Commit | Line | Data |
ba8251e8 |
1 | =head1 NAME |
2 | |
a5222a85 |
3 | perldelta - what's new for perl v5.6 (as of v5.005_62) |
ba8251e8 |
4 | |
5 | =head1 DESCRIPTION |
6 | |
f29c64d6 |
7 | This is an unsupported alpha release, meant for intrepid Perl developers |
8 | only. The included sources may not even build correctly on some platforms. |
9 | Subscribing to perl5-porters is the best way to monitor and contribute |
10 | to the progress of development releases (see www.perl.org for info). |
11 | |
ba8251e8 |
12 | This document describes differences between the 5.005 release and this one. |
13 | |
14 | =head1 Incompatible Changes |
15 | |
e02fdbd2 |
16 | =head2 Perl Source Incompatibilities |
17 | |
a5222a85 |
18 | Beware that any new warnings that have been added are B<not> considered |
19 | incompatible changes. |
20 | |
21 | Since all new warnings must be explicitly requested via the C<-w> |
22 | switch or the C<warnings> pragma, it is ultimately the programmer's |
23 | responsibility to ensure that warnings are enabled judiciously. |
e02fdbd2 |
24 | |
757edf6f |
25 | =over 4 |
26 | |
08cd8952 |
27 | =item Treatment of list slices of undef has changed |
28 | |
29 | When taking a slice of a literal list (as opposed to a slice of |
30 | an array or hash), Perl used to return an empty list if the |
31 | result happened to be composed of all undef values. |
32 | |
33 | The new behavior is to produce an empty list if (and only if) |
34 | the original list was empty. Consider the following example: |
35 | |
36 | @a = (1,undef,undef,2)[2,1,2]; |
37 | |
38 | The old behavior would have resulted in @a having no elements. |
39 | The new behavior ensures it has three undefined elements. |
40 | |
41 | Note in particular that the behavior of slices of the following |
42 | cases remains unchanged: |
43 | |
44 | @a = ()[1,2]; |
45 | @a = (getpwent)[7,0]; |
46 | @a = (anything_returning_empty_list())[2,1,2]; |
47 | @a = @b[2,1,2]; |
48 | @a = @c{'a','b','c'}; |
49 | |
50 | See L<perldata>. |
51 | |
757edf6f |
52 | =item Possibly changed pseudo-random number generator |
53 | |
54 | In 5.005_0x and earlier, perl's rand() function used the C library |
55 | rand(3) function. As of 5.005_52, Configure tests for drand48(), |
56 | random(), and rand() (in that order) and picks the first one it finds. |
57 | Perl programs that depend on reproducing a specific set of pseudo-random |
c35dd67d |
58 | numbers will now likely produce different output. You can use |
59 | C<sh Configure -Drandfunc=rand> to obtain the old behavior. |
757edf6f |
60 | |
a5222a85 |
61 | =item Hashing function for hash keys has changed |
62 | |
63 | Perl hashes are not order preserving. The apparently random order |
64 | encountered when iterating on the contents of a hash is determined |
65 | by the hashing algorithm used. To improve the distribution of lower |
66 | bits in the hashed value, the algorithm has changed slightly as of |
67 | 5.005_52. When iterating over hashes, this may yield a random order |
68 | that is B<different> from that of previous versions. |
69 | |
70 | =item C<undef> fails on read only values |
71 | |
72 | Using the C<undef> operator on a readonly value (such as $1) has |
73 | the same effect as assigning C<undef> to the readonly value--it |
74 | throws an exception. |
75 | |
76 | =item Close-on-exec bit may be set on pipe() handles |
77 | |
78 | On systems that support a close-on-exec flag on filehandles, the |
79 | flag will be set for any handles created by pipe(), if that is |
80 | warranted by the value of $^F that may be in effect. Earlier |
81 | versions neglected to set the flag for handles created with |
82 | pipe(). See L<perlfunc/pipe> and L<perlvar/$^F>. |
83 | |
84 | =item Writing C<"$$1"> to mean C<"${$}1"> is unsupported |
85 | |
86 | Perl 5.004 deprecated the interpretation of C<$$1> and |
87 | similar within interpolated strings to mean C<$$ . "1">, |
88 | but still allowed it. |
89 | |
90 | In Perl 5.6 and later, C<"$$1"> always means C<"${$1}">. |
91 | |
92 | =item values(%h) and C<\(%h)> operate on aliases to values, not copies |
93 | |
94 | each(), values() and hashes in a list context return the actual |
95 | values in the hash, instead of copies (as they used to in earlier |
96 | versions). Typical idioms for using these constructs copy the |
97 | returned values, but this is can make a significant difference when |
98 | creating references to the returned values. |
99 | |
100 | Keys in the hash are still returned as copies when iterating on |
08cd8952 |
101 | a hash. |
a5222a85 |
102 | |
103 | =item vec(EXPR,OFFSET,BITS) enforces powers-of-two BITS |
104 | |
08cd8952 |
105 | vec() generates a run-time error if the BITS argument is not |
a5222a85 |
106 | a valid power-of-two integer. |
107 | |
108 | =item Text of some diagnostic output has changed |
109 | |
110 | Most references to internal Perl operations in diagnostics |
111 | have been changed to be more descriptive. This may be an |
112 | issue for programs that may incorrectly rely on the exact |
113 | text of diagnostics for proper functioning. |
114 | |
115 | =item C<%@> has been removed |
116 | |
117 | The undocumented special variable C<%@> that used to accumulate |
118 | "background" errors (such as those that happen in DESTROY()) |
119 | has been removed, because it could potentially result in memory |
120 | leaks. |
121 | |
757edf6f |
122 | =back |
123 | |
e02fdbd2 |
124 | =head2 C Source Incompatibilities |
125 | |
126 | =over 4 |
127 | |
128 | =item C<PERL_POLLUTE> |
129 | |
130 | Release 5.005 grandfathered old global symbol names by providing preprocessor |
87275199 |
131 | macros for extension source compatibility. As of release 5.6, these |
e02fdbd2 |
132 | preprocessor definitions are not available by default. You need to explicitly |
14218588 |
133 | compile perl with C<-DPERL_POLLUTE> to get these definitions. For |
134 | extensions still using the old symbols, this option can be |
2aea4d40 |
135 | specified via MakeMaker: |
136 | |
14218588 |
137 | perl Makefile.PL POLLUTE=1 |
e02fdbd2 |
138 | |
f29c64d6 |
139 | =item C<PERL_IMPLICIT_CONTEXT> |
140 | |
141 | This new build option provides a set of macros for all API functions |
142 | such that an implicit interpreter/thread context argument is passed to |
143 | every API function. As a result of this, something like C<sv_setsv(foo,bar)> |
2c2d71f5 |
144 | amounts to a macro invocation that actually translates to something like |
f29c64d6 |
145 | C<Perl_sv_setsv(my_perl,foo,bar)>. While this is generally expected |
146 | to not have any significant source compatibility issues, the difference |
147 | between a macro and a real function call will need to be considered. |
148 | |
2c2d71f5 |
149 | This means that there B<is> a source compatibility issue as a result of |
150 | this if your extensions attempt to use pointers to any of the Perl API |
151 | functions. |
152 | |
f29c64d6 |
153 | Note that the above issue is not relevant to the default build of |
154 | Perl, whose interfaces continue to match those of prior versions |
155 | (but subject to the other options described here). |
156 | |
651a3225 |
157 | PERL_IMPLICIT_CONTEXT is automatically enabled whenever Perl is built |
158 | with one of -Dusethreads, -Dusemultiplicity, or both. |
f29c64d6 |
159 | |
2c2d71f5 |
160 | See L<perlguts/"The Perl API"> for detailed information on the |
161 | ramifications of building Perl using this option. |
162 | |
86058a2d |
163 | =item C<PERL_POLLUTE_MALLOC> |
164 | |
14218588 |
165 | Enabling Perl's malloc in release 5.005 and earlier caused |
86058a2d |
166 | the namespace of system versions of the malloc family of functions to |
14218588 |
167 | be usurped by the Perl versions, since by default they used the |
168 | same names. |
86058a2d |
169 | |
170 | Besides causing problems on platforms that do not allow these functions to |
171 | be cleanly replaced, this also meant that the system versions could not |
172 | be called in programs that used Perl's malloc. Previous versions of Perl |
14218588 |
173 | have allowed this behaviour to be suppressed with the HIDEMYMALLOC and |
86058a2d |
174 | EMBEDMYMALLOC preprocessor definitions. |
175 | |
87275199 |
176 | As of release 5.6, Perl's malloc family of functions have default names |
86058a2d |
