Commit | Line | Data |
---|---|---|
6ee623d5 GS |
1 | =head1 NAME |
2 | ||
3 | perldelta - what's new for perl5.004 | |
4 | ||
5 | =head1 DESCRIPTION | |
6 | ||
7 | This document describes differences between the 5.003 release (as | |
8 | documented in I<Programming Perl>, second edition--the Camel Book) and | |
9 | this one. | |
10 | ||
11 | =head1 Supported Environments | |
12 | ||
13 | Perl5.004 builds out of the box on Unix, Plan 9, LynxOS, VMS, OS/2, | |
14 | QNX, AmigaOS, and Windows NT. Perl runs on Windows 95 as well, but it | |
15 | cannot be built there, for lack of a reasonable command interpreter. | |
16 | ||
17 | =head1 Core Changes | |
18 | ||
19 | Most importantly, many bugs were fixed, including several security | |
20 | problems. See the F<Changes> file in the distribution for details. | |
21 | ||
22 | =head2 List assignment to %ENV works | |
23 | ||
24 | C<%ENV = ()> and C<%ENV = @list> now work as expected (except on VMS | |
25 | where it generates a fatal error). | |
26 | ||
13a2d996 SP |
27 | =head2 Change to "Can't locate Foo.pm in @INC" error |
28 | ||
29 | The error "Can't locate Foo.pm in @INC" now lists the contents of @INC | |
30 | for easier debugging. | |
6ee623d5 GS |
31 | |
32 | =head2 Compilation option: Binary compatibility with 5.003 | |
33 | ||
34 | There is a new Configure question that asks if you want to maintain | |
35 | binary compatibility with Perl 5.003. If you choose binary | |
36 | compatibility, you do not have to recompile your extensions, but you | |
37 | might have symbol conflicts if you embed Perl in another application, | |
38 | just as in the 5.003 release. By default, binary compatibility | |
39 | is preserved at the expense of symbol table pollution. | |
40 | ||
41 | =head2 $PERL5OPT environment variable | |
42 | ||
43 | You may now put Perl options in the $PERL5OPT environment variable. | |
44 | Unless Perl is running with taint checks, it will interpret this | |
45 | variable as if its contents had appeared on a "#!perl" line at the | |
46 | beginning of your script, except that hyphens are optional. PERL5OPT | |
47 | may only be used to set the following switches: B<-[DIMUdmw]>. | |
48 | ||
49 | =head2 Limitations on B<-M>, B<-m>, and B<-T> options | |
50 | ||
51 | The C<-M> and C<-m> options are no longer allowed on the C<#!> line of | |
52 | a script. If a script needs a module, it should invoke it with the | |
53 | C<use> pragma. | |
54 | ||
55 | The B<-T> option is also forbidden on the C<#!> line of a script, | |
56 | unless it was present on the Perl command line. Due to the way C<#!> | |
57 | works, this usually means that B<-T> must be in the first argument. | |
58 | Thus: | |
59 | ||
60 | #!/usr/bin/perl -T -w | |
61 | ||
62 | will probably work for an executable script invoked as C<scriptname>, | |
63 | while: | |
64 | ||
65 | #!/usr/bin/perl -w -T | |
66 | ||
67 | will probably fail under the same conditions. (Non-Unix systems will | |
68 | probably not follow this rule.) But C<perl scriptname> is guaranteed | |
69 | to fail, since then there is no chance of B<-T> being found on the | |
70 | command line before it is found on the C<#!> line. | |
71 | ||
72 | =head2 More precise warnings | |
73 | ||
74 | If you removed the B<-w> option from your Perl 5.003 scripts because it | |
75 | made Perl too verbose, we recommend that you try putting it back when | |
76 | you upgrade to Perl 5.004. Each new perl version tends to remove some | |
77 | undesirable warnings, while adding new warnings that may catch bugs in | |
78 | your scripts. | |
79 | ||
80 | =head2 Deprecated: Inherited C<AUTOLOAD> for non-methods | |
81 | ||
82 | Before Perl 5.004, C<AUTOLOAD> functions were looked up as methods | |
83 | (using the C<@ISA> hierarchy), even when the function to be autoloaded | |
84 | was called as a plain function (e.g. C<Foo::bar()>), not a method | |
c47ff5f1 | 85 | (e.g. C<< Foo->bar() >> or C<< $obj->bar() >>). |
6ee623d5 GS |
86 | |
87 | Perl 5.005 will use method lookup only for methods' C<AUTOLOAD>s. | |
88 | However, there is a significant base of existing code that may be using | |
89 | the old behavior. So, as an interim step, Perl 5.004 issues an optional | |
90 | warning when a non-method uses an inherited C<AUTOLOAD>. | |
91 | ||
92 | The simple rule is: Inheritance will not work when autoloading | |
93 | non-methods. The simple fix for old code is: In any module that used to | |
94 | depend on inheriting C<AUTOLOAD> for non-methods from a base class named | |
95 | C<BaseClass>, execute C<*AUTOLOAD = \&BaseClass::AUTOLOAD> during startup. | |
96 | ||
97 | =head2 Previously deprecated %OVERLOAD is no longer usable | |
98 | ||
99 | Using %OVERLOAD to define overloading was deprecated in 5.003. | |
100 | Overloading is now defined using the overload pragma. %OVERLOAD is | |
101 | still used internally but should not be used by Perl scripts. See | |
102 | L<overload> for more details. | |
103 | ||
104 | =head2 Subroutine arguments created only when they're modified | |
105 | ||
106 | In Perl 5.004, nonexistent array and hash elements used as subroutine | |
107 | parameters are brought into existence only if they are actually | |
108 | assigned to (via C<@_>). | |
109 | ||
110 | Earlier versions of Perl vary in their handling of such arguments. | |
111 | Perl versions 5.002 and 5.003 always brought them into existence. | |
112 | Perl versions 5.000 and 5.001 brought them into existence only if | |
113 | they were not the first argument (which was almost certainly a bug). | |
114 | Earlier versions of Perl never brought them into existence. | |
115 | ||
116 | For example, given this code: | |
117 | ||
118 | undef @a; undef %a; | |
119 | sub show { print $_[0] }; | |
120 | sub change { $_[0]++ }; | |
121 | show($a[2]); | |
122 | change($a{b}); | |
123 | ||
124 | After this code executes in Perl 5.004, $a{b} exists but $a[2] does | |
125 | not. In Perl 5.002 and 5.003, both $a{b} and $a[2] would have existed | |
126 | (but $a[2]'s value would have been undefined). | |
127 | ||
128 | =head2 Group vector changeable with C<$)> | |
129 | ||
130 | The C<$)> special variable has always (well, in Perl 5, at least) | |
131 | reflected not only the current effective group, but also the group list | |
132 | as returned by the C<getgroups()> C function (if there is one). | |
133 | However, until this release, there has not been a way to call the | |
134 | C<setgroups()> C function from Perl. | |
135 | ||
136 | In Perl 5.004, assigning to C<$)> is exactly symmetrical with examining | |
137 | it: The first number in its string value is used as the effective gid; | |
138 | if there are any numbers after the first one, they are passed to the | |
139 | C<setgroups()> C function (if there is one). | |
140 | ||
141 | =head2 Fixed parsing of $$<digit>, &$<digit>, etc. | |
142 | ||
143 | Perl versions before 5.004 misinterpreted any type marker followed by | |
144 | "$" and a digit. For example, "$$0" was incorrectly taken to mean | |
145 | "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004. | |
146 | ||
147 | However, the developers of Perl 5.004 could not fix this bug completely, | |
148 | because at least two widely-used modules depend on the old meaning of | |
149 | "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the | |
150 | old (broken) way inside strings; but it generates this message as a | |
151 | warning. And in Perl 5.005, this special treatment will cease. | |
152 | ||
153 | =head2 Fixed localization of $<digit>, $&, etc. | |
154 | ||
155 | Perl versions before 5.004 did not always properly localize the | |
156 | regex-related special variables. Perl 5.004 does localize them, as | |
157 | the documentation has always said it should. This may result in $1, | |
158 | $2, etc. no longer being set where existing programs use them. | |
159 | ||
160 | =head2 No resetting of $. on implicit close | |
161 | ||
162 | The documentation for Perl 5.0 has always stated that C<$.> is I<not> | |
163 | reset when an already-open file handle is reopened with no intervening | |
164 | call to C<close>. Due to a bug, perl versions 5.000 through 5.003 | |
165 | I<did> reset C<$.> under that circumstance; Perl 5.004 does not. | |
166 | ||
167 | =head2 C<wantarray> may return undef | |
168 | ||
169 | The C<wantarray> operator returns true if a subroutine is expected to | |
170 | return a list, and false otherwise. In Perl 5.004, C<wantarray> can | |
171 | also return the undefined value if a subroutine's return value will | |
172 | not be used at all, which allows subroutines to avoid a time-consuming | |
173 | calculation of a return value if it isn't going to be used. | |
174 | ||
175 | =head2 C<eval EXPR> determines value of EXPR in scalar context | |
176 | ||
177 | Perl (version 5) used to determine the value of EXPR inconsistently, | |
178 | sometimes incorrectly using the surrounding context for the determination. | |
