides simple ways to query and possibly |
| 840 | load installed modules. |
| 841 | |
| 842 | =item * |
| 843 | |
| 844 | C<Time::Piece> provides an object oriented interface to time functions, |
| 845 | overriding the built-ins localtime() and gmtime(). |
| 846 | |
| 847 | =item * |
| 848 | |
| 849 | C<IPC::Cmd> helps to find and run external commands, possibly |
| 850 | interactively. |
| 851 | |
| 852 | =item * |
| 853 | |
| 854 | C<File::Fetch> provide a simple generic file fetching mechanism. |
| 855 | |
| 856 | =item * |
| 857 | |
| 858 | C<Log::Message> and C<Log::Message::Simple> are used by the log facility |
| 859 | of C<CPANPLUS>. |
| 860 | |
| 861 | =item * |
| 862 | |
| 863 | C<Archive::Extract> is a generic archive extraction mechanism |
| 864 | for F<.tar> (plain, gzipped or bzipped) or F<.zip> files. |
| 865 | |
| 866 | =item * |
| 867 | |
| 868 | C<CPANPLUS> provides an API and a command-line tool to access the CPAN |
| 869 | mirrors. |
| 870 | |
| 871 | =item * |
| 872 | |
| 873 | C<Pod::Escapes> provides utilities that are useful in decoding Pod |
| 874 | EE<lt>...E<gt> sequences. |
| 875 | |
| 876 | =item * |
| 877 | |
| 878 | C<Pod::Simple> is now the backend for several of the Pod-related modules |
| 879 | included with Perl. |
| 880 | |
| 881 | =back |
| 882 | |
| 883 | =head2 Selected Changes to Core Modules |
| 884 | |
| 885 | =over 4 |
| 886 | |
| 887 | =item C<Attribute::Handlers> |
| 888 | |
| 889 | C<Attribute::Handlers> can now report the caller's file and line number. |
| 890 | (David Feldman) |
| 891 | |
| 892 | All interpreted attributes are now passed as array references. (Damian |
| 893 | Conway) |
| 894 | |
| 895 | =item C<B::Lint> |
| 896 | |
| 897 | C<B::Lint> is now based on C<Module::Pluggable>, and so can be extended |
| 898 | with plugins. (Joshua ben Jore) |
| 899 | |
| 900 | =item C<B> |
| 901 | |
| 902 | It's now possible to access the lexical pragma hints (C<%^H>) by using the |
| 903 | method B::COP::hints_hash(). It returns a C<B::RHE> object, which in turn |
| 904 | can be used to get a hash reference via the method B::RHE::HASH(). (Joshua |
| 905 | ben Jore) |
| 906 | |
| 907 | =item C<Thread> |
| 908 | |
| 909 | As the old 5005thread threading model has been removed, in favor of the |
| 910 | ithreads scheme, the C<Thread> module is now a compatibility wrapper, to |
| 911 | be used in old code only. It has been removed from the default list of |
| 912 | dynamic extensions. |
| 913 | |
| 914 | =back |
| 915 | |
| 916 | =head1 Utility Changes |
| 917 | |
| 918 | =over 4 |
| 919 | |
| 920 | =item perl -d |
| 921 | |
| 922 | The Perl debugger can now save all debugger commands for sourcing later; |
| 923 | notably, it can now emulate stepping backwards, by restarting and |
| 924 | rerunning all bar the last command from a saved command history. |
| 925 | |
| 926 | It can also display the parent inheritance tree of a given class, with the |
| 927 | C<i> command. |
| 928 | |
| 929 | =item ptar |
| 930 | |
| 931 | C<ptar> is a pure perl implementation of C<tar> that comes with |
| 932 | C<Archive::Tar>. |
| 933 | |
| 934 | =item ptardiff |
| 935 | |
| 936 | C<ptardiff> is a small utility used to generate a diff between the contents |
| 937 | of a tar archive and a directory tree. Like C<ptar>, it comes with |
| 938 | C<Archive::Tar>. |
| 939 | |
| 940 | =item shasum |
| 941 | |
| 942 | C<shasum> is a command-line utility, used to print or to check SHA |
| 943 | digests. It comes with the new C<Digest::SHA> module. |
| 944 | |
| 945 | =item corelist |
| 946 | |
| 947 | The C<corelist> utility is now installed with perl (see L</"New modules"> |
| 948 | above). |
| 949 | |
| 950 | =item h2ph and h2xs |
| 951 | |
| 952 | C<h2ph> and C<h2xs> have been made more robust with regard to |
| 953 | "modern" C code. |
| 954 | |
| 955 | C<h2xs> implements a new option C<--use-xsloader> to force use of |
| 956 | C<XSLoader> even in backwards compatible modules. |
| 957 | |
| 958 | The handling of authors' names that had apostrophes has been fixed. |
| 959 | |
| 960 | Any enums with negative values are now skipped. |
| 961 | |
| 962 | =item perlivp |
| 963 | |