177 | distinct from the system versions. You need to explicitly compile perl with |
14218588 |
178 | C<-DPERL_POLLUTE_MALLOC> to get the older behaviour. HIDEMYMALLOC |
179 | and EMBEDMYMALLOC have no effect, since the behaviour they enabled is now |
86058a2d |
180 | the default. |
181 | |
182 | Note that these functions do B<not> constitute Perl's memory allocation API. |
183 | See L<perlguts/"Memory Allocation"> for further information about that. |
184 | |
e02fdbd2 |
185 | =item C<PL_na> and C<dTHR> Issues |
186 | |
187 | The C<PL_na> global is now thread local, so a C<dTHR> declaration is needed |
14218588 |
188 | in the scope in which the global appears. XSUBs should handle this automatically, |
e02fdbd2 |
189 | but if you have used C<PL_na> in support functions, you either need to |
190 | change the C<PL_na> to a local variable (which is recommended), or put in |
191 | a C<dTHR>. |
192 | |
193 | =back |
194 | |
cceca5ed |
195 | =head2 Compatible C Source API Changes |
196 | |
197 | =over |
198 | |
199 | =item C<PATCHLEVEL> is now C<PERL_VERSION> |
200 | |
14218588 |
201 | The cpp macros C<PERL_REVISION>, C<PERL_VERSION>, and C<PERL_SUBVERSION> |
cceca5ed |
202 | are now available by default from perl.h, and reflect the base revision, |
14218588 |
203 | patchlevel, and subversion respectively. C<PERL_REVISION> had no |
cceca5ed |
204 | prior equivalent, while C<PERL_VERSION> and C<PERL_SUBVERSION> were |
205 | previously available as C<PATCHLEVEL> and C<SUBVERSION>. |
206 | |
14218588 |
207 | The new names cause less pollution of the B<cpp> namespace and reflect what |
cceca5ed |
208 | the numbers have come to stand for in common practice. For compatibility, |
14218588 |
209 | the old names are still supported when F<patchlevel.h> is explicitly |
cceca5ed |
210 | included (as required before), so there is no source incompatibility |
14218588 |
211 | from the change. |
cceca5ed |
212 | |
a5222a85 |
213 | =item Support for C++ exceptions |
214 | |
215 | change#3386, also needs perlguts documentation |
216 | [TODO - Chip Salzenberg <chip@perlsupport.com>] |
217 | |
cceca5ed |
218 | =back |
219 | |
e02fdbd2 |
220 | =head2 Binary Incompatibilities |
221 | |
9c107f78 |
222 | The default build of this release is binary compatible with the 5.005 |
223 | release or its maintenance versions. |
f29c64d6 |
224 | |
225 | The usethreads or usemultiplicity builds are B<not> binary compatible |
226 | with the corresponding builds in 5.005. |
e02fdbd2 |
227 | |
a5222a85 |
228 | =head1 Installation and Configuration Improvements |
229 | |
230 | =head2 New Configure flags |
231 | |
232 | The following new flags may be enabled on the Configure command line |
233 | by running Configure with C<-Dflag>. |
234 | |
235 | usemultiplicity |
67d3893f |
236 | |
237 | uselongdouble |
a5222a85 |
238 | usemorebits |
239 | uselargefiles |
a5222a85 |
240 | |
67d3893f |
241 | =head2 -Dusethreads and -Duse64bits now more daring |
242 | |
243 | The Configure options enabling the use of threads and the use of |
244 | 64-bitness are now more daring in the sense that they no more have |
245 | an explicit list of operating systems of known threads/64-bit |
246 | capabilities. In other words: if your operating system has the |
247 | necessary APIs, you should be able just to go ahead and use them. |
248 | See also L<"64-bit support">. |
249 | |
250 | =head2 Long Doubles |
251 | |
252 | Some platforms have "long doubles", floating point numbers of even |
437784d6 |
253 | larger range than ordinary "doubles". To enable using long doubles for |
67d3893f |
254 | Perl's scalars, use -Duselongdouble. |
255 | |
256 | =head2 -Dusemorebits |
257 | |
258 | You can enable both -Duse64bits and -Dlongdouble by -Dusemorebits. |
259 | See also L<"64-bit support">. |
260 | |
261 | =head2 -Duselargefiles |
262 | |
263 | Some platforms support large files, files larger than two gigabytes. |
264 | See L<"Large file support"> for more information. |
a5222a85 |
265 | |
266 | =head2 installusrbinperl |
267 | |
268 | You can use "Configure -Uinstallusrbinperl" which causes installperl |
269 | to skip installing perl also as /usr/bin/perl. This is useful if you |
270 | prefer not to modify /usr/bin for some reason or another but harmful |
271 | because many scripts assume to find Perl in /usr/bin/perl. |
272 | |
273 | =head2 SOCKS support |
274 | |
275 | You can use "Configure -Dusesocks" which causes Perl to probe |
c35dd67d |
276 | for the SOCKS (v5, not v4) proxy protocol library, |
277 | http://www.socks.nec.com/ |
a5222a85 |
278 | |
279 | =head2 C<-A> flag |
280 | |
281 | You can "post-edit" the Configure variables using the Configure C<-A> |
282 | flag. The editing happens immediately after the platform specific |
283 | hints files have been processed but before the actual configuration |
284 | process starts. Run C<Configure -h> to find out the full C<-A> syntax. |
285 | |
c35dd67d |
286 | =head2 Enhanced Installation Directories |
67d3893f |
287 | |
c35dd67d |
288 | The installation structure has been enriched to improve the support for |
289 | maintaining multiple versions of perl, to provide locations for |
290 | vendor-supplied modules and scripts, and to ease maintenance of |
291 | locally-added modules and scripts. See the section on Installation |
292 | Directories in the INSTALL file for complete details. For most users |
293 | building and installing from source, the defaults should be fine. |
67d3893f |
294 | |
ba8251e8 |
295 | =head1 Core Changes |
296 | |
9d73390d |
297 | =head2 Unicode and UTF-8 support |
298 | |
299 | Perl can optionally use UTF-8 as its internal representation for character |
a5222a85 |
300 | strings. The C<utf8> pragma enables this support in the current lexical |
9d73390d |
301 | scope. See L<utf8> for more information. |
302 | |
303 | =head2 Lexically scoped warning categories |
304 | |
305 | You can now control the granularity of warnings emitted by perl at a finer |
4438c4b7 |
306 | level using the C<use warnings> pragma. See L<warnings> and L<perllexwarn> |
0453d815 |
307 | for details. |
9d73390d |
308 | |
a5222a85 |
309 | =head2 Lvalue subroutines |
310 | |
311 | WARNING: This is an experimental feature. |
312 | |
313 | change#4081 |
314 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>, |
315 | Tuomas Lukka <lukka@fas.harvard.edu>)] |
316 | |
317 | =head2 "our" declarations |
318 | |
319 | An "our" declaration introduces a value that can be best understood |
320 | as a lexically scoped symbolic alias to a global variable in the |
321 | current package. This is mostly useful as an alternative to the |
322 | C<vars> pragma, but also provides the opportunity to introduce |
323 | typing and other attributes for such variables. See L<perlfunc/our>. |
324 | |
325 | =head2 Weak references |
326 | |
327 | WARNING: This is an experimental feature. |
328 | |
329 | change#3385, also need perlguts documentation |
330 | |
331 | [TODO - Tuomas Lukka <lukka@fas.harvard.edu>] |
332 | |
becf2bd3 |
333 | =head2 File globbing implemented internally |
334 | |
335 | WARNING: This is currently an experimental feature. Interfaces and |
336 | implementation are likely to change. |
337 | |
338 | Perl can be compiled with -DPERL_INTERNAL_GLOB to use the File::Glob |
339 | implementation of the glob() operator. This avoids using an external |
340 | csh process and the problems associated with it. |
341 | |
5fdc711f |
342 | =head2 Binary numbers supported |
343 | |
4f19785b |
344 | Binary numbers are now supported as literals, in s?printf formats, and |
345 | C<oct()>: |
346 | |
14218588 |
347 | $answer = 0b101010; |
348 | printf "The answer is: %b\n", oct("0b101010"); |
4f19785b |
349 | |
a5222a85 |
350 | =head2 Some arrows may be omitted in calls through references |
351 | |
352 | Perl now allows the arrow to be omitted in many constructs |
353 | involving subroutine calls through references. For example, |
354 | C<$foo[10]->('foo')> may now be written C<$foo[10]('foo')>. |
355 | This is rather similar to how the arrow may be omitted from |
356 | C<$foo[10]->{'foo'}>. Note however, that the arrow is still |
357 | required for C<foo(10)->('bar')>. |
358 | |
5fdc711f |
359 | =head2 syswrite() ease-of-use |
360 | |
a5222a85 |
361 | The length argument of C<syswrite()> has become optional. |
362 | |