179 | Now, the value of EXPR (before being parsed by eval) is always determined in | |
180 | a scalar context. Once parsed, it is executed as before, by providing | |
181 | the context that the scope surrounding the eval provided. This change | |
182 | makes the behavior Perl4 compatible, besides fixing bugs resulting from | |
183 | the inconsistent behavior. This program: | |
184 | ||
185 | @a = qw(time now is time); | |
186 | print eval @a; | |
187 | print '|', scalar eval @a; | |
188 | ||
189 | used to print something like "timenowis881399109|4", but now (and in perl4) | |
190 | prints "4|4". | |
191 | ||
192 | =head2 Changes to tainting checks | |
193 | ||
194 | A bug in previous versions may have failed to detect some insecure | |
195 | conditions when taint checks are turned on. (Taint checks are used | |
196 | in setuid or setgid scripts, or when explicitly turned on with the | |
197 | C<-T> invocation option.) Although it's unlikely, this may cause a | |
198 | previously-working script to now fail -- which should be construed | |
199 | as a blessing, since that indicates a potentially-serious security | |
200 | hole was just plugged. | |
201 | ||
202 | The new restrictions when tainting include: | |
203 | ||
13a2d996 | 204 | =over 4 |
6ee623d5 GS |
205 | |
206 | =item No glob() or <*> | |
207 | ||
208 | These operators may spawn the C shell (csh), which cannot be made | |
209 | safe. This restriction will be lifted in a future version of Perl | |
210 | when globbing is implemented without the use of an external program. | |
211 | ||
212 | =item No spawning if tainted $CDPATH, $ENV, $BASH_ENV | |
213 | ||
214 | These environment variables may alter the behavior of spawned programs | |
215 | (especially shells) in ways that subvert security. So now they are | |
216 | treated as dangerous, in the manner of $IFS and $PATH. | |
217 | ||
218 | =item No spawning if tainted $TERM doesn't look like a terminal name | |
219 | ||
220 | Some termcap libraries do unsafe things with $TERM. However, it would be | |
221 | unnecessarily harsh to treat all $TERM values as unsafe, since only shell | |
222 | metacharacters can cause trouble in $TERM. So a tainted $TERM is | |
223 | considered to be safe if it contains only alphanumerics, underscores, | |
224 | dashes, and colons, and unsafe if it contains other characters (including | |
225 | whitespace). | |
226 | ||
227 | =back | |
228 | ||
229 | =head2 New Opcode module and revised Safe module | |
230 | ||
231 | A new Opcode module supports the creation, manipulation and | |
232 | application of opcode masks. The revised Safe module has a new API | |
233 | and is implemented using the new Opcode module. Please read the new | |
234 | Opcode and Safe documentation. | |
235 | ||
236 | =head2 Embedding improvements | |
237 | ||
238 | In older versions of Perl it was not possible to create more than one | |
239 | Perl interpreter instance inside a single process without leaking like a | |
240 | sieve and/or crashing. The bugs that caused this behavior have all been | |
241 | fixed. However, you still must take care when embedding Perl in a C | |
242 | program. See the updated perlembed manpage for tips on how to manage | |
243 | your interpreters. | |
244 | ||
245 | =head2 Internal change: FileHandle class based on IO::* classes | |
246 | ||
247 | File handles are now stored internally as type IO::Handle. The | |
248 | FileHandle module is still supported for backwards compatibility, but | |
249 | it is now merely a front end to the IO::* modules -- specifically, | |
250 | IO::Handle, IO::Seekable, and IO::File. We suggest, but do not | |
251 | require, that you use the IO::* modules in new code. | |
252 | ||
253 | In harmony with this change, C<*GLOB{FILEHANDLE}> is now just a | |
254 | backward-compatible synonym for C<*GLOB{IO}>. | |
255 | ||
256 | =head2 Internal change: PerlIO abstraction interface | |
257 | ||
258 | It is now possible to build Perl with AT&T's sfio IO package | |
259 | instead of stdio. See L<perlapio> for more details, and | |
260 | the F<INSTALL> file for how to use it. | |
261 | ||
262 | =head2 New and changed syntax | |
263 | ||
13a2d996 | 264 | =over 4 |
6ee623d5 GS |
265 | |
266 | =item $coderef->(PARAMS) | |
267 | ||
268 | A subroutine reference may now be suffixed with an arrow and a | |
269 | (possibly empty) parameter list. This syntax denotes a call of the | |
270 | referenced subroutine, with the given parameters (if any). | |
271 | ||
c47ff5f1 GS |
272 | This new syntax follows the pattern of S<C<< $hashref->{FOO} >>> and |
273 | S<C<< $aryref->[$foo] >>>: You may now write S<C<&$subref($foo)>> as | |
274 | S<C<< $subref->($foo) >>>. All these arrow terms may be chained; | |
275 | thus, S<C<< &{$table->{FOO}}($bar) >>> may now be written | |
276 | S<C<< $table->{FOO}->($bar) >>>. | |
6ee623d5 GS |
277 | |
278 | =back | |
279 | ||
280 | =head2 New and changed builtin constants | |
281 | ||
13a2d996 | 282 | =over 4 |
6ee623d5 GS |
283 | |
284 | =item __PACKAGE__ | |
285 | ||
286 | The current package name at compile time, or the undefined value if | |
287 | there is no current package (due to a C<package;> directive). Like | |
288 | C<__FILE__> and C<__LINE__>, C<__PACKAGE__> does I<not> interpolate | |
289 | into strings. | |
290 | ||
291 | =back | |
292 | ||
293 | =head2 New and changed builtin variables | |
294 | ||
13a2d996 | 295 | =over 4 |
6ee623d5 GS |
296 | |
297 | =item $^E | |
298 | ||
299 | Extended error message on some platforms. (Also known as | |
300 | $EXTENDED_OS_ERROR if you C<use English>). | |
301 | ||
302 | =item $^H | |
303 | ||
304 | The current set of syntax checks enabled by C<use strict>. See the | |
305 | documentation of C<strict> for more details. Not actually new, but | |
306 | newly documented. | |
307 | Because it is intended for internal use by Perl core components, | |
308 | there is no C<use English> long name for this variable. | |
309 | ||
310 | =item $^M | |
311 | ||
312 | By default, running out of memory it is not trappable. However, if | |
313 | compiled for this, Perl may use the contents of C<$^M> as an emergency | |
314 | pool after die()ing with this message. Suppose that your Perl were | |
315 | compiled with -DPERL_EMERGENCY_SBRK and used Perl's malloc. Then | |
316 | ||
317 | $^M = 'a' x (1<<16); | |
318 | ||
319 | would allocate a 64K buffer for use when in emergency. | |
320 | See the F<INSTALL> file for information on how to enable this option. | |
321 | As a disincentive to casual use of this advanced feature, | |
322 | there is no C<use English> long name for this variable. | |
323 | ||
324 | =back | |
325 | ||
326 | =head2 New and changed builtin functions | |
327 | ||
13a2d996 | 328 | =over 4 |
6ee623d5 GS |
329 | |
330 | =item delete on slices | |
331 | ||
332 | This now works. (e.g. C<delete @ENV{'PATH', 'MANPATH'}>) | |
333 | ||
334 | =item flock | |
335 | ||
336 | is now supported on more platforms, prefers fcntl to lockf when | |
337 | emulating, and always flushes before (un)locking. | |
338 | ||
339 | =item printf and sprintf | |
340 | ||
341 | Perl now implements these functions itself; it doesn't use the C | |
342 | library function sprintf() any more, except for floating-point | |
343 | numbers, and even then only known flags are allowed. As a result, it | |
344 | is now possible to know which conversions and flags will work, and | |
345 | what they will do. | |
346 | ||
347 | The new conversions in Perl's sprintf() are: | |
348 | ||
349 | %i a synonym for %d | |
350 | %p a pointer (the address of the Perl value, in hexadecimal) | |
351 | %n special: *stores* the number of characters output so far | |
352 | into the next variable in the parameter list | |
353 | ||
354 | The new flags that go between the C<%> and the conversion are: | |
355 | ||
356 | # prefix octal with "0", hex with "0x" | |
357 | h interpret integer as C type "short" or "unsigned short" | |
358 | V interpret integer as Perl's standard integer type | |
359 | ||
360 | Also, where a number would appear in the flags, an asterisk ("*") may | |
361 | be used instead, in which case Perl uses the next item in the | |
362 | parameter list as the given number (that is, as the field width or | |
363 | precision). If a field width obtained through "*" is negative, it has | |
364 | the same effect as the '-' flag: left-justification. | |
365 | ||
366 | See L<perlfunc/sprintf> for a complete list of conversion and flags. | |
367 | ||
368 | =item keys as an lvalue | |
369 | ||
370 | As an lvalue, C<keys> allows you to increase the number of hash buckets | |
371 | allocated for the given hash. This can gain you a measure of efficiency if | |
372 | you know the hash is going to get big. (This is similar to pre-extending | |
373 | an array by assigning a larger number to $#array.) If you say | |
374 | ||
375 | keys %hash = 200; | |
376 | ||