| 964 | C<perlivp> no longer checks for F<*.ph> files by default. Use the new C<-a> |
| 965 | option to run I<all> tests. |
| 966 | |
| 967 | =item find2perl |
| 968 | |
| 969 | C<find2perl> now assumes C<-print> as a default action. Previously, it |
| 970 | needed to be specified explicitly. |
| 971 | |
| 972 | Several bugs have been fixed in C<find2perl>, regarding C<-exec> and |
| 973 | C<-eval>. Also the options C<-path>, C<-ipath> and C<-iname> have been |
| 974 | added. |
| 975 | |
| 976 | =item config_data |
| 977 | |
| 978 | C<config_data> is a new utility that comes with C<Module::Build>. It |
| 979 | provides a command-line interface to the configuration of Perl modules |
| 980 | that use Module::Build's framework of configurability (that is, |
| 981 | C<*::ConfigData> modules that contain local configuration information for |
| 982 | their parent modules.) |
| 983 | |
| 984 | =item cpanp |
| 985 | |
| 986 | C<cpanp>, the CPANPLUS shell, has been added. (C<cpanp-run-perl>, a |
| 987 | helper for CPANPLUS operation, has been added too, but isn't intended for |
| 988 | direct use). |
| 989 | |
| 990 | =item cpan2dist |
| 991 | |
| 992 | C<cpan2dist> is a new utility that comes with CPANPLUS. It's a tool to |
| 993 | create distributions (or packages) from CPAN modules. |
| 994 | |
| 995 | =item pod2html |
| 996 | |
| 997 | The output of C<pod2html> has been enhanced to be more customizable via |
| 998 | CSS. Some formatting problems were also corrected. (Jari Aalto) |
| 999 | |
| 1000 | =back |
| 1001 | |
| 1002 | =head1 New Documentation |
| 1003 | |
| 1004 | The L<perlpragma> manpage documents how to write one's own lexical |
| 1005 | pragmas in pure Perl (something that is possible starting with 5.9.4). |
| 1006 | |
| 1007 | The new L<perlglossary> manpage is a glossary of terms used in the Perl |
| 1008 | documentation, technical and otherwise, kindly provided by O'Reilly Media, |
| 1009 | Inc. |
| 1010 | |
| 1011 | The L<perlreguts> manpage, courtesy of Yves Orton, describes internals of the |
| 1012 | Perl regular expression engine. |
| 1013 | |
| 1014 | The L<perlreapi> manpage describes the interface to the perl interpreter |
| 1015 | used to write pluggable regular expression engines (by Ævar Arnfjörð |
| 1016 | Bjarmason). |
| 1017 | |
| 1018 | The L<perlunitut> manpage is an tutorial for programming with Unicode and |
| 1019 | string encodings in Perl, courtesy of Juerd Waalboer. |
| 1020 | |
| 1021 | A new manual page, L<perlunifaq> (the Perl Unicode FAQ), has been added |
| 1022 | (Juerd Waalboer). |
| 1023 | |
| 1024 | The L<perlcommunity> manpage gives a description of the Perl community |
| 1025 | on the Internet and in real life. (Edgar "Trizor" Bering) |
| 1026 | |
| 1027 | The L<CORE> manual page documents the C<CORE::> namespace. (Tels) |
| 1028 | |
| 1029 | The long-existing feature of C</(?{...})/> regexps setting C<$_> and pos() |
| 1030 | is now documented. |
| 1031 | |
| 1032 | =head1 Performance Enhancements |
| 1033 | |
| 1034 | =head2 In-place sorting |
| 1035 | |
| 1036 | Sorting arrays in place (C<@a = sort @a>) is now optimized to avoid |
| 1037 | making a temporary copy of the array. |
| 1038 | |
| 1039 | Likewise, C<reverse sort ...> is now optimized to sort in reverse, |
| 1040 | avoiding the generation of a temporary intermediate list. |
| 1041 | |
| 1042 | =head2 Lexical array access |
| 1043 | |
| 1044 | Access to elements of lexical arrays via a numeric constant between 0 and |
| 1045 | 255 is now faster. (This used to be only the case for global arrays.) |
| 1046 | |
| 1047 | =head2 XS-assisted SWASHGET |
| 1048 | |
| 1049 | Some pure-perl code that perl was using to retrieve Unicode properties and |
| 1050 | transliteration mappings has been reimplemented in XS. |
| 1051 | |
| 1052 | =head2 Constant subroutines |
| 1053 | |
| 1054 | The interpreter internals now support a far more memory efficient form of |
| 1055 | inlineable constants. Storing a reference to a constant value in a symbol |
| 1056 | table is equivalent to a full typeglob referencing a constant subroutine, |
| 1057 | but using about 400 bytes less memory. This proxy constant subroutine is |
| 1058 | automatically upgraded to a real typeglob with subroutine if necessary. |