363 | =head2 Filehandles can be autovivified |
364 | |
365 | The construct C<open(my $fh, ...)> can be used to create filehandles |
366 | more easily. The filehandle will be automatically closed at the end |
367 | of the scope of $fh, provided there are no other references to it. This |
368 | largely eliminates the need for typeglobs when opening filehandles |
369 | that must be passed around, as in the following example: |
370 | |
371 | sub myopen { |
372 | open my $fh, "@_" |
373 | or die "Can't open '@_': $!"; |
374 | return $fh; |
375 | } |
376 | |
377 | { |
378 | my $f = myopen("</etc/motd"); |
379 | print <$f>; |
380 | # $f implicitly closed here |
381 | } |
382 | |
383 | [TODO - this idiom needs more pod penetration] |
6c67e1bb |
384 | |
5fdc711f |
385 | =head2 64-bit support |
386 | |
9c107f78 |
387 | All platforms that have 64-bit integers either (a) natively as longs |
388 | or ints (b) via special compiler flags (c) using long long are able to |
389 | use "quads" (64-integers) as follows: |
390 | |
391 | =over 4 |
392 | |
a5222a85 |
393 | =item * |
394 | |
395 | constants (decimal, hexadecimal, octal, binary) in the code |
396 | |
397 | =item * |
9c107f78 |
398 | |
a5222a85 |
399 | arguments to oct() and hex() |
9c107f78 |
400 | |
a5222a85 |
401 | =item * |
402 | |
403 | arguments to print(), printf() and sprintf() (flag prefixes ll, L, q) |
404 | |
405 | =item * |
9c107f78 |
406 | |
a5222a85 |
407 | printed as such |
9c107f78 |
408 | |
a5222a85 |
409 | =item * |
410 | |
411 | pack() and unpack() "q" and "Q" formats |
412 | |
413 | =item * |
414 | |
415 | in basic arithmetics: + - * / % |
416 | |
417 | =item * |
1fad5d67 |
418 | |
a5222a85 |
419 | vec() (but see the below note about bit arithmetics) |
9c107f78 |
420 | |
421 | =back |
422 | |
423 | Note that unless you have the case (a) you will have to configure |
424 | and compile Perl using the -Duse64bits Configure flag. |
425 | |
3175b8cd |
426 | Unfortunately bit arithmetics (&, |, ^, ~, <<, >>) for numbers are not |
427 | 64-bit clean, they are explictly forced to be 32-bit. Bit arithmetics |
428 | for bit vectors (created by vec()) are not limited in their width. |
d0ba1bd2 |
429 | |
2d4389e4 |
430 | Last but not least: note that due to Perl's habit of always using |
d0ba1bd2 |
431 | floating point numbers the quads are still not true integers. |
432 | When quads overflow their limits (0...18_446_744_073_709_551_615 unsigned, |
433 | -9_223_372_036_854_775_808...9_223_372_036_854_775_807 signed), they |
434 | are silently promoted to floating point numbers, after which they will |
435 | start losing precision (their lower digits). |
2d4389e4 |
436 | |
437 | =head2 Large file support |
438 | |
439 | If you have filesystems that support "large files" (files larger than |
aa855319 |
440 | 2 gigabytes), you may now also be able to create and access them from |
249b38c6 |
441 | Perl. You have to use Configure -Duselargefiles. Turning on the |
442 | large file support turns on also the 64-bit support, for obvious reasons. |
2d4389e4 |
443 | |
eed7fde4 |
444 | Note that in addition to requiring a proper file system to do large |
445 | files you may also need to adjust your per-process (or your |
446 | per-system, or per-process-group, or per-user-group) maximum filesize |
447 | limits before running Perl scripts that try to handle large files, |
448 | especially if you intend to write such files. |
449 | |
450 | Finally, in addition to your process/process group maximum filesize |
451 | limits, you may have quota limits on your filesystems that stop you |
452 | (your user id or your user group id) from using large files. |
453 | |
454 | Adjusting your process/user/group/file system/operating system limits |
455 | is outside the scope of Perl core language. For process limits, you |
456 | may try increasing the limits using your shell's limits/limit/ulimit |
457 | command before running Perl. The BSD::Resource extension (not |
458 | included with the standard Perl distribution) may also be of use, it |
459 | offers the getrlimit/setrlimit interface that can be used to adjust |
460 | process resource usage limits, including the maximum filesize limit. |
2d4389e4 |
461 | |
aa855319 |
462 | =head2 Long doubles |
463 | |
464 | In some systems you may be able to use long doubles to enhance the |
465 | range of precision of your double precision floating point numbers |
466 | (that is, Perl's numbers). Use Configure -Duselongdouble to enable |
467 | this support (if it is available). |
468 | |
469 | =head2 "more bits" |
470 | |
471 | You can Configure -Dusemorebits to turn on both the 64-bit support |
472 | and the long double support. |
09bef843 |
473 | |
62c18ce2 |
474 | =head2 Better syntax checks on parenthesized unary operators |
475 | |
476 | Expressions such as: |
477 | |
14218588 |
478 | print defined(&foo,&bar,&baz); |
479 | print uc("foo","bar","baz"); |
480 | undef($foo,&bar); |
62c18ce2 |
481 | |
7711098a |
482 | used to be accidentally allowed in earlier versions, and produced |
14218588 |
483 | unpredictable behaviour. Some produced ancillary warnings |
484 | when used in this way; others silently did the wrong thing. |
62c18ce2 |
485 | |
486 | The parenthesized forms of most unary operators that expect a single |
14218588 |
487 | argument now ensure that they are not called with more than one |
488 | argument, making the cases shown above syntax errors. The usual |
489 | behaviour of: |
62c18ce2 |
490 | |
14218588 |
491 | print defined &foo, &bar, &baz; |
492 | print uc "foo", "bar", "baz"; |
493 | undef $foo, &bar; |
62c18ce2 |
494 | |
495 | remains unchanged. See L<perlop>. |
496 | |
3e3318e7 |
497 | =head2 POSIX character class syntax [: :] supported |
498 | |
499 | For example to match alphabetic characters use /[[:alpha:]]/. |
500 | See L<perlre> for details. |
501 | |
5a929a98 |
502 | =head2 Improved C<qw//> operator |
8127e0e3 |
503 | |
26ef7447 |
504 | The C<qw//> operator is now evaluated at compile time into a true list |
505 | instead of being replaced with a run time call to C<split()>. This |
14218588 |
506 | removes the confusing misbehaviour of C<qw//> in scalar context, which |
507 | had inherited that behaviour from split(). |
26ef7447 |
508 | |
509 | Thus: |
510 | |
511 | $foo = ($bar) = qw(a b c); print "$foo|$bar\n"; |
512 | |
513 | now correctly prints "3|a", instead of "2|a". |
8127e0e3 |
514 | |
5a929a98 |
515 | =head2 pack() format 'Z' supported |
516 | |
517 | The new format type 'Z' is useful for packing and unpacking null-terminated |
518 | strings. See L<perlfunc/"pack">. |
519 | |
4d0c1c44 |
520 | =head2 pack() format modifier '!' supported |
ee3907e2 |
521 | |
14218588 |
522 | The new format type modifier '!' is useful for packing and unpacking |
ee3907e2 |
523 | native shorts, ints, and longs. See L<perlfunc/"pack">. |
524 | |
f29c64d6 |
525 | =head2 pack() and unpack() support counted strings |
526 | |
a5222a85 |
527 | The template character '/' can be used to specify a counted string |
f29c64d6 |
528 | type to be packed or unpacked. See L<perlfunc/"pack">. |
529 | |
a5222a85 |
530 | =head2 Comments in pack() templates |
531 | |
532 | The '#' character in a template introduces a comment up to |
533 | end of the line. This facilitates documentation of pack() |
534 | templates. |
535 | |
2b92dfce |
536 | =head2 $^X variables may now have names longer than one character |
537 | |
538 | Formerly, $^X was synonymous with ${"\cX"}, but $^XY was a syntax |
539 | error. Now variable names that begin with a control character may be |
540 | arbitrarily long. However, for compatibility reasons, these variables |
541 | I<must> be written with explicit braces, as C<${^XY}> for example. |
14218588 |
542 | C<${^XYZ}> is synonymous with ${"\cXYZ"}. Variable names with more |
2b92dfce |
543 | than one control character, such as C<${^XY^Z}>, are illegal. |
544 | |
14218588 |
545 | The old syntax has not changed. As before, `^X' may be either a |
546 | literal control-X character or the two-character sequence `caret' plus |
547 | `X'. When braces are omitted, the variable name stops after the |
2b92dfce |
548 | control character. Thus C<"$^XYZ"> continues to be synonymous with |
7711098a |
549 | C<$^X . "YZ"> as before. |
2b92dfce |
550 | |