377 | then C<%hash> will have at least 200 buckets allocated for it. These | |
378 | buckets will be retained even if you do C<%hash = ()>; use C<undef | |
379 | %hash> if you want to free the storage while C<%hash> is still in scope. | |
380 | You can't shrink the number of buckets allocated for the hash using | |
381 | C<keys> in this way (but you needn't worry about doing this by accident, | |
382 | as trying has no effect). | |
383 | ||
384 | =item my() in Control Structures | |
385 | ||
386 | You can now use my() (with or without the parentheses) in the control | |
387 | expressions of control structures such as: | |
388 | ||
389 | while (defined(my $line = <>)) { | |
390 | $line = lc $line; | |
391 | } continue { | |
392 | print $line; | |
393 | } | |
394 | ||
395 | if ((my $answer = <STDIN>) =~ /^y(es)?$/i) { | |
396 | user_agrees(); | |
397 | } elsif ($answer =~ /^n(o)?$/i) { | |
398 | user_disagrees(); | |
399 | } else { | |
400 | chomp $answer; | |
401 | die "`$answer' is neither `yes' nor `no'"; | |
402 | } | |
403 | ||
404 | Also, you can declare a foreach loop control variable as lexical by | |
405 | preceding it with the word "my". For example, in: | |
406 | ||
407 | foreach my $i (1, 2, 3) { | |
408 | some_function(); | |
409 | } | |
410 | ||
411 | $i is a lexical variable, and the scope of $i extends to the end of | |
412 | the loop, but not beyond it. | |
413 | ||
414 | Note that you still cannot use my() on global punctuation variables | |
415 | such as $_ and the like. | |
416 | ||
417 | =item pack() and unpack() | |
418 | ||
419 | A new format 'w' represents a BER compressed integer (as defined in | |
420 | ASN.1). Its format is a sequence of one or more bytes, each of which | |
421 | provides seven bits of the total value, with the most significant | |
422 | first. Bit eight of each byte is set, except for the last byte, in | |
423 | which bit eight is clear. | |
424 | ||
425 | If 'p' or 'P' are given undef as values, they now generate a NULL | |
426 | pointer. | |
427 | ||
428 | Both pack() and unpack() now fail when their templates contain invalid | |
429 | types. (Invalid types used to be ignored.) | |
430 | ||
431 | =item sysseek() | |
432 | ||
433 | The new sysseek() operator is a variant of seek() that sets and gets the | |
434 | file's system read/write position, using the lseek(2) system call. It is | |
435 | the only reliable way to seek before using sysread() or syswrite(). Its | |
436 | return value is the new position, or the undefined value on failure. | |
437 | ||
438 | =item use VERSION | |
439 | ||
440 | If the first argument to C<use> is a number, it is treated as a version | |
441 | number instead of a module name. If the version of the Perl interpreter | |
442 | is less than VERSION, then an error message is printed and Perl exits | |
443 | immediately. Because C<use> occurs at compile time, this check happens | |
444 | immediately during the compilation process, unlike C<require VERSION>, | |
445 | which waits until runtime for the check. This is often useful if you | |
446 | need to check the current Perl version before C<use>ing library modules | |
447 | which have changed in incompatible ways from older versions of Perl. | |
448 | (We try not to do this more than we have to.) | |
449 | ||
450 | =item use Module VERSION LIST | |
451 | ||
452 | If the VERSION argument is present between Module and LIST, then the | |
453 | C<use> will call the VERSION method in class Module with the given | |
454 | version as an argument. The default VERSION method, inherited from | |
455 | the UNIVERSAL class, croaks if the given version is larger than the | |
456 | value of the variable $Module::VERSION. (Note that there is not a | |
457 | comma after VERSION!) | |
458 | ||
459 | This version-checking mechanism is similar to the one currently used | |
460 | in the Exporter module, but it is faster and can be used with modules | |
461 | that don't use the Exporter. It is the recommended method for new | |
462 | code. | |
463 | ||
464 | =item prototype(FUNCTION) | |
465 | ||
466 | Returns the prototype of a function as a string (or C<undef> if the | |
467 | function has no prototype). FUNCTION is a reference to or the name of the | |
468 | function whose prototype you want to retrieve. | |
469 | (Not actually new; just never documented before.) | |
470 | ||
471 | =item srand | |
472 | ||
473 | The default seed for C<srand>, which used to be C<time>, has been changed. | |
474 | Now it's a heady mix of difficult-to-predict system-dependent values, | |
475 | which should be sufficient for most everyday purposes. | |
476 | ||
477 | Previous to version 5.004, calling C<rand> without first calling C<srand> | |
478 | would yield the same sequence of random numbers on most or all machines. | |
479 | Now, when perl sees that you're calling C<rand> and haven't yet called | |
480 | C<srand>, it calls C<srand> with the default seed. You should still call | |
481 | C<srand> manually if your code might ever be run on a pre-5.004 system, | |
482 | of course, or if you want a seed other than the default. | |
483 | ||
484 | =item $_ as Default | |
485 | ||
486 | Functions documented in the Camel to default to $_ now in | |
487 | fact do, and all those that do are so documented in L<perlfunc>. | |
488 | ||
489 | =item C<m//gc> does not reset search position on failure | |
490 | ||
491 | The C<m//g> match iteration construct has always reset its target | |
492 | string's search position (which is visible through the C<pos> operator) | |
493 | when a match fails; as a result, the next C<m//g> match after a failure | |
494 | starts again at the beginning of the string. With Perl 5.004, this | |
495 | reset may be disabled by adding the "c" (for "continue") modifier, | |
496 | i.e. C<m//gc>. This feature, in conjunction with the C<\G> zero-width | |
497 | assertion, makes it possible to chain matches together. See L<perlop> | |
498 | and L<perlre>. | |
499 | ||
500 | =item C<m//x> ignores whitespace before ?*+{} | |
501 | ||
502 | The C<m//x> construct has always been intended to ignore all unescaped | |
503 | whitespace. However, before Perl 5.004, whitespace had the effect of | |
504 | escaping repeat modifiers like "*" or "?"; for example, C</a *b/x> was | |
505 | (mis)interpreted as C</a\*b/x>. This bug has been fixed in 5.004. | |
506 | ||
507 | =item nested C<sub{}> closures work now | |
508 | ||
509 | Prior to the 5.004 release, nested anonymous functions didn't work | |
510 | right. They do now. | |
511 | ||
512 | =item formats work right on changing lexicals | |
513 | ||
514 | Just like anonymous functions that contain lexical variables | |
515 | that change (like a lexical index variable for a C<foreach> loop), | |
516 | formats now work properly. For example, this silently failed | |
517 | before (printed only zeros), but is fine now: | |
518 | ||
519 | my $i; | |
520 | foreach $i ( 1 .. 10 ) { | |
521 | write; | |
522 | } | |
523 | format = | |
524 | my i is @# | |
525 | $i | |
526 | . | |
527 | ||
528 | However, it still fails (without a warning) if the foreach is within a | |
529 | subroutine: | |
530 | ||
531 | my $i; | |
532 | sub foo { | |
533 | foreach $i ( 1 .. 10 ) { | |
534 | write; | |
535 | } | |
536 | } | |
537 | foo; | |
538 | format = | |
539 | my i is @# | |
540 | $i | |
541 | . | |
542 | ||
543 | =back | |
544 | ||
545 | =head2 New builtin methods | |
546 | ||
547 | The C<UNIVERSAL> package automatically contains the following methods that | |
548 | are inherited by all other classes: | |
549 | ||
13a2d996 | 550 | =over 4 |
6ee623d5 GS |
551 | |
552 | =item isa(CLASS) | |
553 | ||
554 | C<isa> returns I<true> if its object is blessed into a subclass of C<CLASS> | |
555 | ||
556 | C<isa> is also exportable and can be called as a sub with two arguments. This | |
557 | allows the ability to check what a reference points to. Example: | |
558 | ||
559 | use UNIVERSAL qw(isa); | |
560 | ||
561 | if(isa($ref, 'ARRAY')) { | |
562 | ... | |
563 | } | |
564 | ||
565 | =item can(METHOD) | |
566 | ||
567 | C<can> checks to see if its object has a method called C<METHOD>, | |
568 | if it does then a reference to the sub is returned; if it does not then | |
569 | I<undef> is returned. | |
570 | ||
571 | =item VERSION( [NEED] ) | |
572 | ||
573 | C<VERSION> returns the version number of the class (package). If the | |
574 | NEED argument is given then it will check that the current version (as | |
575 | defined by the $VERSION variable in the given package) not less than | |
576 | NEED; it will die if this is not the case. This method is normally | |
577 | called as a class method. This method is called automatically by the | |
578 | C<VERSION> form of C<use>. | |
579 | ||
580 | use A 1.2 qw(some imported subs); | |
581 | # implies: | |
582 | A->VERSION(1.2); | |
583 | ||
584 | =back | |
585 | ||