| 1059 | The approach taken is analogous to the existing space optimisation for |
| 1060 | subroutine stub declarations, which are stored as plain scalars in place |
| 1061 | of the full typeglob. |
| 1062 | |
| 1063 | Several of the core modules have been converted to use this feature for |
| 1064 | their system dependent constants - as a result C<use POSIX;> now takes about |
| 1065 | 200K less memory. |
| 1066 | |
| 1067 | =head2 C<PERL_DONT_CREATE_GVSV> |
| 1068 | |
| 1069 | The new compilation flag C<PERL_DONT_CREATE_GVSV>, introduced as an option |
| 1070 | in perl 5.8.8, is turned on by default in perl 5.9.3. It prevents perl |
| 1071 | from creating an empty scalar with every new typeglob. See L<perl589delta> |
| 1072 | for details. |
| 1073 | |
| 1074 | =head2 Weak references are cheaper |
| 1075 | |
| 1076 | Weak reference creation is now I<O(1)> rather than I<O(n)>, courtesy of |
| 1077 | Nicholas Clark. Weak reference deletion remains I<O(n)>, but if deletion only |
| 1078 | happens at program exit, it may be skipped completely. |
| 1079 | |
| 1080 | =head2 sort() enhancements |
| 1081 | |
| 1082 | Salvador Fandiño provided improvements to reduce the memory usage of C<sort> |
| 1083 | and to speed up some cases. |
| 1084 | |
| 1085 | =head2 Memory optimisations |
| 1086 | |
| 1087 | Several internal data structures (typeglobs, GVs, CVs, formats) have been |
| 1088 | restructured to use less memory. (Nicholas Clark) |
| 1089 | |
| 1090 | =head2 UTF-8 cache optimisation |
| 1091 | |
| 1092 | The UTF-8 caching code is now more efficient, and used more often. |
| 1093 | (Nicholas Clark) |
| 1094 | |
| 1095 | =head2 Sloppy stat on Windows |
| 1096 | |
| 1097 | On Windows, perl's stat() function normally opens the file to determine |
| 1098 | the link count and update attributes that may have been changed through |
| 1099 | hard links. Setting ${^WIN32_SLOPPY_STAT} to a true value speeds up |
| 1100 | stat() by not performing this operation. (Jan Dubois) |
| 1101 | |
| 1102 | =head2 Regular expressions optimisations |
| 1103 | |
| 1104 | =over 4 |
| 1105 | |
| 1106 | =item Engine de-recursivised |
| 1107 | |
| 1108 | The regular expression engine is no longer recursive, meaning that |
| 1109 | patterns that used to overflow the stack will either die with useful |
| 1110 | explanations, or run to completion, which, since they were able to blow |
| 1111 | the stack before, will likely take a very long time to happen. If you were |
| 1112 | experiencing the occasional stack overflow (or segfault) and upgrade to |
| 1113 | discover that now perl apparently hangs instead, look for a degenerate |
| 1114 | regex. (Dave Mitchell) |
| 1115 | |
| 1116 | =item Single char char-classes treated as literals |
| 1117 | |
| 1118 | Classes of a single character are now treated the same as if the character |
| 1119 | had been used as a literal, meaning that code that uses char-classes as an |
| 1120 | escaping mechanism will see a speedup. (Yves Orton) |
| 1121 | |
| 1122 | =item Trie optimisation of literal string alternations |
| 1123 | |
| 1124 | Alternations, where possible, are optimised into more efficient matching |
| 1125 | structures. String literal alternations are merged into a trie and are |
| 1126 | matched simultaneously. This means that instead of O(N) time for matching |
| 1127 | N alternations at a given point, the new code performs in O(1) time. |
| 1128 | A new special variable, ${^RE_TRIE_MAXBUF}, has been added to fine-tune |
| 1129 | this optimization. (Yves Orton) |
| 1130 | |
| 1131 | B<Note:> Much code exists that works around perl's historic poor |
| 1132 | performance on alternations. Often the tricks used to do so will disable |
| 1133 | the new optimisations. Hopefully the utility modules used for this purpose |
| 1134 | will be educated about these new optimisations. |
| 1135 | |
| 1136 | =item Aho-Corasick start-point optimisation |
| 1137 | |
| 1138 | When a pattern starts with a trie-able alternation and there aren't |
| 1139 | better optimisations available, the regex engine will use Aho-Corasick |
| 1140 | matching to find the start point. (Yves Orton) |
| 1141 | |
| 1142 | =back |
| 1143 | |