551 | As before, lexical variables may not have names beginning with control |
552 | characters. As before, variables whose names begin with a control |
14218588 |
553 | character are always forced to be in package `main'. All such variables |
554 | are reserved for future extensions, except those that begin with |
09bef843 |
555 | C<^_>, which may be used by user programs and are guaranteed not to |
14218588 |
556 | acquire special meaning in any future version of Perl. |
2b92dfce |
557 | |
09bef843 |
558 | =head2 C<use attrs> implicit in subroutine attributes |
559 | |
560 | Formerly, if you wanted to mark a subroutine as being a method call or |
561 | as requiring an automatic lock() when it is entered, you had to declare |
562 | that with a C<use attrs> pragma in the body of the subroutine. |
563 | That can now be accomplished with a declaration syntax, like this: |
564 | |
565 | sub mymethod : locked, method ; |
566 | ... |
567 | sub mymethod : locked, method { |
568 | ... |
569 | } |
570 | |
571 | F<AutoSplit.pm> and F<SelfLoader.pm> have been updated to keep the attributes |
572 | with the stubs they provide. See L<attributes>. |
573 | |
a5222a85 |
574 | =head2 Regular expression improvements |
575 | |
576 | change#2827,2373,2372,2365,1813,1800,4112,4158,4215,4301 |
577 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
578 | |
579 | =head2 Overloading improvements |
580 | |
581 | change#2150 |
582 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
583 | |
584 | =head2 open() with more than two arguments |
585 | |
586 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
587 | |
588 | =head2 Support for interpolating named characters |
589 | |
590 | change#4052 |
591 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
592 | |
08cd8952 |
593 | =head2 Experimental support for user-hooks in @INC |
a5222a85 |
594 | |
595 | [TODO - Ken Fox <kfox@ford.com>] |
596 | |
597 | =head2 C<require> and C<do> may be overridden |
598 | |
599 | C<require> and C<do 'file'> operations may be overridden locally |
600 | by importing subroutines of the same name into the current package |
601 | (or globally by importing them into the CORE::GLOBAL:: namespace). |
602 | Overriding C<require> will also affect C<use>, provided the override |
603 | is visible at compile-time. |
604 | See L<perlsub/"Overriding Built-in Functions">. |
605 | |
606 | =head2 New variable $^C reflects C<-c> switch |
607 | |
08cd8952 |
608 | C<$^C> has a boolean value that reflects whether perl is being run |
a5222a85 |
609 | in compile-only mode (i.e. via the C<-c> switch). Since |
610 | BEGIN blocks are executed under such conditions, this variable |
611 | enables perl code to determine whether actions that make sense |
612 | only during normal running are warranted. See L<perlvar>. |
613 | |
614 | =head2 Optional Y2K warnings |
615 | |
616 | If Perl is built with the cpp macro C<PERL_Y2KWARN> defined, |
617 | it emits optional warnings when concatenating the number 19 |
618 | with another number. |
619 | |
620 | This behavior must be specifically enabled when running Configure. |
621 | See L<INSTALL> and L<README.Y2K>. |
622 | |
fbad3eb5 |
623 | =head1 Significant bug fixes |
624 | |
625 | =head2 E<lt>HANDLEE<gt> on empty files |
626 | |
627 | With C<$/> set to C<undef>, slurping an empty file returns a string of |
14218588 |
628 | zero length (instead of C<undef>, as it used to) the first time the |
629 | HANDLE is read. Further reads yield C<undef>. |
fbad3eb5 |
630 | |
631 | This means that the following will append "foo" to an empty file (it used |
14218588 |
632 | to do nothing): |
fbad3eb5 |
633 | |
634 | perl -0777 -pi -e 's/^/foo/' empty_file |
635 | |
14218588 |
636 | The behaviour of: |
fbad3eb5 |
637 | |
638 | perl -pi -e 's/^/foo/' empty_file |
639 | |
640 | is unchanged (it continues to leave the file empty). |
641 | |
0244c3a4 |
642 | =head2 C<eval '...'> improvements |
643 | |
644 | Line numbers (as reflected by caller() and most diagnostics) within |
645 | C<eval '...'> were often incorrect when here documents were involved. |
646 | This has been corrected. |
647 | |
648 | Lexical lookups for variables appearing in C<eval '...'> within |
649 | functions that were themselves called within an C<eval '...'> were |
14218588 |
650 | searching the wrong place for lexicals. The lexical search now |
651 | correctly ends at the subroutine's block boundary. |
0244c3a4 |
652 | |
653 | Parsing of here documents used to be flawed when they appeared as |
654 | the replacement expression in C<eval 's/.../.../e'>. This has |
655 | been fixed. |
656 | |
a5222a85 |
657 | =head2 All compilation errors are true errors |
658 | |
659 | Some "errors" encountered at compile time were by neccessity |
660 | generated as warnings followed by eventual termination of the |
661 | program. This enabled more such errors to be reported in a |
662 | single run, rather than causing a hard stop at the first error |
663 | that was encountered. |
664 | |
665 | The mechanism for reporting such errors has been reimplemented |
666 | to queue compile-time errors and report them at the end of the |
667 | compilation as true errors rather than as warnings. This fixes |
08cd8952 |
668 | cases where error messages leaked through in the form of warnings |
669 | when code was compiled at run time using C<eval STRING>, and |
670 | also allows such errors to be reliably trapped using __DIE__ hooks. |
a5222a85 |
671 | |
45bc9206 |
672 | =head2 Automatic flushing of output buffers |
673 | |
14218588 |
674 | fork(), exec(), system(), qx//, and pipe open()s now flush buffers |
675 | of all files opened for output when the operation |
676 | was attempted. This mostly eliminates confusing |
45bc9206 |
677 | buffering mishaps suffered by users unaware of how Perl internally |
14218588 |
678 | handles I/O. |
45bc9206 |
679 | |
af8c498a |
680 | =head2 Better diagnostics on meaningless filehandle operations |
681 | |
682 | Constructs such as C<open(E<lt>FHE<gt>)> and C<close(E<lt>FHE<gt>)> |
683 | are compile time errors. Attempting to read from filehandles that |
684 | were opened only for writing will now produce warnings (just as |
685 | writing to read-only filehandles does). |
686 | |
a5222a85 |
687 | =head2 Where possible, buffered data discarded from duped input filehandle |
688 | |
689 | C<open(NEW, "E<lt>&OLD")> now attempts to discard any data that |
690 | was previously read and buffered in C<OLD> before duping the handle. |
691 | On platforms where doing this is allowed, the next read operation |
692 | on C<NEW> will return the same data as the corresponding operation |
693 | on C<OLD>. Formerly, it would have returned the data from the start |
694 | of the following disk block instead. |
695 | |
696 | =head2 system(), backticks and pipe open now reflect exec() failure |
697 | |
698 | On Unix and similar platforms, system(), qx() and open(FOO, "cmd |") |
699 | etc., are implemented via fork() and exec(). When the underlying |
700 | exec() fails, earlier versions did not report the error properly, |
701 | since the exec() happened to be in a different process. |
702 | |
703 | The child process now communicates with the parent about the |
437784d6 |
704 | error in launching the external command, which allows these |
a5222a85 |
705 | constructs to return with their usual error value and set $!. |
706 | |
707 | =head2 Implicitly closed filehandles are safer |
708 | |
709 | Sometimes implicitly closed filehandles (as when they are localized, |
710 | and Perl automatically closes them on exiting the scope) could |
711 | inadvertently set $? or $!. This has been corrected. |
712 | |
713 | =head2 C<(\$)> prototype and C<$foo{a}> |
714 | |
715 | An scalar reference prototype now correctly allows a hash or |
716 | array element in that slot. |
717 | |
718 | =head2 Pseudo-hashes work better |
719 | |
720 | Dereferencing some types of reference values in a pseudo-hash, |
721 | such as C<$ph->{foo}[1]>, was accidentally disallowed. This has |
722 | been corrected. |
723 | |
724 | When applied to a pseudo-hash element, exists() now reports whether |
725 | the specified value exists, not merely if the key is valid. |
726 | |
727 | =head2 C<goto &sub> and AUTOLOAD |
728 | |
08cd8952 |
729 | The C<goto &sub> construct works correctly when C<&sub> happens |
a5222a85 |