586 | B<NOTE:> C<can> directly uses Perl's internal code for method lookup, and | |
587 | C<isa> uses a very similar method and caching strategy. This may cause | |
588 | strange effects if the Perl code dynamically changes @ISA in any package. | |
589 | ||
590 | You may add other methods to the UNIVERSAL class via Perl or XS code. | |
591 | You do not need to C<use UNIVERSAL> in order to make these methods | |
592 | available to your program. This is necessary only if you wish to | |
593 | have C<isa> available as a plain subroutine in the current package. | |
594 | ||
595 | =head2 TIEHANDLE now supported | |
596 | ||
597 | See L<perltie> for other kinds of tie()s. | |
598 | ||
13a2d996 | 599 | =over 4 |
6ee623d5 GS |
600 | |
601 | =item TIEHANDLE classname, LIST | |
602 | ||
603 | This is the constructor for the class. That means it is expected to | |
604 | return an object of some sort. The reference can be used to | |
605 | hold some internal information. | |
606 | ||
607 | sub TIEHANDLE { | |
608 | print "<shout>\n"; | |
609 | my $i; | |
610 | return bless \$i, shift; | |
611 | } | |
612 | ||
613 | =item PRINT this, LIST | |
614 | ||
615 | This method will be triggered every time the tied handle is printed to. | |
616 | Beyond its self reference it also expects the list that was passed to | |
617 | the print function. | |
618 | ||
619 | sub PRINT { | |
620 | $r = shift; | |
621 | $$r++; | |
622 | return print join( $, => map {uc} @_), $\; | |
623 | } | |
624 | ||
625 | =item PRINTF this, LIST | |
626 | ||
627 | This method will be triggered every time the tied handle is printed to | |
628 | with the C<printf()> function. | |
629 | Beyond its self reference it also expects the format and list that was | |
630 | passed to the printf function. | |
631 | ||
632 | sub PRINTF { | |
633 | shift; | |
634 | my $fmt = shift; | |
635 | print sprintf($fmt, @_)."\n"; | |
636 | } | |
637 | ||
638 | =item READ this LIST | |
639 | ||
640 | This method will be called when the handle is read from via the C<read> | |
641 | or C<sysread> functions. | |
642 | ||
643 | sub READ { | |
644 | $r = shift; | |
645 | my($buf,$len,$offset) = @_; | |
646 | print "READ called, \$buf=$buf, \$len=$len, \$offset=$offset"; | |
647 | } | |
648 | ||
649 | =item READLINE this | |
650 | ||
651 | This method will be called when the handle is read from. The method | |
652 | should return undef when there is no more data. | |
653 | ||
654 | sub READLINE { | |
655 | $r = shift; | |
656 | return "PRINT called $$r times\n" | |
657 | } | |
658 | ||
659 | =item GETC this | |
660 | ||
661 | This method will be called when the C<getc> function is called. | |
662 | ||
663 | sub GETC { print "Don't GETC, Get Perl"; return "a"; } | |
664 | ||
665 | =item DESTROY this | |
666 | ||
667 | As with the other types of ties, this method will be called when the | |
668 | tied handle is about to be destroyed. This is useful for debugging and | |
669 | possibly for cleaning up. | |
670 | ||
671 | sub DESTROY { | |
672 | print "</shout>\n"; | |
673 | } | |
674 | ||
675 | =back | |
676 | ||
677 | =head2 Malloc enhancements | |
678 | ||
679 | If perl is compiled with the malloc included with the perl distribution | |
680 | (that is, if C<perl -V:d_mymalloc> is 'define') then you can print | |
681 | memory statistics at runtime by running Perl thusly: | |
682 | ||
683 | env PERL_DEBUG_MSTATS=2 perl your_script_here | |
684 | ||
685 | The value of 2 means to print statistics after compilation and on | |
686 | exit; with a value of 1, the statistics are printed only on exit. | |
687 | (If you want the statistics at an arbitrary time, you'll need to | |
688 | install the optional module Devel::Peek.) | |
689 | ||
690 | Three new compilation flags are recognized by malloc.c. (They have no | |
691 | effect if perl is compiled with system malloc().) | |
692 | ||
13a2d996 | 693 | =over 4 |
6ee623d5 GS |
694 | |
695 | =item -DPERL_EMERGENCY_SBRK | |
696 | ||
697 | If this macro is defined, running out of memory need not be a fatal | |
698 | error: a memory pool can allocated by assigning to the special | |
699 | variable C<$^M>. See L<"$^M">. | |
700 | ||
701 | =item -DPACK_MALLOC | |
702 | ||
703 | Perl memory allocation is by bucket with sizes close to powers of two. | |
704 | Because of these malloc overhead may be big, especially for data of | |
705 | size exactly a power of two. If C<PACK_MALLOC> is defined, perl uses | |
706 | a slightly different algorithm for small allocations (up to 64 bytes | |
707 | long), which makes it possible to have overhead down to 1 byte for | |
708 | allocations which are powers of two (and appear quite often). | |
709 | ||
710 | Expected memory savings (with 8-byte alignment in C<alignbytes>) is | |
711 | about 20% for typical Perl usage. Expected slowdown due to additional | |
712 | malloc overhead is in fractions of a percent (hard to measure, because | |
713 | of the effect of saved memory on speed). | |
714 | ||
715 | =item -DTWO_POT_OPTIMIZE | |
716 | ||
717 | Similarly to C<PACK_MALLOC>, this macro improves allocations of data | |
718 | with size close to a power of two; but this works for big allocations | |
719 | (starting with 16K by default). Such allocations are typical for big | |
720 | hashes and special-purpose scripts, especially image processing. | |
721 | ||
722 | On recent systems, the fact that perl requires 2M from system for 1M | |
723 | allocation will not affect speed of execution, since the tail of such | |
724 | a chunk is not going to be touched (and thus will not require real | |
725 | memory). However, it may result in a premature out-of-memory error. | |
726 | So if you will be manipulating very large blocks with sizes close to | |
727 | powers of two, it would be wise to define this macro. | |
728 | ||
729 | Expected saving of memory is 0-100% (100% in applications which | |
730 | require most memory in such 2**n chunks); expected slowdown is | |
731 | negligible. | |
732 | ||
733 | =back | |
734 | ||
735 | =head2 Miscellaneous efficiency enhancements | |
736 | ||
737 | Functions that have an empty prototype and that do nothing but return | |
738 | a fixed value are now inlined (e.g. C<sub PI () { 3.14159 }>). | |
739 | ||
740 | Each unique hash key is only allocated once, no matter how many hashes | |
741 | have an entry with that key. So even if you have 100 copies of the | |
742 | same hash, the hash keys never have to be reallocated. | |
743 | ||
744 | =head1 Support for More Operating Systems | |
745 | ||
746 | Support for the following operating systems is new in Perl 5.004. | |
747 | ||
748 | =head2 Win32 | |
749 | ||
750 | Perl 5.004 now includes support for building a "native" perl under | |
751 | Windows NT, using the Microsoft Visual C++ compiler (versions 2.0 | |
752 | and above) or the Borland C++ compiler (versions 5.02 and above). | |
753 | The resulting perl can be used under Windows 95 (if it | |
754 | is installed in the same directory locations as it got installed | |
755 | in Windows NT). This port includes support for perl extension | |
756 | building tools like L<MakeMaker> and L<h2xs>, so that many extensions | |
757 | available on the Comprehensive Perl Archive Network (CPAN) can now be | |
758 | readily built under Windows NT. See http://www.perl.com/ for more | |
7b8d334a GS |
759 | information on CPAN and F<README.win32> in the perl distribution for more |
760 | details on how to get started with building this port. | |
6ee623d5 GS |
761 | |
762 | There is also support for building perl under the Cygwin32 environment. | |
763 | Cygwin32 is a set of GNU tools that make it possible to compile and run | |
19799a22 | 764 | many Unix programs under Windows NT by providing a mostly Unix-like |
7b8d334a GS |
765 | interface for compilation and execution. See F<README.cygwin32> in the |
766 | perl distribution for more details on this port and how to obtain the | |
767 | Cygwin32 toolkit. | |
6ee623d5 GS |
768 | |
769 | =head2 Plan 9 | |
770 | ||
7b8d334a | 771 | See F<README.plan9> in the perl distribution. |
6ee623d5 GS |
772 | |
773 | =head2 QNX | |
774 | ||
7b8d334a | 775 | See F<README.qnx> in the perl distribution. |
6ee623d5 GS |
776 | |
777 | =head2 AmigaOS | |
778 | ||
7b8d334a | 779 | See F<README.amigaos> in the perl distribution. |
6ee623d5 GS |
780 | |
781 | =head1 Pragmata | |
782 | ||
783 | Six new pragmatic modules exist: | |
784 | ||
13a2d996 | 785 | =over 4 |
6ee623d5 GS |
786 | |
787 | =item use autouse MODULE => qw(sub1 sub2 sub3) | |
788 | ||
789 | Defers C<require MODULE> until someone calls one of the specified | |
790 | subroutines (which must be exported by MODULE). This pragma should be | |
791 | used with caution, and only when necessary. | |
792 | ||
793 | =item use blib | |
794 | ||
795 | =item use blib 'dir' | |
796 | ||
797 | Looks for MakeMaker-like I<'blib'> directory structure starting in | |
798 | I<dir> (or current directory) and working back up to five levels of | |