| 1144 | =head1 Installation and Configuration Improvements |
| 1145 | |
| 1146 | =head2 Configuration improvements |
| 1147 | |
| 1148 | =over 4 |
| 1149 | |
| 1150 | =item C<-Dusesitecustomize> |
| 1151 | |
| 1152 | Run-time customization of @INC can be enabled by passing the |
| 1153 | C<-Dusesitecustomize> flag to Configure. When enabled, this will make perl |
| 1154 | run F<$sitelibexp/sitecustomize.pl> before anything else. This script can |
| 1155 | then be set up to add additional entries to @INC. |
| 1156 | |
| 1157 | =item Relocatable installations |
| 1158 | |
| 1159 | There is now Configure support for creating a relocatable perl tree. If |
| 1160 | you Configure with C<-Duserelocatableinc>, then the paths in @INC (and |
| 1161 | everything else in %Config) can be optionally located via the path of the |
| 1162 | perl executable. |
| 1163 | |
| 1164 | That means that, if the string C<".../"> is found at the start of any |
| 1165 | path, it's substituted with the directory of $^X. So, the relocation can |
| 1166 | be configured on a per-directory basis, although the default with |
| 1167 | C<-Duserelocatableinc> is that everything is relocated. The initial |
| 1168 | install is done to the original configured prefix. |
| 1169 | |
| 1170 | =item strlcat() and strlcpy() |
| 1171 | |
| 1172 | The configuration process now detects whether strlcat() and strlcpy() are |
| 1173 | available. When they are not available, perl's own version is used (from |
| 1174 | Russ Allbery's public domain implementation). Various places in the perl |
| 1175 | interpreter now use them. (Steve Peters) |
| 1176 | |
| 1177 | =item C<d_pseudofork> and C<d_printf_format_null> |
| 1178 | |
| 1179 | A new configuration variable, available as C<$Config{d_pseudofork}> in |
| 1180 | the L<Config> module, has been added, to distinguish real fork() support |
| 1181 | from fake pseudofork used on Windows platforms. |
| 1182 | |
| 1183 | A new configuration variable, C<d_printf_format_null>, has been added, |
| 1184 | to see if printf-like formats are allowed to be NULL. |
| 1185 | |
| 1186 | =item Configure help |
| 1187 | |
| 1188 | C<Configure -h> has been extended with the most commonly used options. |
| 1189 | |
| 1190 | =back |
| 1191 | |
| 1192 | =head2 Compilation improvements |
| 1193 | |
| 1194 | =over 4 |
| 1195 | |
| 1196 | =item Parallel build |
| 1197 | |
| 1198 | Parallel makes should work properly now, although there may still be problems |
| 1199 | if C<make test> is instructed to run in parallel. |
| 1200 | |
| 1201 | =item Borland's compilers support |
| 1202 | |
| 1203 | Building with Borland's compilers on Win32 should work more smoothly. In |
| 1204 | particular Steve Hay has worked to side step many warnings emitted by their |
| 1205 | compilers and at least one C compiler internal error. |
| 1206 | |
| 1207 | =item Static build on Windows |
| 1208 | |
| 1209 | Perl extensions on Windows now can be statically built into the Perl DLL. |
| 1210 | |
| 1211 | Also, it's now possible to build a C<perl-static.exe> that doesn't depend |
| 1212 | on the Perl DLL on Win32. See the Win32 makefiles for details. |
| 1213 | (Vadim Konovalov) |
| 1214 | |
| 1215 | =item ppport.h files |
| 1216 | |
| 1217 | All F<ppport.h> files in the XS modules bundled with perl are now |
| 1218 | autogenerated at build time. (Marcus Holland-Moritz) |
| 1219 | |
| 1220 | =item C++ compatibility |
| 1221 | |
| 1222 | Efforts have been made to make perl and the core XS modules compilable |
| 1223 | with various C++ compilers (although the situation is not perfect with |
| 1224 | some of the compilers on some of the platforms tested.) |
| 1225 | |
| 1226 | =item Support for Microsoft 64-bit compiler |
| 1227 | |
| 1228 | Support for building perl with Microsoft's 64-bit compiler has been |
| 1229 | improved. (ActiveState) |
| 1230 | |
| 1231 | =item Visual C++ |
| 1232 | |
| 1233 | Perl can now be compiled with Microsoft Visual C++ 2005 (and 2008 Beta 2). |
| 1234 | |
| 1235 | =item Win32 builds |
| 1236 | |
| 1237 | All win32 builds (MS-Win, WinCE) have been merged and cleaned up. |
| 1238 | |
| 1239 | =back |
| 1240 | |
| 1241 | =head2 Installation improvements |