730 | to be autoloaded. |
731 | |
732 | =head2 C<-bareword> allowed under C<use integer> |
733 | |
734 | The autoquoting of barewords preceded by C<-> did not work |
735 | in prior versions when the C<integer> pragma was enabled. |
736 | This has been fixed. |
737 | |
738 | =head2 Boolean assignment operators are legal lvalues |
739 | |
740 | Constructs such as C<($a ||= 2) += 1> are now allowed. |
741 | |
742 | =head2 C<sort $coderef @foo> allowed |
743 | |
744 | sort() did not accept a subroutine reference as the comparison |
08cd8952 |
745 | function in earlier versions. This is now permitted. |
a5222a85 |
746 | |
747 | =head2 Failures in DESTROY() |
748 | |
749 | When code in a destructor threw an exception, it went unnoticed |
750 | in earlier versions of Perl, unless someone happened to be |
751 | looking in $@ just after the point the destructor happened to |
752 | run. Such failures are now visible as warnings when warnings are |
753 | enabled. |
754 | |
755 | =head2 Locale bugs fixed |
54195c32 |
756 | |
437784d6 |
757 | printf() and sprintf() previously reset the numeric locale |
67d3893f |
758 | back to the default "C" locale. This has been fixed. |
759 | |
760 | Numbers formatted according to the local numeric locale |
761 | (such as using a decimal comma instead of a decimal dot) caused |
762 | "isn't numeric" warnings, even while the operations accessing |
763 | those numbers produced correct results. The warnings are gone. |
54195c32 |
764 | |
a5222a85 |
765 | =head2 Memory leaks |
766 | |
767 | The C<eval 'return sub {...}'> construct could sometimes leak |
768 | memory. This has been fixed. |
769 | |
770 | Operations that aren't filehandle constructors used to leak memory |
771 | when used on invalid filehandles. This has been fixed. |
772 | |
773 | Constructs that modified C<@_> could fail to deallocate values |
774 | in C<@_> and thus leak memory. This has been corrected. |
775 | |
776 | =head2 Spurious subroutine stubs after failed subroutine calls |
777 | |
778 | Perl could sometimes create empty subroutine stubs when a |
779 | subroutine was not found in the package. Such cases stopped |
780 | later method lookups from progressing into base packages. |
781 | This has been corrected. |
782 | |
783 | =head2 Consistent numeric conversions |
784 | |
785 | change#3378,3318 |
786 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
787 | |
788 | =head2 Taint failures under C<-U> |
789 | |
790 | When running in unsafe mode, taint violations could sometimes |
791 | cause silent failures. This has been fixed. |
792 | |
793 | =head2 END blocks and the C<-c> switch |
794 | |
795 | Prior versions used to run BEGIN B<and> END blocks when Perl was |
796 | run in compile-only mode. Since this is typically not the expected |
08cd8952 |
797 | behavior, END blocks are not executed anymore when the C<-c> switch |
a5222a85 |
798 | is used. |
799 | |
800 | Note that something resembling the previous behavior can still be |
801 | obtained by putting C<BEGIN { $^C = 0; exit; } at the very end of |
802 | the top level source file. |
803 | |
804 | =head2 Potential to leak DATA filehandles |
805 | |
806 | Using the C<__DATA__> token creates an implicit filehandle to |
807 | the file that contains the token. It is the program's |
808 | responsibility to close it when it is done reading from it. |
809 | |
810 | This caveat is now better explained in the documentation. |
811 | See L<perldata>. |
812 | |
813 | =head2 Diagnostics follow STDERR |
814 | |
815 | Diagnostic output now goes to whichever file the C<STDERR> handle |
816 | is pointing at, instead of always going to the underlying C runtime |
817 | library's C<stderr>. |
818 | |
819 | =head2 Other fixes for better diagnostics |
820 | |
437784d6 |
821 | Line numbers are no longer suppressed (under most likely circumstances) |
a5222a85 |
822 | during the global destruction phase. |
823 | |
824 | Diagnostics emitted from code running in threads other than the main |
825 | thread are now accompanied by the thread ID. |
826 | |
827 | Embedded null characters in diagnostics now actually show up. They |
828 | used to truncate the message in prior versions. |
829 | |
830 | $foo::a and $foo::b are now exempt from "possible typo" warnings only |
831 | if sort() is encountered in package foo. |
832 | |
833 | Unrecognized alphabetic escapes encountered when parsing quoting |
834 | constructs now generate a warning, since they may take on new |
835 | semantics in later versions of Perl. |
836 | |
837 | =head1 Performance enhancements |
838 | |
839 | =head2 Simple sort() using { $a <=> $b } and the like are optimized |
840 | |
08cd8952 |
841 | Many common sort() operations using a simple inlined block are now |
a5222a85 |
842 | optimized for faster performance. |
843 | |
844 | =head2 Optimized assignments to lexical variables |
845 | |
846 | Certain operations in the RHS of assignment statements have been |
847 | optimized to directly set the lexical variable on the LHS, |
848 | eliminating redundant copying overheads. |
849 | |
850 | =head2 Method lookups optimized |
851 | |
852 | [TODO - Chip Salzenberg <chip@perlsupport.com>] |
853 | |
854 | =head2 Faster mechanism to invoke XSUBs |
855 | |
856 | change#4044,4125 |
857 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
858 | |
859 | =head2 Perl_malloc() improvements |
860 | |
861 | change#4237 |
862 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
863 | |
864 | =head2 Faster subroutine calls |
865 | |
866 | Minor changes in how subroutine calls are handled internally |
867 | provide marginal improvements in performance. |
868 | |
869 | =head1 Platform specific changes |
870 | |
871 | =head2 Additional supported platforms |
ba8251e8 |
872 | |
5fdc711f |
873 | =over 4 |
874 | |
875 | =item * |
876 | |
6c67e1bb |
877 | VM/ESA is now supported. |
878 | |
5fdc711f |
879 | =item * |
880 | |
ee3907e2 |
881 | Siemens BS2000 is now supported under the POSIX Shell. |
882 | |
883 | =item * |
884 | |
2bb14304 |
885 | The Mach CThreads (NEXTSTEP, OPENSTEP) are now supported by the Thread |
886 | extension. |
6c67e1bb |
887 | |
5fdc711f |
888 | =item * |
889 | |
ee3907e2 |
890 | GNU/Hurd is now supported. |
6c67e1bb |
891 | |
00ad96e1 |
892 | =item * |
893 | |
894 | Rhapsody is now supported. |
895 | |
27806c82 |
896 | =item * |
897 | |
898 | EPOC is is now supported (on Psion 5). |
899 | |
5fdc711f |
900 | =back |
901 | |
a5222a85 |
902 | =head2 DOS |
903 | |
904 | [TODO - Laszlo Molnar <laszlo.molnar@eth.ericsson.se>] |
905 | |
906 | =head2 OS/2 |
907 | |
908 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
909 | |
910 | =head2 VMS |
911 | |
912 | [TODO - Charles Bailey <bailey@newman.upenn.edu>] |
913 | |
914 | =head2 Win32 |
915 | |
916 | Site library searches failed to look for ".../site/5.XXX/lib" |
917 | if ".../site/5.XXXYY/lib" wasn't found. This has been corrected. |
918 | |
919 | When given a pathname that consists only of a drivename, such |
920 | as C<A:>, opendir() and stat() now use the current working |
921 | directory for the drive rather than the drive root. |
922 | |
923 | The builtin XSUB functions in the Win32:: namespace are |
924 | documented. See L<Win32>. |
925 | |
926 | $^X now contains the full path name of the running executable. |
927 | |
928 | A Win32::GetLongPathName() function is provided to complement |
929 | Win32::GetFullPathName() and Win32::GetShortPathName(). See L<Win32>. |
930 | |
931 | POSIX::uname() is supported. |
932 | |
933 | system(1,...) now returns true process IDs rather than process |
934 | handles. kill() accepts any real process id, rather than strictly |
935 | return values from system(1,...). |
936 | |
937 | The C<Shell> module is supported. |
938 | |
883d36a6 |
939 | Rudimentary support for building under command.com in Windows 95 |
940 | has been added. |
941 | |
a5222a85 |
942 | [TODO - GSAR] |
943 | |
6c67e1bb |
944 | =head1 New tests |
945 | |
946 | =over 4 |
947 | |
09bef843 |
948 | =item lib/attrs |
949 | |
950 | Compatibility tests for C<sub : attrs> vs the older C<use attrs>. |
951 | |
952 | =item lib/io_const |
6c67e1bb |
953 | |
954 | IO constants (SEEK_*, _IO*). |
14218588 |
955 | |
09bef843 |
956 | =item lib/io_dir |
6c67e1bb |
957 | |
958 | Directory-related IO methods (new, read, close, rewind, tied delete). |
959 | |