799 | parent directories. | |
800 | ||
801 | Intended for use on command line with B<-M> option as a way of testing | |
802 | arbitrary scripts against an uninstalled version of a package. | |
803 | ||
804 | =item use constant NAME => VALUE | |
805 | ||
806 | Provides a convenient interface for creating compile-time constants, | |
807 | See L<perlsub/"Constant Functions">. | |
808 | ||
809 | =item use locale | |
810 | ||
811 | Tells the compiler to enable (or disable) the use of POSIX locales for | |
812 | builtin operations. | |
813 | ||
814 | When C<use locale> is in effect, the current LC_CTYPE locale is used | |
815 | for regular expressions and case mapping; LC_COLLATE for string | |
a31a806a | 816 | ordering; and LC_NUMERIC for numeric formatting in printf and sprintf |
6ee623d5 GS |
817 | (but B<not> in print). LC_NUMERIC is always used in write, since |
818 | lexical scoping of formats is problematic at best. | |
819 | ||
820 | Each C<use locale> or C<no locale> affects statements to the end of | |
821 | the enclosing BLOCK or, if not inside a BLOCK, to the end of the | |
822 | current file. Locales can be switched and queried with | |
823 | POSIX::setlocale(). | |
824 | ||
825 | See L<perllocale> for more information. | |
826 | ||
827 | =item use ops | |
828 | ||
829 | Disable unsafe opcodes, or any named opcodes, when compiling Perl code. | |
830 | ||
831 | =item use vmsish | |
832 | ||
833 | Enable VMS-specific language features. Currently, there are three | |
834 | VMS-specific features available: 'status', which makes C<$?> and | |
835 | C<system> return genuine VMS status values instead of emulating POSIX; | |
836 | 'exit', which makes C<exit> take a genuine VMS status value instead of | |
837 | assuming that C<exit 1> is an error; and 'time', which makes all times | |
838 | relative to the local time zone, in the VMS tradition. | |
839 | ||
840 | =back | |
841 | ||
842 | =head1 Modules | |
843 | ||
844 | =head2 Required Updates | |
845 | ||
846 | Though Perl 5.004 is compatible with almost all modules that work | |
847 | with Perl 5.003, there are a few exceptions: | |
848 | ||
849 | Module Required Version for Perl 5.004 | |
850 | ------ ------------------------------- | |
851 | Filter Filter-1.12 | |
852 | LWP libwww-perl-5.08 | |
853 | Tk Tk400.202 (-w makes noise) | |
854 | ||
855 | Also, the majordomo mailing list program, version 1.94.1, doesn't work | |
856 | with Perl 5.004 (nor with perl 4), because it executes an invalid | |
857 | regular expression. This bug is fixed in majordomo version 1.94.2. | |
858 | ||
859 | =head2 Installation directories | |
860 | ||
861 | The I<installperl> script now places the Perl source files for | |
862 | extensions in the architecture-specific library directory, which is | |
863 | where the shared libraries for extensions have always been. This | |
864 | change is intended to allow administrators to keep the Perl 5.004 | |
865 | library directory unchanged from a previous version, without running | |
866 | the risk of binary incompatibility between extensions' Perl source and | |
867 | shared libraries. | |
868 | ||
869 | =head2 Module information summary | |
870 | ||
871 | Brand new modules, arranged by topic rather than strictly | |
872 | alphabetically: | |
873 | ||
874 | CGI.pm Web server interface ("Common Gateway Interface") | |
875 | CGI/Apache.pm Support for Apache's Perl module | |
876 | CGI/Carp.pm Log server errors with helpful context | |
877 | CGI/Fast.pm Support for FastCGI (persistent server process) | |
878 | CGI/Push.pm Support for server push | |
879 | CGI/Switch.pm Simple interface for multiple server types | |
880 | ||
881 | CPAN Interface to Comprehensive Perl Archive Network | |
882 | CPAN::FirstTime Utility for creating CPAN configuration file | |
883 | CPAN::Nox Runs CPAN while avoiding compiled extensions | |
884 | ||
885 | IO.pm Top-level interface to IO::* classes | |
886 | IO/File.pm IO::File extension Perl module | |
887 | IO/Handle.pm IO::Handle extension Perl module | |
888 | IO/Pipe.pm IO::Pipe extension Perl module | |
889 | IO/Seekable.pm IO::Seekable extension Perl module | |
890 | IO/Select.pm IO::Select extension Perl module | |
891 | IO/Socket.pm IO::Socket extension Perl module | |
892 | ||
893 | Opcode.pm Disable named opcodes when compiling Perl code | |
894 | ||
895 | ExtUtils/Embed.pm Utilities for embedding Perl in C programs | |
896 | ExtUtils/testlib.pm Fixes up @INC to use just-built extension | |
897 | ||
898 | FindBin.pm Find path of currently executing program | |
899 | ||
900 | Class/Struct.pm Declare struct-like datatypes as Perl classes | |
901 | File/stat.pm By-name interface to Perl's builtin stat | |
902 | Net/hostent.pm By-name interface to Perl's builtin gethost* | |
903 | Net/netent.pm By-name interface to Perl's builtin getnet* | |
904 | Net/protoent.pm By-name interface to Perl's builtin getproto* | |
905 | Net/servent.pm By-name interface to Perl's builtin getserv* | |
906 | Time/gmtime.pm By-name interface to Perl's builtin gmtime | |
907 | Time/localtime.pm By-name interface to Perl's builtin localtime | |
908 | Time/tm.pm Internal object for Time::{gm,local}time | |
909 | User/grent.pm By-name interface to Perl's builtin getgr* | |
910 | User/pwent.pm By-name interface to Perl's builtin getpw* | |
911 | ||
912 | Tie/RefHash.pm Base class for tied hashes with references as keys | |
913 | ||
914 | UNIVERSAL.pm Base class for *ALL* classes | |
915 | ||
916 | =head2 Fcntl | |
917 | ||
918 | New constants in the existing Fcntl modules are now supported, | |
919 | provided that your operating system happens to support them: | |
920 | ||
921 | F_GETOWN F_SETOWN | |
922 | O_ASYNC O_DEFER O_DSYNC O_FSYNC O_SYNC | |
923 | O_EXLOCK O_SHLOCK | |
924 | ||
925 | These constants are intended for use with the Perl operators sysopen() | |
926 | and fcntl() and the basic database modules like SDBM_File. For the | |
927 | exact meaning of these and other Fcntl constants please refer to your | |
928 | operating system's documentation for fcntl() and open(). | |
929 | ||
930 | In addition, the Fcntl module now provides these constants for use | |
931 | with the Perl operator flock(): | |
932 | ||
933 | LOCK_SH LOCK_EX LOCK_NB LOCK_UN | |
934 | ||
935 | These constants are defined in all environments (because where there is | |
936 | no flock() system call, Perl emulates it). However, for historical | |
937 | reasons, these constants are not exported unless they are explicitly | |
938 | requested with the ":flock" tag (e.g. C<use Fcntl ':flock'>). | |
939 | ||
940 | =head2 IO | |
941 | ||
19799a22 | 942 | The IO module provides a simple mechanism to load all the IO modules at one |
6ee623d5 GS |
943 | go. Currently this includes: |
944 | ||
945 | IO::Handle | |
946 | IO::Seekable | |
947 | IO::File | |
948 | IO::Pipe | |
949 | IO::Socket | |
950 | ||
951 | For more information on any of these modules, please see its | |
952 | respective documentation. | |
953 | ||
954 | =head2 Math::Complex | |
955 | ||
956 | The Math::Complex module has been totally rewritten, and now supports | |
957 | more operations. These are overloaded: | |
958 | ||
959 | + - * / ** <=> neg ~ abs sqrt exp log sin cos atan2 "" (stringify) | |
960 | ||
961 | And these functions are now exported: | |
962 | ||
963 | pi i Re Im arg | |
964 | log10 logn ln cbrt root | |
965 | tan | |
966 | csc sec cot | |
967 | asin acos atan | |
968 | acsc asec acot | |
969 | sinh cosh tanh | |
970 | csch sech coth | |
971 | asinh acosh atanh | |
972 | acsch asech acoth | |
973 | cplx cplxe | |
974 | ||
975 | =head2 Math::Trig | |
976 | ||
977 | This new module provides a simpler interface to parts of Math::Complex for | |
978 | those who need trigonometric functions only for real numbers. | |
979 | ||
980 | =head2 DB_File | |
981 | ||
982 | There have been quite a few changes made to DB_File. Here are a few of | |
983 | the highlights: | |
984 | ||
13a2d996 | 985 | =over 4 |
6ee623d5 GS |
986 | |
987 | =item * | |
988 | ||
989 | Fixed a handful of bugs. | |
990 | ||
991 | =item * | |
992 | ||
993 | By public demand, added support for the standard hash function exists(). | |
994 | ||
995 | =item * | |
996 | ||
997 | Made it compatible with Berkeley DB 1.86. | |
998 | ||
999 | =item * | |
1000 | ||
1001 | Made negative subscripts work with RECNO interface. | |
1002 | ||
1003 | =item * | |
1004 | ||
1005 | Changed the default flags from O_RDWR to O_CREAT|O_RDWR and the default | |
1006 | mode from 0640 to 0666. | |
1007 | ||
1008 | =item * | |
1009 | ||
1010 | Made DB_File automatically import the open() constants (O_RDWR, | |
1011 | O_CREAT etc.) from Fcntl, if available. | |
1012 | ||
1013 | =item * | |
1014 | ||
1015 | Updated documentation. | |
1016 | ||
1017 | =back | |
1018 | ||
1019 | Refer to the HISTORY section in DB_File.pm for a complete list of | |