| 1242 | |
| 1243 | =over 4 |
| 1244 | |
| 1245 | =item Module auxiliary files |
| 1246 | |
| 1247 | README files and changelogs for CPAN modules bundled with perl are no |
| 1248 | longer installed. |
| 1249 | |
| 1250 | =back |
| 1251 | |
| 1252 | =head2 New Or Improved Platforms |
| 1253 | |
| 1254 | Perl has been reported to work on Symbian OS. See L<perlsymbian> for more |
| 1255 | information. |
| 1256 | |
| 1257 | Many improvements have been made towards making Perl work correctly on |
| 1258 | z/OS. |
| 1259 | |
| 1260 | Perl has been reported to work on DragonFlyBSD and MidnightBSD. |
| 1261 | |
| 1262 | Perl has also been reported to work on NexentaOS |
| 1263 | ( http://www.gnusolaris.org/ ). |
| 1264 | |
| 1265 | The VMS port has been improved. See L<perlvms>. |
| 1266 | |
| 1267 | Support for Cray XT4 Catamount/Qk has been added. See |
| 1268 | F<hints/catamount.sh> in the source code distribution for more |
| 1269 | information. |
| 1270 | |
| 1271 | Vendor patches have been merged for RedHat and Gentoo. |
| 1272 | |
| 1273 | DynaLoader::dl_unload_file() now works on Windows. |
| 1274 | |
| 1275 | =head1 Selected Bug Fixes |
| 1276 | |
| 1277 | =over 4 |
| 1278 | |
| 1279 | =item strictures in regexp-eval blocks |
| 1280 | |
| 1281 | C<strict> wasn't in effect in regexp-eval blocks (C</(?{...})/>). |
| 1282 | |
| 1283 | =item Calling CORE::require() |
| 1284 | |
| 1285 | CORE::require() and CORE::do() were always parsed as require() and do() |
| 1286 | when they were overridden. This is now fixed. |
| 1287 | |
| 1288 | =item Subscripts of slices |
| 1289 | |
| 1290 | You can now use a non-arrowed form for chained subscripts after a list |
| 1291 | slice, like in: |
| 1292 | |
| 1293 | ({foo => "bar"})[0]{foo} |
| 1294 | |
| 1295 | This used to be a syntax error; a C<< -> >> was required. |
| 1296 | |
| 1297 | =item C<no warnings 'category'> works correctly with -w |
| 1298 | |
| 1299 | Previously when running with warnings enabled globally via C<-w>, selective |
| 1300 | disabling of specific warning categories would actually turn off all warnings. |
| 1301 | This is now fixed; now C<no warnings 'io';> will only turn off warnings in the |
| 1302 | C<io> class. Previously it would erroneously turn off all warnings. |
| 1303 | |
| 1304 | =item threads improvements |
| 1305 | |
| 1306 | Several memory leaks in ithreads were closed. Also, ithreads were made |
| 1307 | less memory-intensive. |
| 1308 | |
| 1309 | C<threads> is now a dual-life module, also available on CPAN. It has been |
| 1310 | expanded in many ways. A kill() method is available for thread signalling. |
| 1311 | One can get thread status, or the list of running or joinable threads. |
| 1312 | |
| 1313 | A new C<< threads->exit() >> method is used to exit from the application |
| 1314 | (this is the default for the main thread) or from the current thread only |
| 1315 | (this is the default for all other threads). On the other hand, the exit() |
| 1316 | built-in now always causes the whole application to terminate. (Jerry |
| 1317 | D. Hedden) |
| 1318 | |
| 1319 | =item chr() and negative values |
| 1320 | |
| 1321 | chr() on a negative value now gives C<\x{FFFD}>, the Unicode replacement |
| 1322 | character, unless when the C<bytes> pragma is in effect, where the low |
| 1323 | eight bits of the value are used. |
| 1324 | |
| 1325 | =item PERL5SHELL and tainting |
| 1326 | |
| 1327 | On Windows, the PERL5SHELL environment variable is now checked for |
| 1328 | taintedness. (Rafael Garcia-Suarez) |
| 1329 | |
| 1330 | =item Using *FILE{IO} |
| 1331 | |
| 1332 | C<stat()> and C<-X> filetests now treat *FILE{IO} filehandles like *FILE |
| 1333 | filehandles. (Steve Peters) |
| 1334 | |
| 1335 | =item Overloading and reblessing |
| 1336 | |
| 1337 | Overloading now works when references are reblessed into another class. |
| 1338 | Internally, this has been implemented by moving the flag for "overloading" |
| 1339 | from the reference to the referent, which logically is where it should |
| 1340 | always have been. (Nicholas Clark) |
| 1341 | |
| 1342 | =item Overloading and UTF-8 |
| 1343 | |
| 1344 | A few bugs related to UTF-8 handling with objects that have |