09bef843 |
960 | =item lib/io_multihomed |
6c67e1bb |
961 | |
962 | INET sockets with multi-homed hosts. |
963 | |
09bef843 |
964 | =item lib/io_poll |
6c67e1bb |
965 | |
966 | IO poll(). |
967 | |
09bef843 |
968 | =item lib/io_unix |
6c67e1bb |
969 | |
970 | UNIX sockets. |
971 | |
09bef843 |
972 | =item op/attrs |
973 | |
974 | Regression tests for C<my ($x,@y,%z) : attrs> and <sub : attrs>. |
975 | |
6c67e1bb |
976 | =item op/filetest |
977 | |
978 | File test operators. |
979 | |
980 | =item op/lex_assign |
981 | |
5fdc711f |
982 | Verify operations that access pad objects (lexicals and temporaries). |
6c67e1bb |
983 | |
984 | =back |
e02fdbd2 |
985 | |
ba8251e8 |
986 | =head1 Modules and Pragmata |
987 | |
3e8c4fa0 |
988 | =head2 Modules |
989 | |
b7d8191e |
990 | =over 4 |
991 | |
09bef843 |
992 | =item attributes |
993 | |
994 | While used internally by Perl as a pragma, this module also |
995 | provides a way to fetch subroutine and variable attributes. |
996 | See L<attributes>. |
997 | |
a5222a85 |
998 | =item B |
999 | |
1000 | [TODO - Vishal Bhatia <vishal@gol.com>, |
1001 | Nick Ing-Simmons <nick@ni-s.u-net.com>] |
1002 | |
f29c64d6 |
1003 | =item ByteLoader |
1004 | |
a5222a85 |
1005 | The ByteLoader is a dedicated extension to generate and run |
f29c64d6 |
1006 | Perl bytecode. See L<ByteLoader>. |
1007 | |
1008 | =item B |
1009 | |
1010 | The Perl Compiler suite has been extensively reworked for this |
1011 | release. |
1012 | |
a5222a85 |
1013 | =item constant |
1014 | |
1015 | References can now be used. See L<constant>. |
1016 | |
1017 | =item charnames |
1018 | |
1019 | change#4052 |
1020 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
1021 | |
1022 | =item Data::Dumper |
1023 | |
1024 | A C<Maxdepth> setting can be specified to avoid venturing |
1025 | too deeply into data structures that may be very deep. |
1026 | See L<Data::Dumper>. |
1027 | |
1028 | Dumping C<qr//> objects works correctly. |
1029 | |
1030 | =item DB |
1031 | |
1032 | C<DB> is an experimental module that exposes a clean abstraction |
1033 | to Perl's debugging API. |
1034 | |
1035 | =item DB_File |
1036 | |
1037 | [TODO - Paul Marquess <paul.marquess@bt.com>] |
1038 | |
f29c64d6 |
1039 | =item Devel::DProf |
1040 | |
a5222a85 |
1041 | Devel::DProf, a Perl source code profiler has been added. See L<DProf>. |
f29c64d6 |
1042 | |
b7d8191e |
1043 | =item Dumpvalue |
1044 | |
437784d6 |
1045 | The Dumpvalue module provides screen dumps of Perl data. |
b7d8191e |
1046 | |
1047 | =item Benchmark |
1048 | |
868cb350 |
1049 | You can now run tests for I<n> seconds instead of guessing the right |
14218588 |
1050 | number of tests to run: e.g. timethese(-5, ...) will run each |
1051 | code for at least 5 CPU seconds. Zero as the "number of repetitions" |
155776c0 |
1052 | means "for at least 3 CPU seconds". The output format has also |
14218588 |
1053 | changed. For example: |
155776c0 |
1054 | |
1055 | use Benchmark;$x=3;timethese(-5,{a=>sub{$x*$x},b=>sub{$x**2}}) |
1056 | |
1057 | will now output something like this: |
1058 | |
1059 | Benchmark: running a, b, each for at least 5 CPU seconds... |
1060 | a: 5 wallclock secs ( 5.77 usr + 0.00 sys = 5.77 CPU) @ 200551.91/s (n=1156516) |
1061 | b: 4 wallclock secs ( 5.00 usr + 0.02 sys = 5.02 CPU) @ 159605.18/s (n=800686) |
1062 | |
1063 | New features: "each for at least N CPU seconds...", "wallclock secs", |
1064 | and the "@ operations/CPU second (n=operations)". |
b7d8191e |
1065 | |
a5222a85 |
1066 | change#4265,4266,4292 |
1067 | [TODO - Barrie Slaymaker <barries@slaysys.com>] |
1068 | |
f505c983 |
1069 | =item Devel::Peek |
1070 | |
1071 | The Devel::Peek module provides access to the internal representation |
14218588 |
1072 | of Perl variables and data. It is a data debugging tool for the XS programmer. |
f505c983 |
1073 | |
a5222a85 |
1074 | =item ExtUtils::MakeMaker |
1075 | |
1076 | change#4135, also needs docs in module pod |
1077 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
1078 | |
b7d8191e |
1079 | =item Fcntl |
1080 | |
1081 | More Fcntl constants added: F_SETLK64, F_SETLKW64, O_LARGEFILE for |
14218588 |
1082 | large (more than 4G) file access (64-bit support is not yet |
b7d8191e |
1083 | working, though, so no need to get overly excited), Free/Net/OpenBSD |
1084 | locking behaviour flags F_FLOCK, F_POSIX, Linux F_SHLCK, and |
1085 | O_ACCMODE: the mask of O_RDONLY, O_WRONLY, and O_RDWR. |
1086 | |
a5222a85 |
1087 | =item File::Compare |
1088 | |
1089 | A compare_text() function has been added, which allows custom |
1090 | comparison functions. See L<File::Compare>. |
1091 | |
1092 | =item File::Find |
1093 | |
1094 | File::Find now works correctly when the wanted() function is either |
1095 | autoloaded or is a symbolic reference. |
1096 | |
08cd8952 |
1097 | A bug that caused File::Find to lose track of the working directory |
a5222a85 |
1098 | when pruning top-level directories has been fixed. |
1099 | |
becf2bd3 |
1100 | =item File::Glob |
1101 | |
1102 | This extension implements BSD-style file globbing. It will also be |
1103 | used for the internal implementation of the glob() operator if |
1104 | Perl was compiled with -DPERL_INTERNAL_GLOB. See L<File::Glob>. |
1105 | |
f505c983 |
1106 | =item File::Spec |
1107 | |
1108 | New methods have been added to the File::Spec module: devnull() returns |
19799a22 |
1109 | the name of the null device (/dev/null on Unix) and tmpdir() the name of |
14218588 |
1110 | the temp directory (normally /tmp on Unix). There are now also methods |
f505c983 |
1111 | to convert between absolute and relative filenames: abs2rel() and |
14218588 |
1112 | rel2abs(). For compatibility with operating systems that specify volume |
1113 | names in file paths, the splitpath(), splitdir(), and catdir() methods |
f505c983 |
1114 | have been added. |
1115 | |
1116 | =item File::Spec::Functions |
1117 | |
1118 | The new File::Spec::Functions modules provides a function interface |
14218588 |
1119 | to the File::Spec module. Allows shorthand |
f505c983 |
1120 | |
14218588 |
1121 | $fullname = catfile($dir1, $dir2, $file); |
f505c983 |
1122 | |
1123 | instead of |
1124 | |
14218588 |
1125 | $fullname = File::Spec->catfile($dir1, $dir2, $file); |
f505c983 |
1126 | |
a5222a85 |
1127 | =item Getopt::Long |
1128 | |
c6edd1b7 |
1129 | Getopt::Long licensing has changed to allow the Perl Artistic License |
1130 | as well as the GPL. It used to be GPL only, which got in the way of |
1131 | non-GPL applications that wanted to use Getopt::Long. |
1132 | |
1133 | Getopt::Long encourages the use of Pod::Usage to produce help |
1134 | messages. For example: |
1135 | |
1136 | use Getopt::Long; |
1137 | use Pod::Usage; |
1138 | my $man = 0; |
1139 | my $help = 0; |
1140 | GetOptions('help|?' => \$help, man => \$man) or pod2usage(2); |
1141 | pod2usage(1) if $help; |
1142 | pod2usage(-exitstatus => 0, -verbose => 2) if $man; |
1143 | |
1144 | __END__ |
1145 | |
1146 | =head1 NAME |
1147 | |
1148 | sample - Using GetOpt::Long and Pod::Usage |
1149 | |
1150 | =head1 SYNOPSIS |
1151 | |
1152 | sample [options] [file ...] |
1153 | |
1154 | Options: |
1155 | -help brief help message |
1156 | -man full documentation |
1157 | |
1158 | =head1 OPTIONS |
1159 | |
1160 | =over 8 |
1161 | |
1162 | =item B<-help> |
1163 | |
1164 | Print a brief help message and exits. |
1165 | |
1166 | =item B<-man> |
1167 | |
1168 | Prints the manual page and exits. |
1169 | |
1170 | =back |
1171 | |
1172 | =head1 DESCRIPTION |
1173 | |
1174 | B<This program> will read the given input file(s) and do someting |
1175 | useful with the contents thereof. |
1176 | |
1177 | =cut |
1178 | |
1179 | See L<Pod::Usage> for details. |
1180 | |
1181 | A bug that prevented the non-option call-back E<lt>E<gt> from being |
1182 | specified as the first argument has been fixed. |
1183 | |
1184 | To specify the characters E<lt> and E<gt> as option starters, use |
1185 | E<gt>E<lt>. Note, however, that changing option starters is strongly |
1186 | deprecated. |
a5222a85 |
1187 | |
1188 | =item IO |
1189 | |
1190 | write() and syswrite() will now accept a single-argument |
1191 | form of the call, for consistency with Perl's syswrite(). |
1192 | |
1193 | You can now create a TCP-based IO::Socket::INET without forcing |