1020 | changes. Everything after DB_File 1.01 has been added since 5.003. | |
1021 | ||
1022 | =head2 Net::Ping | |
1023 | ||
1024 | Major rewrite - support added for both udp echo and real icmp pings. | |
1025 | ||
1026 | =head2 Object-oriented overrides for builtin operators | |
1027 | ||
1028 | Many of the Perl builtins returning lists now have | |
1029 | object-oriented overrides. These are: | |
1030 | ||
1031 | File::stat | |
1032 | Net::hostent | |
1033 | Net::netent | |
1034 | Net::protoent | |
1035 | Net::servent | |
1036 | Time::gmtime | |
1037 | Time::localtime | |
1038 | User::grent | |
1039 | User::pwent | |
1040 | ||
1041 | For example, you can now say | |
1042 | ||
1043 | use File::stat; | |
1044 | use User::pwent; | |
1045 | $his = (stat($filename)->st_uid == pwent($whoever)->pw_uid); | |
1046 | ||
1047 | =head1 Utility Changes | |
1048 | ||
1049 | =head2 pod2html | |
1050 | ||
13a2d996 | 1051 | =over 4 |
6ee623d5 GS |
1052 | |
1053 | =item Sends converted HTML to standard output | |
1054 | ||
1055 | The I<pod2html> utility included with Perl 5.004 is entirely new. | |
1056 | By default, it sends the converted HTML to its standard output, | |
1057 | instead of writing it to a file like Perl 5.003's I<pod2html> did. | |
1058 | Use the B<--outfile=FILENAME> option to write to a file. | |
1059 | ||
1060 | =back | |
1061 | ||
1062 | =head2 xsubpp | |
1063 | ||
13a2d996 | 1064 | =over 4 |
6ee623d5 GS |
1065 | |
1066 | =item C<void> XSUBs now default to returning nothing | |
1067 | ||
1068 | Due to a documentation/implementation bug in previous versions of | |
1069 | Perl, XSUBs with a return type of C<void> have actually been | |
1070 | returning one value. Usually that value was the GV for the XSUB, | |
1071 | but sometimes it was some already freed or reused value, which would | |
1072 | sometimes lead to program failure. | |
1073 | ||
1074 | In Perl 5.004, if an XSUB is declared as returning C<void>, it | |
1075 | actually returns no value, i.e. an empty list (though there is a | |
1076 | backward-compatibility exception; see below). If your XSUB really | |
1077 | does return an SV, you should give it a return type of C<SV *>. | |
1078 | ||
1079 | For backward compatibility, I<xsubpp> tries to guess whether a | |
1080 | C<void> XSUB is really C<void> or if it wants to return an C<SV *>. | |
1081 | It does so by examining the text of the XSUB: if I<xsubpp> finds | |
1082 | what looks like an assignment to C<ST(0)>, it assumes that the | |
1083 | XSUB's return type is really C<SV *>. | |
1084 | ||
1085 | =back | |
1086 | ||
1087 | =head1 C Language API Changes | |
1088 | ||
13a2d996 | 1089 | =over 4 |
6ee623d5 GS |
1090 | |
1091 | =item C<gv_fetchmethod> and C<perl_call_sv> | |
1092 | ||
1093 | The C<gv_fetchmethod> function finds a method for an object, just like | |
1094 | in Perl 5.003. The GV it returns may be a method cache entry. | |
1095 | However, in Perl 5.004, method cache entries are not visible to users; | |
1096 | therefore, they can no longer be passed directly to C<perl_call_sv>. | |
1097 | Instead, you should use the C<GvCV> macro on the GV to extract its CV, | |
1098 | and pass the CV to C<perl_call_sv>. | |
1099 | ||
1100 | The most likely symptom of passing the result of C<gv_fetchmethod> to | |
1101 | C<perl_call_sv> is Perl's producing an "Undefined subroutine called" | |
1102 | error on the I<second> call to a given method (since there is no cache | |
1103 | on the first call). | |
1104 | ||
1105 | =item C<perl_eval_pv> | |
1106 | ||
1107 | A new function handy for eval'ing strings of Perl code inside C code. | |
1108 | This function returns the value from the eval statement, which can | |
1109 | be used instead of fetching globals from the symbol table. See | |
1110 | L<perlguts>, L<perlembed> and L<perlcall> for details and examples. | |
1111 | ||
1112 | =item Extended API for manipulating hashes | |
1113 | ||
1114 | Internal handling of hash keys has changed. The old hashtable API is | |
1115 | still fully supported, and will likely remain so. The additions to the | |
1116 | API allow passing keys as C<SV*>s, so that C<tied> hashes can be given | |
1117 | real scalars as keys rather than plain strings (nontied hashes still | |
1118 | can only use strings as keys). New extensions must use the new hash | |
1119 | access functions and macros if they wish to use C<SV*> keys. These | |
1120 | additions also make it feasible to manipulate C<HE*>s (hash entries), | |
1121 | which can be more efficient. See L<perlguts> for details. | |
1122 | ||
1123 | =back | |
1124 | ||
1125 | =head1 Documentation Changes | |
1126 | ||
1127 | Many of the base and library pods were updated. These | |
1128 | new pods are included in section 1: | |
1129 | ||
13a2d996 | 1130 | =over 4 |
6ee623d5 GS |
1131 | |
1132 | =item L<perldelta> | |
1133 | ||
1134 | This document. | |
1135 | ||
1136 | =item L<perlfaq> | |
1137 | ||
1138 | Frequently asked questions. | |
1139 | ||
1140 | =item L<perllocale> | |
1141 | ||
1142 | Locale support (internationalization and localization). | |
1143 | ||
1144 | =item L<perltoot> | |
1145 | ||
1146 | Tutorial on Perl OO programming. | |
1147 | ||
1148 | =item L<perlapio> | |
1149 | ||
1150 | Perl internal IO abstraction interface. | |
1151 | ||
1152 | =item L<perlmodlib> | |
1153 | ||
1154 | Perl module library and recommended practice for module creation. | |
1155 | Extracted from L<perlmod> (which is much smaller as a result). | |
1156 | ||
1157 | =item L<perldebug> | |
1158 | ||
1159 | Although not new, this has been massively updated. | |
1160 | ||
1161 | =item L<perlsec> | |
1162 | ||
1163 | Although not new, this has been massively updated. | |
1164 | ||
1165 | =back | |
1166 | ||
1167 | =head1 New Diagnostics | |
1168 | ||
1169 | Several new conditions will trigger warnings that were | |
1170 | silent before. Some only affect certain platforms. | |
1171 | The following new warnings and errors outline these. | |
1172 | These messages are classified as follows (listed in | |
1173 | increasing order of desperation): | |
1174 | ||
1175 | (W) A warning (optional). | |
1176 | (D) A deprecation (optional). | |
1177 | (S) A severe warning (mandatory). | |
1178 | (F) A fatal error (trappable). | |
1179 | (P) An internal error you should never see (trappable). | |
1180 | (X) A very fatal error (nontrappable). | |
1181 | (A) An alien error message (not generated by Perl). | |
1182 | ||
13a2d996 | 1183 | =over 4 |
6ee623d5 GS |
1184 | |
1185 | =item "my" variable %s masks earlier declaration in same scope | |
1186 | ||
1187 | (W) A lexical variable has been redeclared in the same scope, effectively | |
1188 | eliminating all access to the previous instance. This is almost always | |
1189 | a typographical error. Note that the earlier variable will still exist | |
1190 | until the end of the scope or until all closure referents to it are | |
1191 | destroyed. | |
1192 | ||
1193 | =item %s argument is not a HASH element or slice | |
1194 | ||
1195 | (F) The argument to delete() must be either a hash element, such as | |
1196 | ||
1197 | $foo{$bar} | |
1198 | $ref->[12]->{"susie"} | |
1199 | ||
1200 | or a hash slice, such as | |
1201 | ||
1202 | @foo{$bar, $baz, $xyzzy} | |
1203 | @{$ref->[12]}{"susie", "queue"} | |
1204 | ||
1205 | =item Allocation too large: %lx | |
1206 | ||
1207 | (X) You can't allocate more than 64K on an MS-DOS machine. | |
1208 | ||
1209 | =item Allocation too large | |
1210 | ||
1211 | (F) You can't allocate more than 2^31+"small amount" bytes. | |
1212 | ||
1213 | =item Applying %s to %s will act on scalar(%s) | |
1214 | ||
1215 | (W) The pattern match (//), substitution (s///), and transliteration (tr///) | |
1216 | operators work on scalar values. If you apply one of them to an array | |
1217 | or a hash, it will convert the array or hash to a scalar value -- the | |
1218 | length of an array, or the population info of a hash -- and then work on | |
1219 | that scalar value. This is probably not what you meant to do. See | |
1220 | L<perlfunc/grep> and L<perlfunc/map> for alternatives. | |
1221 | ||
1222 | =item Attempt to free nonexistent shared string | |
1223 | ||
1224 | (P) Perl maintains a reference counted internal table of strings to | |
1225 | optimize the storage and access of hash keys and other strings. This | |
1226 | indicates someone tried to decrement the reference count of a string | |
1227 | that can no longer be found in the table. | |
1228 | ||
1229 | =item Attempt to use reference as lvalue in substr | |
1230 | ||
1231 | (W) You supplied a reference as the first argument to substr() used | |
1232 | as an lvalue, which is pretty strange. Perhaps you forgot to | |
1233 | dereference it first. See L<perlfunc/substr>. | |
1234 | ||
1235 | =item Bareword "%s" refers to nonexistent package | |
1236 | ||
1237 | (W) You used a qualified bareword of the form C<Foo::>, but | |