| 1345 | stringification overloaded have been fixed. (Nicholas Clark) |
| 1346 | |
| 1347 | =item eval memory leaks fixed |
| 1348 | |
| 1349 | Traditionally, C<eval 'syntax error'> has leaked badly. Many (but not all) |
| 1350 | of these leaks have now been eliminated or reduced. (Dave Mitchell) |
| 1351 | |
| 1352 | =item Random device on Windows |
| 1353 | |
| 1354 | In previous versions, perl would read the file F</dev/urandom> if it |
| 1355 | existed when seeding its random number generator. That file is unlikely |
| 1356 | to exist on Windows, and if it did would probably not contain appropriate |
| 1357 | data, so perl no longer tries to read it on Windows. (Alex Davies) |
| 1358 | |
| 1359 | =item PERLIO_DEBUG |
| 1360 | |
| 1361 | The C<PERLIO_DEBUG> environment variable no longer has any effect for |
| 1362 | setuid scripts and for scripts run with B<-T>. |
| 1363 | |
| 1364 | Moreover, with a thread-enabled perl, using C<PERLIO_DEBUG> could lead to |
| 1365 | an internal buffer overflow. This has been fixed. |
| 1366 | |
| 1367 | =item PerlIO::scalar and read-only scalars |
| 1368 | |
| 1369 | PerlIO::scalar will now prevent writing to read-only scalars. Moreover, |
| 1370 | seek() is now supported with PerlIO::scalar-based filehandles, the |
| 1371 | underlying string being zero-filled as needed. (Rafael, Jarkko Hietaniemi) |
| 1372 | |
| 1373 | =item study() and UTF-8 |
| 1374 | |
| 1375 | study() never worked for UTF-8 strings, but could lead to false results. |
| 1376 | It's now a no-op on UTF-8 data. (Yves Orton) |
| 1377 | |
| 1378 | =item Critical signals |
| 1379 | |
| 1380 | The signals SIGILL, SIGBUS and SIGSEGV are now always delivered in an |
| 1381 | "unsafe" manner (contrary to other signals, that are deferred until the |
| 1382 | perl interpreter reaches a reasonably stable state; see |
| 1383 | L<perlipc/"Deferred Signals (Safe Signals)">). (Rafael) |
| 1384 | |
| 1385 | =item @INC-hook fix |
| 1386 | |
| 1387 | When a module or a file is loaded through an @INC-hook, and when this hook |
| 1388 | has set a filename entry in %INC, __FILE__ is now set for this module |
| 1389 | accordingly to the contents of that %INC entry. (Rafael) |
| 1390 | |
| 1391 | =item C<-t> switch fix |
| 1392 | |
| 1393 | The C<-w> and C<-t> switches can now be used together without messing |
| 1394 | up which categories of warnings are activated. (Rafael) |
| 1395 | |
| 1396 | =item Duping UTF-8 filehandles |
| 1397 | |
| 1398 | Duping a filehandle which has the C<:utf8> PerlIO layer set will now |
| 1399 | properly carry that layer on the duped filehandle. (Rafael) |
| 1400 | |
| 1401 | =item Localisation of hash elements |
| 1402 | |
| 1403 | Localizing a hash element whose key was given as a variable didn't work |
| 1404 | correctly if the variable was changed while the local() was in effect (as |
| 1405 | in C<local $h{$x}; ++$x>). (Bo Lindbergh) |
| 1406 | |
| 1407 | =back |
| 1408 | |
| 1409 | =head1 New or Changed Diagnostics |
| 1410 | |
| 1411 | =over 4 |
| 1412 | |
| 1413 | =item Use of uninitialized value |
| 1414 | |
| 1415 | Perl will now try to tell you the name of the variable (if any) that was |
| 1416 | undefined. |
| 1417 | |
| 1418 | =item Deprecated use of my() in false conditional |
| 1419 | |
| 1420 | A new deprecation warning, I<Deprecated use of my() in false conditional>, |
| 1421 | has been added, to warn against the use of the dubious and deprecated |
| 1422 | construct |
| 1423 | |
| 1424 | my $x if 0; |
| 1425 | |
| 1426 | See L<perldiag>. Use C<state> variables instead. |
| 1427 | |
| 1428 | =item !=~ should be !~ |
| 1429 | |
| 1430 | A new warning, C<!=~ should be !~>, is emitted to prevent this misspelling |
| 1431 | of the non-matching operator. |
| 1432 | |
| 1433 | =item Newline in left-justified string |
| 1434 | |
| 1435 | The warning I<Newline in left-justified string> has been removed. |
| 1436 | |
| 1437 | =item Too late for "-T" option |
| 1438 | |
| 1439 | The error I<Too late for "-T" option> has been reformulated to be more |
| 1440 | descriptive. |
| 1441 | |
| 1442 | =item "%s" variable %s masks earlier declaration |
| 1443 | |