1194 | a connect attempt. This allows you to configure its options |
1195 | (like making it non-blocking) and then call connect() manually. |
1196 | |
1197 | A bug that prevented the IO::Socket::protocol() accessor |
1198 | from ever returning the correct value has been corrected. |
1199 | |
1200 | =item JPL |
1201 | |
1202 | Java Perl Lingo is now distributed with Perl. See jpl/README |
1203 | for more information. |
1204 | |
883d36a6 |
1205 | =item lib |
1206 | |
1207 | C<use lib> now weeds out any trailing duplicate entries. |
1208 | C<no lib> removes all named entries. |
1209 | |
e16b8f49 |
1210 | =item Math::BigInt |
1211 | |
437784d6 |
1212 | The bitwise operations C<E<lt>E<lt>>, C<E<gt>E<gt>>, C<&>, C<|>, |
e16b8f49 |
1213 | and C<~> are now supported on bigints. |
1214 | |
b7d8191e |
1215 | =item Math::Complex |
7711098a |
1216 | |
14218588 |
1217 | The accessor methods Re, Im, arg, abs, rho, and theta can now also |
868cb350 |
1218 | act as mutators (accessor $z->Re(), mutator $z->Re(3)). |
b7d8191e |
1219 | |
1220 | =item Math::Trig |
1221 | |
14218588 |
1222 | A little bit of radial trigonometry (cylindrical and spherical), |
1223 | radial coordinate conversions, and the great circle distance were added. |
b7d8191e |
1224 | |
a5222a85 |
1225 | =item Pod::Parser |
1226 | |
1227 | [TODO - Brad Appleton <bradapp@enteract.com>] |
1228 | |
1229 | =item Pod::Text and Pod::Man |
1230 | |
1231 | [TODO - Russ Allbery <rra@stanford.edu>] |
1232 | |
f4b9d880 |
1233 | =item SDBM_File |
1234 | |
1235 | An EXISTS method has been added to this module (and sdbm_exists() has |
1236 | been added to the underlying sdbm library), so one can now call exists |
14218588 |
1237 | on an SDBM_File tied hash and get the correct result, rather than a |
f4b9d880 |
1238 | runtime error. |
1239 | |
a5222a85 |
1240 | A bug that may have caused data loss when more than one disk block |
1241 | happens to be read from the database in a single FETCH() has been |
1242 | fixed. |
1243 | |
06ef4121 |
1244 | =item Time::Local |
1245 | |
1246 | The timelocal() and timegm() functions used to silently return bogus |
437784d6 |
1247 | results when the date fell outside the machine's integer range. They |
a5222a85 |
1248 | now consistently croak() if the date falls in an unsupported range. |
06ef4121 |
1249 | |
8fe0a5c4 |
1250 | =item Win32 |
1251 | |
1252 | The error return value in list context has been changed for all functions |
14218588 |
1253 | that return a list of values. Previously these functions returned a list |
1254 | with a single element C<undef> if an error occurred. Now these functions |
1255 | return the empty list in these situations. This applies to the following |
8fe0a5c4 |
1256 | functions: |
1257 | |
14218588 |
1258 | Win32::FsType |
1259 | Win32::GetOSVersion |
8fe0a5c4 |
1260 | |
1261 | The remaining functions are unchanged and continue to return C<undef> on |
1262 | error even in list context. |
1263 | |
1264 | The Win32::SetLastError(ERROR) function has been added as a complement |
1265 | to the Win32::GetLastError() function. |
1266 | |
1267 | The new Win32::GetFullPathName(FILENAME) returns the full absolute |
14218588 |
1268 | pathname for FILENAME in scalar context. In list context it returns |
1269 | a two-element list containing the fully qualified directory name and |
8fe0a5c4 |
1270 | the filename. |
1271 | |
9fe6733a |
1272 | =item DBM Filters |
1273 | |
1274 | A new feature called "DBM Filters" has been added to all the |
14218588 |
1275 | DBM modules--DB_File, GDBM_File, NDBM_File, ODBM_File, and SDBM_File. |
1276 | DBM Filters add four new methods to each DBM module: |
9fe6733a |
1277 | |
1278 | filter_store_key |
1279 | filter_store_value |
1280 | filter_fetch_key |
1281 | filter_fetch_value |
1282 | |
14218588 |
1283 | These can be used to filter key-value pairs before the pairs are |
9fe6733a |
1284 | written to the database or just after they are read from the database. |
1285 | See L<perldbmfilter> for further information. |
1286 | |
b7d8191e |
1287 | =back |
3e8c4fa0 |
1288 | |
1289 | =head2 Pragmata |
1290 | |
437784d6 |
1291 | C<use attrs> is now obsolete, and is only provided for |
09bef843 |
1292 | backward-compatibility. It's been replaced by the C<sub : attributes> |
1293 | syntax. See L<perlsub/"Subroutine Attributes"> and L<attributes>. |
1294 | |
14218588 |
1295 | C<use utf8> to enable UTF-8 and Unicode support. |
43165c05 |
1296 | |
1297 | C<use caller 'encoding'> allows modules to inherit pragmatic attributes |
1298 | from the caller's context. C<encoding> is currently the only supported |
1299 | attribute. |
9d73390d |
1300 | |
4438c4b7 |
1301 | Lexical warnings pragma, C<use warnings;>, to control optional warnings. |
a5222a85 |
1302 | See L<perllexwarn>. |
6c67e1bb |
1303 | |
67d3893f |
1304 | C<use filetest> to control the behaviour of filetests (C<-r> C<-w> |
1305 | ...). Currently only one subpragma implemented, "use filetest |
1306 | 'access';", that uses access(2) or equivalent to check permissions |
1307 | instead of using stat(2) as usual. This matters in filesystems |
1308 | where there are ACLs (access control lists): the stat(2) might lie, |
1309 | but access(2) knows better. |
6c67e1bb |
1310 | |
ba8251e8 |
1311 | =head1 Utility Changes |
1312 | |
a5222a85 |
1313 | =head2 h2ph |
1314 | |
1315 | [TODO - Kurt Starsinic <kstar@chapin.edu>] |
1316 | |
1317 | =head2 perlcc |
1318 | |
1319 | C<perlcc> now supports the C and Bytecode backends. By default, |
1320 | it generates output from the simple C backend rather than the |
1321 | optimized C backend. |
1322 | |
1323 | Support for non-Unix platforms has been improved. |
1324 | |
1325 | =head2 h2xs |
1326 | |
1327 | change#4232 |
1328 | [TODO - Ilya Zakharevich <ilya@math.ohio-state.edu>] |
e02fdbd2 |
1329 | |
ba8251e8 |
1330 | =head1 Documentation Changes |
1331 | |
5fdc711f |
1332 | =over 4 |
1333 | |
883d36a6 |
1334 | =item perlcompile.pod |
1335 | |
1336 | An introduction to using the Perl Compiler suite. |
1337 | |
1338 | =item perlhack.pod |
1339 | |
1340 | Some guidelines for hacking the Perl source code. |
1341 | |
5fdc711f |
1342 | =item perlopentut.pod |
f8284313 |
1343 | |
5fdc711f |
1344 | A tutorial on using open() effectively. |
1345 | |
1346 | =item perlreftut.pod |
1347 | |
1348 | A tutorial that introduces the essentials of references. |
1349 | |
14218588 |
1350 | =item perltootc.pod |
1351 | |
1352 | A tutorial on managing class data for object modules. |
1353 | |
5fdc711f |
1354 | =back |
e02fdbd2 |
1355 | |
ba8251e8 |
1356 | =head1 New Diagnostics |
1357 | |
a99ba403 |
1358 | =over 4 |
1359 | |
09bef843 |
1360 | =item "my sub" not yet implemented |
1361 | |
1362 | (F) Lexically scoped subroutines are not yet implemented. Don't try that |
1363 | yet. |
1364 | |
a99ba403 |
1365 | =item '!' allowed only after types %s |
1366 | |
1367 | (F) The '!' is allowed in pack() and unpack() only after certain types. |
1368 | See L<perlfunc/pack>. |
1369 | |
1370 | =item / cannot take a count |
1371 | |
1372 | (F) You had an unpack template indicating a counted-length string, |
1373 | but you have also specified an explicit size for the string. |
1374 | See L<perlfunc/pack>. |
1375 | |
1376 | =item / must be followed by a, A or Z |
1377 | |
1378 | (F) You had an unpack template indicating a counted-length string, |
1379 | which must be followed by one of the letters a, A or Z |
1380 | to indicate what sort of string is to be unpacked. |
1381 | See L<perlfunc/pack>. |
1382 | |
1383 | =item / must be followed by a*, A* or Z* |
1384 | |
437784d6 |
1385 | (F) You had a pack template indicating a counted-length string, |
a99ba403 |
1386 | Currently the only things that can have their length counted are a*, A* or Z*. |
1387 | See L<perlfunc/pack>. |
1388 | |
1389 | =item / must follow a numeric type |
1390 | |
1391 | (F) You had an unpack template that contained a '#', |
1392 | but this did not follow some numeric unpack specification. |
1393 | See L<perlfunc/pack>. |
1394 | |
1395 | =item Repeat count in pack overflows |
1396 | |
1397 | (F) You can't specify a repeat count so large that it overflows |
1398 | your signed integers. See L<perlfunc/pack>. |
1399 | |
1400 | =item Repeat count in unpack overflows |
1401 | |
1402 | (F) You can't specify a repeat count so large that it overflows |