1238 | the compiler saw no other uses of that namespace before that point. | |
1239 | Perhaps you need to predeclare a package? | |
1240 | ||
1241 | =item Can't redefine active sort subroutine %s | |
1242 | ||
1243 | (F) Perl optimizes the internal handling of sort subroutines and keeps | |
1244 | pointers into them. You tried to redefine one such sort subroutine when it | |
1245 | was currently active, which is not allowed. If you really want to do | |
1246 | this, you should write C<sort { &func } @x> instead of C<sort func @x>. | |
1247 | ||
1248 | =item Can't use bareword ("%s") as %s ref while "strict refs" in use | |
1249 | ||
1250 | (F) Only hard references are allowed by "strict refs". Symbolic references | |
1251 | are disallowed. See L<perlref>. | |
1252 | ||
1253 | =item Cannot resolve method `%s' overloading `%s' in package `%s' | |
1254 | ||
1255 | (P) Internal error trying to resolve overloading specified by a method | |
1256 | name (as opposed to a subroutine reference). | |
1257 | ||
1258 | =item Constant subroutine %s redefined | |
1259 | ||
1260 | (S) You redefined a subroutine which had previously been eligible for | |
1261 | inlining. See L<perlsub/"Constant Functions"> for commentary and | |
1262 | workarounds. | |
1263 | ||
1264 | =item Constant subroutine %s undefined | |
1265 | ||
1266 | (S) You undefined a subroutine which had previously been eligible for | |
1267 | inlining. See L<perlsub/"Constant Functions"> for commentary and | |
1268 | workarounds. | |
1269 | ||
1270 | =item Copy method did not return a reference | |
1271 | ||
1272 | (F) The method which overloads "=" is buggy. See L<overload/Copy Constructor>. | |
1273 | ||
1274 | =item Died | |
1275 | ||
1276 | (F) You passed die() an empty string (the equivalent of C<die "">) or | |
1277 | you called it with no args and both C<$@> and C<$_> were empty. | |
1278 | ||
1279 | =item Exiting pseudo-block via %s | |
1280 | ||
1281 | (W) You are exiting a rather special block construct (like a sort block or | |
1282 | subroutine) by unconventional means, such as a goto, or a loop control | |
1283 | statement. See L<perlfunc/sort>. | |
1284 | ||
1285 | =item Identifier too long | |
1286 | ||
1287 | (F) Perl limits identifiers (names for variables, functions, etc.) to | |
1288 | 252 characters for simple names, somewhat more for compound names (like | |
1289 | C<$A::B>). You've exceeded Perl's limits. Future versions of Perl are | |
1290 | likely to eliminate these arbitrary limitations. | |
1291 | ||
1292 | =item Illegal character %s (carriage return) | |
1293 | ||
1294 | (F) A carriage return character was found in the input. This is an | |
1295 | error, and not a warning, because carriage return characters can break | |
c47ff5f1 | 1296 | multi-line strings, including here documents (e.g., C<print <<EOF;>). |
6ee623d5 GS |
1297 | |
1298 | =item Illegal switch in PERL5OPT: %s | |
1299 | ||
1300 | (X) The PERL5OPT environment variable may only be used to set the | |
1301 | following switches: B<-[DIMUdmw]>. | |
1302 | ||
1303 | =item Integer overflow in hex number | |
1304 | ||
1305 | (S) The literal hex number you have specified is too big for your | |
1306 | architecture. On a 32-bit architecture the largest hex literal is | |
1307 | 0xFFFFFFFF. | |
1308 | ||
1309 | =item Integer overflow in octal number | |
1310 | ||
1311 | (S) The literal octal number you have specified is too big for your | |
1312 | architecture. On a 32-bit architecture the largest octal literal is | |
1313 | 037777777777. | |
1314 | ||
1315 | =item internal error: glob failed | |
1316 | ||
1317 | (P) Something went wrong with the external program(s) used for C<glob> | |
c47ff5f1 | 1318 | and C<< <*.c> >>. This may mean that your csh (C shell) is |
6ee623d5 GS |
1319 | broken. If so, you should change all of the csh-related variables in |
1320 | config.sh: If you have tcsh, make the variables refer to it as if it | |
1321 | were csh (e.g. C<full_csh='/usr/bin/tcsh'>); otherwise, make them all | |
1322 | empty (except that C<d_csh> should be C<'undef'>) so that Perl will | |
1323 | think csh is missing. In either case, after editing config.sh, run | |
1324 | C<./Configure -S> and rebuild Perl. | |
1325 | ||
1326 | =item Invalid conversion in %s: "%s" | |
1327 | ||
1328 | (W) Perl does not understand the given format conversion. | |
1329 | See L<perlfunc/sprintf>. | |
1330 | ||
1331 | =item Invalid type in pack: '%s' | |
1332 | ||
1333 | (F) The given character is not a valid pack type. See L<perlfunc/pack>. | |
1334 | ||
1335 | =item Invalid type in unpack: '%s' | |
1336 | ||
1337 | (F) The given character is not a valid unpack type. See L<perlfunc/unpack>. | |
1338 | ||
1339 | =item Name "%s::%s" used only once: possible typo | |
1340 | ||
1341 | (W) Typographical errors often show up as unique variable names. | |
1342 | If you had a good reason for having a unique name, then just mention | |
1343 | it again somehow to suppress the message (the C<use vars> pragma is | |
1344 | provided for just this purpose). | |
1345 | ||
1346 | =item Null picture in formline | |
1347 | ||
1348 | (F) The first argument to formline must be a valid format picture | |
1349 | specification. It was found to be empty, which probably means you | |
1350 | supplied it an uninitialized value. See L<perlform>. | |
1351 | ||
1352 | =item Offset outside string | |
1353 | ||
1354 | (F) You tried to do a read/write/send/recv operation with an offset | |
1355 | pointing outside the buffer. This is difficult to imagine. | |
1356 | The sole exception to this is that C<sysread()>ing past the buffer | |
1357 | will extend the buffer and zero pad the new area. | |
1358 | ||
1359 | =item Out of memory! | |
1360 | ||
1361 | (X|F) The malloc() function returned 0, indicating there was insufficient | |
1362 | remaining memory (or virtual memory) to satisfy the request. | |
1363 | ||
1364 | The request was judged to be small, so the possibility to trap it | |
1365 | depends on the way Perl was compiled. By default it is not trappable. | |
1366 | However, if compiled for this, Perl may use the contents of C<$^M> as | |
1367 | an emergency pool after die()ing with this message. In this case the | |
1368 | error is trappable I<once>. | |
1369 | ||
1370 | =item Out of memory during request for %s | |
1371 | ||
1372 | (F) The malloc() function returned 0, indicating there was insufficient | |
1373 | remaining memory (or virtual memory) to satisfy the request. However, | |
1374 | the request was judged large enough (compile-time default is 64K), so | |
1375 | a possibility to shut down by trapping this error is granted. | |
1376 | ||
1377 | =item panic: frexp | |
1378 | ||
1379 | (P) The library function frexp() failed, making printf("%f") impossible. | |
1380 | ||
1381 | =item Possible attempt to put comments in qw() list | |
1382 | ||
1383 | (W) qw() lists contain items separated by whitespace; as with literal | |
1384 | strings, comment characters are not ignored, but are instead treated | |
1385 | as literal data. (You may have used different delimiters than the | |
7b8d334a | 1386 | parentheses shown here; braces are also frequently used.) |
6ee623d5 GS |
1387 | |
1388 | You probably wrote something like this: | |
1389 | ||
1390 | @list = qw( | |
1391 | a # a comment | |
1392 | b # another comment | |
1393 | ); | |
1394 | ||
1395 | when you should have written this: | |
1396 | ||
1397 | @list = qw( | |
1398 | a | |
1399 | b | |
1400 | ); | |
1401 | ||
1402 | If you really want comments, build your list the | |
1403 | old-fashioned way, with quotes and commas: | |
1404 | ||
1405 | @list = ( | |
1406 | 'a', # a comment | |
1407 | 'b', # another comment | |
1408 | ); | |
1409 | ||
1410 | =item Possible attempt to separate words with commas | |
1411 | ||
1412 | (W) qw() lists contain items separated by whitespace; therefore commas | |
1413 | aren't needed to separate the items. (You may have used different | |
1414 | delimiters than the parentheses shown here; braces are also frequently | |
1415 | used.) | |
1416 | ||
1417 | You probably wrote something like this: | |
1418 | ||
1419 | qw! a, b, c !; | |
1420 | ||
1421 | which puts literal commas into some of the list items. Write it without | |
1422 | commas if you don't want them to appear in your data: | |
1423 | ||
1424 | qw! a b c !; | |
1425 | ||
1426 | =item Scalar value @%s{%s} better written as $%s{%s} | |
1427 | ||
1428 | (W) You've used a hash slice (indicated by @) to select a single element of | |
1429 | a hash. Generally it's better to ask for a scalar value (indicated by $). | |
1430 | The difference is that C<$foo{&bar}> always behaves like a scalar, both when | |
1431 | assigning to it and when evaluating its argument, while C<@foo{&bar}> behaves | |
1432 | like a list when you assign to it, and provides a list context to its | |
1433 | subscript, which can do weird things if you're expecting only one subscript. | |