| 1444 | This warning is now emitted in more consistent cases; in short, when one |
| 1445 | of the declarations involved is a C<my> variable: |
| 1446 | |
| 1447 | my $x; my $x; # warns |
| 1448 | my $x; our $x; # warns |
| 1449 | our $x; my $x; # warns |
| 1450 | |
| 1451 | On the other hand, the following: |
| 1452 | |
| 1453 | our $x; our $x; |
| 1454 | |
| 1455 | now gives a C<"our" variable %s redeclared> warning. |
| 1456 | |
| 1457 | =item readdir()/closedir()/etc. attempted on invalid dirhandle |
| 1458 | |
| 1459 | These new warnings are now emitted when a dirhandle is used but is |
| 1460 | either closed or not really a dirhandle. |
| 1461 | |
| 1462 | =item Opening dirhandle/filehandle %s also as a file/directory |
| 1463 | |
| 1464 | Two deprecation warnings have been added: (Rafael) |
| 1465 | |
| 1466 | Opening dirhandle %s also as a file |
| 1467 | Opening filehandle %s also as a directory |
| 1468 | |
| 1469 | =item Use of -P is deprecated |
| 1470 | |
| 1471 | Perl's command-line switch C<-P> is now deprecated. |
| 1472 | |
| 1473 | =item v-string in use/require is non-portable |
| 1474 | |
| 1475 | Perl will warn you against potential backwards compatibility problems with |
| 1476 | the C<use VERSION> syntax. |
| 1477 | |
| 1478 | =item perl -V |
| 1479 | |
| 1480 | C<perl -V> has several improvements, making it more useable from shell |
| 1481 | scripts to get the value of configuration variables. See L<perlrun> for |
| 1482 | details. |
| 1483 | |
| 1484 | =back |
| 1485 | |
| 1486 | =head1 Changed Internals |
| 1487 | |
| 1488 | In general, the source code of perl has been refactored, tidied up, |
| 1489 | and optimized in many places. Also, memory management and allocation |
| 1490 | has been improved in several points. |
| 1491 | |
| 1492 | When compiling the perl core with gcc, as many gcc warning flags are |
| 1493 | turned on as is possible on the platform. (This quest for cleanliness |
| 1494 | doesn't extend to XS code because we cannot guarantee the tidiness of |
| 1495 | code we didn't write.) Similar strictness flags have been added or |
| 1496 | tightened for various other C compilers. |
| 1497 | |
| 1498 | =head2 Reordering of SVt_* constants |
| 1499 | |
| 1500 | The relative ordering of constants that define the various types of C<SV> |
| 1501 | have changed; in particular, C<SVt_PVGV> has been moved before C<SVt_PVLV>, |
| 1502 | C<SVt_PVAV>, C<SVt_PVHV> and C<SVt_PVCV>. This is unlikely to make any |
| 1503 | difference unless you have code that explicitly makes assumptions about that |
| 1504 | ordering. (The inheritance hierarchy of C<B::*> objects has been changed |
| 1505 | to reflect this.) |
| 1506 | |
| 1507 | =head2 Elimination of SVt_PVBM |
| 1508 | |
| 1509 | Related to this, the internal type C<SVt_PVBM> has been removed. This |
| 1510 | dedicated type of C<SV> was used by the C<index> operator and parts of the |
| 1511 | regexp engine to facilitate fast Boyer-Moore matches. Its use internally has |
| 1512 | been replaced by C<SV>s of type C<SVt_PVGV>. |
| 1513 | |
| 1514 | =head2 New type SVt_BIND |
| 1515 | |
| 1516 | A new type C<SVt_BIND> has been added, in readiness for the project to |
| 1517 | implement Perl 6 on 5. There deliberately is no implementation yet, and |
| 1518 | they cannot yet be created or destroyed. |
| 1519 | |
| 1520 | =head2 Removal of CPP symbols |
| 1521 | |
| 1522 | The C preprocessor symbols C<PERL_PM_APIVERSION> and |
| 1523 | C<PERL_XS_APIVERSION>, which were supposed to give the version number of |
| 1524 | the oldest perl binary-compatible (resp. source-compatible) with the |
| 1525 | present one, were not used, and sometimes had misleading values. They have |
| 1526 | been removed. |
| 1527 | |
| 1528 | =head2 Less space is used by ops |
| 1529 | |
| 1530 | The C<BASEOP> structure now uses less space. The C<op_seq> field has been |
| 1531 | removed and replaced by a single bit bit-field C<op_opt>. C<op_type> is now 9 |
| 1532 | bits long. (Consequently, the C<B::OP> class doesn't provide an C<seq> |
| 1533 | method anymore.) |
| 1534 | |
| 1535 | =head2 New parser |
| 1536 | |
| 1537 | perl's parser is now generated by bison (it used to be generated by |
| 1538 | byacc.) As a result, it seems to be a bit more robust. |
| 1539 | |
| 1540 | Also, Dave Mitchell improved the lexer debugging output under C<-DT>. |
| 1541 | |
| 1542 | =head2 Use of C<const> |
| 1543 | |
| 1544 | Andy Lester supplied many improvements to determine which function |
| 1545 | parameters and local variables could actually be declared C<const> to the C |
| 1546 | compiler. Steve Peters provided new C<*_set> macros and reworked the core to |
| 1547 | use these rather than assigning to macros in LVALUE context. |
| 1548 | |
| 1549 | =head2 Mathoms |
| 1550 | |
| 1551 | A new file, F<mathoms.c>, has been added. It contains functions that are |
| 1552 | no longer used in the perl core, but that remain available for binary or |
| 1553 | source compatibility reasons. However, those functions will not be |
| 1554 | compiled in if you add C<-DNO_MATHOMS> in the compiler flags. |
| 1555 | |
| 1556 | =head2 C<AvFLAGS> has been removed |
| 1557 | |
| 1558 | The C<AvFLAGS> macro has been removed. |
| 1559 | |
| 1560 | =head2 C<av_*> changes |
| 1561 | |
| 1562 | The C<av_*()> functions, used to manipulate arrays, no longer accept null |
| 1563 | C<AV*> parameters. |
| 1564 | |
| 1565 | =head2 $^H and %^H |
| 1566 | |
| 1567 | The implementation of the special variables $^H and %^H has changed, to |
| 1568 | allow implementing lexical pragmas in pure Perl. |
| 1569 | |
| 1570 | =head2 B:: modules inheritance changed |
| 1571 | |
| 1572 | The inheritance hierarchy of C<B::> modules has changed; C<B::NV> now |
| 1573 | inherits from C<B::SV> (it used to inherit from C<B::IV>). |
| 1574 | |
| 1575 | =head2 Anonymous hash and array constructors |
| 1576 | |
| 1577 | The anonymous hash and array constructors now take 1 op in the optree |
| 1578 | instead of 3, now that pp_anonhash and pp_anonlist return a reference to |
| 1579 | an hash/array when the op is flagged with OPf_SPECIAL. (Nicholas Clark) |
| 1580 | |
| 1581 | =head1 Known Problems |
| 1582 | |
| 1583 | There's still a remaining problem in the implementation of the lexical |
| 1584 | C<$_>: it doesn't work inside C</(?{...})/> blocks. (See the TODO test in |
| 1585 | F<t/op/mydef.t>.) |
| 1586 | |
| 1587 | Stacked filetest operators won't work when the C<filetest> pragma is in |
| 1588 | effect, because they rely on the stat() buffer C<_> being populated, and |
| 1589 | filetest bypasses stat(). |
| 1590 | |
| 1591 | =head2 UTF-8 problems |
| 1592 | |
| 1593 | The handling of Unicode still is unclean in several places, where it's |
| 1594 | dependent on whether a string is internally flagged as UTF-8. This will |
| 1595 | be made more consistent in perl 5.12, but that won't be possible without |
| 1596 | a certain amount of backwards incompatibility. |
| 1597 | |
| 1598 | =head1 Platform Specific Problems |
| 1599 | |
| 1600 | When compiled with g++ and thread support on Linux, it's reported that the |
| 1601 | C<$!> stops working correctly. This is related to the fact that the glibc |
| 1602 | provides two strerror_r(3) implementation, and perl selects the wrong |
| 1603 | one. |
| 1604 | |
| 1605 | =head1 Reporting Bugs |
| 1606 | |
| 1607 | If you find what you think is a bug, you might check the articles |
| 1608 | recently posted to the comp.lang.perl.misc newsgroup and the perl |
| 1609 | bug database at http://rt.perl.org/rt3/ . There may also be |
| 1610 | information at http://www.perl.org/ , the Perl Home Page. |
| 1611 | |
| 1612 | If you believe you have an unreported bug, please run the B<perlbug> |
| 1613 | program included with your release. Be sure to trim your bug down |
| 1614 | to a tiny but sufficient test case. Your bug report, along with the |
| 1615 | output of C<perl -V>, will be sent off to perlbug@perl.org to be |
| 1616 | analysed by the Perl porting team. |
| 1617 | |
| 1618 | =head1 SEE ALSO |
| 1619 | |
| 1620 | The F<Changes> file and the perl590delta to perl595delta man pages for |
| 1621 | exhaustive details on what changed. |
| 1622 | |
| 1623 | The F<INSTALL> file for how to build Perl. |
| 1624 | |
| 1625 | The F<README> file for general stuff. |
| 1626 | |
| 1627 | The F<Artistic> and F<Copying> files for copyright information. |
| 1628 | |
| 1629 | =cut |