1434 | ||
551e1d92 | 1435 | =item Stub found while resolving method `%s' overloading `%s' in %s |
6ee623d5 GS |
1436 | |
1437 | (P) Overloading resolution over @ISA tree may be broken by importing stubs. | |
c2611fb3 | 1438 | Stubs should never be implicitly created, but explicit calls to C<can> |
6ee623d5 GS |
1439 | may break this. |
1440 | ||
1441 | =item Too late for "B<-T>" option | |
1442 | ||
1443 | (X) The #! line (or local equivalent) in a Perl script contains the | |
1444 | B<-T> option, but Perl was not invoked with B<-T> in its argument | |
1445 | list. This is an error because, by the time Perl discovers a B<-T> in | |
1446 | a script, it's too late to properly taint everything from the | |
1447 | environment. So Perl gives up. | |
1448 | ||
1449 | =item untie attempted while %d inner references still exist | |
1450 | ||
1451 | (W) A copy of the object returned from C<tie> (or C<tied>) was still | |
1452 | valid when C<untie> was called. | |
1453 | ||
1454 | =item Unrecognized character %s | |
1455 | ||
1456 | (F) The Perl parser has no idea what to do with the specified character | |
1457 | in your Perl script (or eval). Perhaps you tried to run a compressed | |
1458 | script, a binary program, or a directory as a Perl program. | |
1459 | ||
1460 | =item Unsupported function fork | |
1461 | ||
1462 | (F) Your version of executable does not support forking. | |
1463 | ||
1464 | Note that under some systems, like OS/2, there may be different flavors of | |
1465 | Perl executables, some of which may support fork, some not. Try changing | |
1466 | the name you call Perl by to C<perl_>, C<perl__>, and so on. | |
1467 | ||
1468 | =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated | |
1469 | ||
1470 | (D) Perl versions before 5.004 misinterpreted any type marker followed | |
1471 | by "$" and a digit. For example, "$$0" was incorrectly taken to mean | |
1472 | "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004. | |
1473 | ||
1474 | However, the developers of Perl 5.004 could not fix this bug completely, | |
1475 | because at least two widely-used modules depend on the old meaning of | |
1476 | "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the | |
1477 | old (broken) way inside strings; but it generates this message as a | |
1478 | warning. And in Perl 5.005, this special treatment will cease. | |
1479 | ||
1480 | =item Value of %s can be "0"; test with defined() | |
1481 | ||
1482 | (W) In a conditional expression, you used <HANDLE>, <*> (glob), C<each()>, | |
1483 | or C<readdir()> as a boolean value. Each of these constructs can return a | |
1484 | value of "0"; that would make the conditional expression false, which is | |
1485 | probably not what you intended. When using these constructs in conditional | |
1486 | expressions, test their values with the C<defined> operator. | |
1487 | ||
1488 | =item Variable "%s" may be unavailable | |
1489 | ||
1490 | (W) An inner (nested) I<anonymous> subroutine is inside a I<named> | |
1491 | subroutine, and outside that is another subroutine; and the anonymous | |
1492 | (innermost) subroutine is referencing a lexical variable defined in | |
1493 | the outermost subroutine. For example: | |
1494 | ||
1495 | sub outermost { my $a; sub middle { sub { $a } } } | |
1496 | ||
1497 | If the anonymous subroutine is called or referenced (directly or | |
1498 | indirectly) from the outermost subroutine, it will share the variable | |
1499 | as you would expect. But if the anonymous subroutine is called or | |
1500 | referenced when the outermost subroutine is not active, it will see | |
1501 | the value of the shared variable as it was before and during the | |
1502 | *first* call to the outermost subroutine, which is probably not what | |
1503 | you want. | |
1504 | ||
1505 | In these circumstances, it is usually best to make the middle | |
1506 | subroutine anonymous, using the C<sub {}> syntax. Perl has specific | |
1507 | support for shared variables in nested anonymous subroutines; a named | |
1508 | subroutine in between interferes with this feature. | |
1509 | ||
1510 | =item Variable "%s" will not stay shared | |
1511 | ||
1512 | (W) An inner (nested) I<named> subroutine is referencing a lexical | |
1513 | variable defined in an outer subroutine. | |
1514 | ||
1515 | When the inner subroutine is called, it will probably see the value of | |
1516 | the outer subroutine's variable as it was before and during the | |
1517 | *first* call to the outer subroutine; in this case, after the first | |
1518 | call to the outer subroutine is complete, the inner and outer | |
1519 | subroutines will no longer share a common value for the variable. In | |
1520 | other words, the variable will no longer be shared. | |
1521 | ||
1522 | Furthermore, if the outer subroutine is anonymous and references a | |
1523 | lexical variable outside itself, then the outer and inner subroutines | |
1524 | will I<never> share the given variable. | |
1525 | ||
1526 | This problem can usually be solved by making the inner subroutine | |
1527 | anonymous, using the C<sub {}> syntax. When inner anonymous subs that | |
1528 | reference variables in outer subroutines are called or referenced, | |
1529 | they are automatically rebound to the current values of such | |
1530 | variables. | |
1531 | ||
1532 | =item Warning: something's wrong | |
1533 | ||
1534 | (W) You passed warn() an empty string (the equivalent of C<warn "">) or | |
1535 | you called it with no args and C<$_> was empty. | |
1536 | ||
1537 | =item Ill-formed logical name |%s| in prime_env_iter | |
1538 | ||
1539 | (W) A warning peculiar to VMS. A logical name was encountered when preparing | |
1540 | to iterate over %ENV which violates the syntactic rules governing logical | |
1541 | names. Since it cannot be translated normally, it is skipped, and will not | |
1542 | appear in %ENV. This may be a benign occurrence, as some software packages | |
1543 | might directly modify logical name tables and introduce nonstandard names, | |
1544 | or it may indicate that a logical name table has been corrupted. | |
1545 | ||
1546 | =item Got an error from DosAllocMem | |
1547 | ||
1548 | (P) An error peculiar to OS/2. Most probably you're using an obsolete | |
1549 | version of Perl, and this should not happen anyway. | |
1550 | ||
1551 | =item Malformed PERLLIB_PREFIX | |
1552 | ||
1553 | (F) An error peculiar to OS/2. PERLLIB_PREFIX should be of the form | |
1554 | ||
1555 | prefix1;prefix2 | |
1556 | ||
1557 | or | |
1558 | ||
1559 | prefix1 prefix2 | |
1560 | ||
1561 | with nonempty prefix1 and prefix2. If C<prefix1> is indeed a prefix | |
1562 | of a builtin library search path, prefix2 is substituted. The error | |
1563 | may appear if components are not found, or are too long. See | |
1564 | "PERLLIB_PREFIX" in F<README.os2>. | |
1565 | ||
1566 | =item PERL_SH_DIR too long | |
1567 | ||
1568 | (F) An error peculiar to OS/2. PERL_SH_DIR is the directory to find the | |
1569 | C<sh>-shell in. See "PERL_SH_DIR" in F<README.os2>. | |
1570 | ||
1571 | =item Process terminated by SIG%s | |
1572 | ||
1573 | (W) This is a standard message issued by OS/2 applications, while *nix | |
1574 | applications die in silence. It is considered a feature of the OS/2 | |
1575 | port. One can easily disable this by appropriate sighandlers, see | |
1576 | L<perlipc/"Signals">. See also "Process terminated by SIGTERM/SIGINT" | |
1577 | in F<README.os2>. | |
1578 | ||
1579 | =back | |
1580 | ||
1581 | =head1 BUGS | |
1582 | ||
1583 | If you find what you think is a bug, you might check the headers of | |
1584 | recently posted articles in the comp.lang.perl.misc newsgroup. | |
1585 | There may also be information at http://www.perl.com/perl/, the Perl | |
1586 | Home Page. | |
1587 | ||
1588 | If you believe you have an unreported bug, please run the B<perlbug> | |
1589 | program included with your release. Make sure you trim your bug down | |
1590 | to a tiny but sufficient test case. Your bug report, along with the | |
1591 | output of C<perl -V>, will be sent off to <F<perlbug@perl.com>> to be | |
1592 | analysed by the Perl porting team. | |
1593 | ||
1594 | =head1 SEE ALSO | |
1595 | ||
1596 | The F<Changes> file for exhaustive details on what changed. | |
1597 | ||
1598 | The F<INSTALL> file for how to build Perl. This file has been | |
1599 | significantly updated for 5.004, so even veteran users should | |
1600 | look through it. | |
1601 | ||
1602 | The F<README> file for general stuff. | |
1603 | ||
1604 | The F<Copying> file for copyright information. | |
1605 | ||
1606 | =head1 HISTORY | |
1607 | ||
1608 | Constructed by Tom Christiansen, grabbing material with permission | |
1609 | from innumerable contributors, with kibitzing by more than a few Perl | |
1610 | porters. | |
1611 | ||
1612 | Last update: Wed May 14 11:14:09 EDT 1997 |