1403 | your signed integers. See L<perlfunc/unpack>. |
1404 | |
1405 | =item /%s/: Unrecognized escape \\%c passed through |
1406 | |
1407 | (W) You used a backslash-character combination which is not recognized |
1408 | by Perl. This combination appears in an interpolated variable or a |
1409 | C<'>-delimited regular expression. |
1410 | |
1411 | =item /%s/ should probably be written as "%s" |
1412 | |
1413 | (W) You have used a pattern where Perl expected to find a string, |
437784d6 |
1414 | as in the first argument to C<join>. Perl will treat the true |
a99ba403 |
1415 | or false result of matching the pattern against $_ as the string, |
1416 | which is probably not what you had in mind. |
1417 | |
1418 | =item %s() called too early to check prototype |
1419 | |
1420 | (W) You've called a function that has a prototype before the parser saw a |
1421 | definition or declaration for it, and Perl could not check that the call |
1422 | conforms to the prototype. You need to either add an early prototype |
1423 | declaration for the subroutine in question, or move the subroutine |
1424 | definition ahead of the call to get proper prototype checking. Alternatively, |
1425 | if you are certain that you're calling the function correctly, you may put |
1426 | an ampersand before the name to avoid the warning. See L<perlsub>. |
1427 | |
09bef843 |
1428 | =item %s package attribute may clash with future reserved word: %s |
1429 | |
1430 | (W) A lowercase attribute name was used that had a package-specific handler. |
1431 | That name might have a meaning to Perl itself some day, even though it |
1432 | doesn't yet. Perhaps you should use a mixed-case attribute name, instead. |
1433 | See L<attributes>. |
1434 | |
a99ba403 |
1435 | =item (in cleanup) %s |
6b121555 |
1436 | |
a99ba403 |
1437 | (W) This prefix usually indicates that a DESTROY() method raised |
1438 | the indicated exception. Since destructors are usually called by |
1439 | the system at arbitrary points during execution, and often a vast |
1440 | number of times, the warning is issued only once for any number |
1441 | of failures that would otherwise result in the same message being |
1442 | repeated. |
1443 | |
1444 | Failure of user callbacks dispatched using the C<G_KEEPERR> flag |
1445 | could also result in this warning. See L<perlcall/G_KEEPERR>. |
1446 | |
1447 | =item <> should be quotes |
1448 | |
1449 | (F) You wrote C<require E<lt>fileE<gt>> when you should have written |
1450 | C<require 'file'>. |
1451 | |
1452 | =item Attempt to join self |
1453 | |
1454 | (F) You tried to join a thread from within itself, which is an |
1455 | impossible task. You may be joining the wrong thread, or you may |
1456 | need to move the join() to some other thread. |
1457 | |
1458 | =item Bad evalled substitution pattern |
1459 | |
1460 | (F) You've used the /e switch to evaluate the replacement for a |
1461 | substitution, but perl found a syntax error in the code to evaluate, |
1462 | most likely an unexpected right brace '}'. |
1463 | |
1464 | =item Bad realloc() ignored |
1465 | |
1466 | (S) An internal routine called realloc() on something that had never been |
1467 | malloc()ed in the first place. Mandatory, but can be disabled by |
1468 | setting environment variable C<PERL_BADFREE> to 1. |
1469 | |
1470 | =item Binary number > 0b11111111111111111111111111111111 non-portable |
1471 | |
1472 | (W) The binary number you specified is larger than 2**32-1 |
1473 | (4294967295) and therefore non-portable between systems. See |
1474 | L<perlport> for more on portability concerns. |
1475 | |
1476 | =item Bit vector size > 32 non-portable |
1477 | |
1478 | (W) Using bit vector sizes larger than 32 is non-portable. |
1479 | |
1480 | =item Buffer overflow in prime_env_iter: %s |
1481 | |
1482 | (W) A warning peculiar to VMS. While Perl was preparing to iterate over |
1483 | %ENV, it encountered a logical name or symbol definition which was too long, |
1484 | so it was truncated to the string shown. |
1485 | |
1486 | =item Can't check filesystem of script "%s" |
1487 | |
1488 | (P) For some reason you can't check the filesystem of the script for nosuid. |
1489 | |
1490 | =item Can't modify non-lvalue subroutine call |
1491 | |
437784d6 |
1492 | (F) Subroutines meant to be used in lvalue context should be declared as |
1493 | such, see L<perlsub/"Lvalue subroutines">. |
a99ba403 |
1494 | |
1495 | =item Can't read CRTL environ |
1496 | |
1497 | (S) A warning peculiar to VMS. Perl tried to read an element of %ENV |
1498 | from the CRTL's internal environment array and discovered the array was |
1499 | missing. You need to figure out where your CRTL misplaced its environ |
1500 | or define F<PERL_ENV_TABLES> (see L<perlvms>) so that environ is not searched. |
1501 | |
1502 | =item Can't remove %s: %s, skipping file |
1503 | |
1504 | (S) You requested an inplace edit without creating a backup file. Perl |
1505 | was unable to remove the original file to replace it with the modified |
1506 | file. The file was left unmodified. |
1507 | |
1508 | =item Can't return %s from lvalue subroutine |
1509 | |
1510 | (F) Perl detected an attempt to return illegal lvalues (such |
1511 | as temporary or readonly values) from a subroutine used as an lvalue. |
1512 | This is not allowed. |
1513 | |
1514 | =item Can't weaken a nonreference |
1515 | |
1516 | (F) You attempted to weaken something that was not a reference. Only |
1517 | references can be weakened. |
1518 | |
1519 | =item Character class [:%s:] unknown |
1520 | |
1521 | (F) The class in the character class [: :] syntax is unknown. |
437784d6 |
1522 | See L<perlre>. |
a99ba403 |
1523 | |
1524 | =item Character class syntax [%s] belongs inside character classes |
1525 | |
1526 | (W) The character class constructs [: :], [= =], and [. .] go |
1527 | I<inside> character classes, the [] are part of the construct, |
437784d6 |
1528 | for example: /[012[:alpha:]345]/. Note that [= =] and [. .] |
1529 | are not currently implemented; they are simply placeholders for |
1530 | future extensions. |
a99ba403 |
1531 | |
1532 | =item Constant is not %s reference |
1533 | |
1534 | (F) A constant value (perhaps declared using the C<use constant> pragma) |
1535 | is being dereferenced, but it amounts to the wrong type of reference. The |
1536 | message indicates the type of reference that was expected. This usually |
1537 | indicates a syntax error in dereferencing the constant value. |
1538 | See L<perlsub/"Constant Functions"> and L<constant>. |
1539 | |
1540 | =item constant(%s): %%^H is not localized |
1541 | |
1542 | (F) When setting compile-time-lexicalized hash %^H one should set the |
1543 | corresponding bit of $^H as well. |
1544 | |
1545 | =item constant(%s): %s |
1546 | |
1547 | (F) Compile-time-substitutions (such as overloaded constants and |
1548 | character names) were not correctly set up. |
1549 | |
1550 | =item defined(@array) is deprecated |
1551 | |
1552 | (D) defined() is not usually useful on arrays because it checks for an |
1553 | undefined I<scalar> value. If you want to see if the array is empty, |
1554 | just use C<if (@array) { # not empty }> for example. |
1555 | |
1556 | =item defined(%hash) is deprecated |
1557 | |
1558 | (D) defined() is not usually useful on hashes because it checks for an |
1559 | undefined I<scalar> value. If you want to see if the hash is empty, |
1560 | just use C<if (%hash) { # not empty }> for example. |
1561 | |
1562 | =item Did not produce a valid header |
1563 | |
1564 | See Server error. |
1565 | |
1566 | =item Document contains no data |
1567 | |
1568 | See Server error. |
1569 | |
1570 | =item entering effective %s failed |
1571 | |
1572 | (F) While under the C<use filetest> pragma, switching the real and |
1573 | effective uids or gids failed. |
6b121555 |
1574 | |
af8c498a |
1575 | =item Filehandle %s opened only for output |
6b121555 |
1576 | |
af8c498a |
1577 | (W) You tried to read from a filehandle opened only for writing. If you |
437784d6 |
1578 | intended it to be a read/write filehandle, you needed to open it with |
af8c498a |
1579 | "+E<lt>" or "+E<gt>" or "+E<gt>E<gt>" instead of with "E<lt>" or nothing. If |
1580 | you intended only to read from the file, use "E<lt>". See |
1581 | L<perlfunc/open>. |
e02fdbd2 |
1582 | |
a99ba403 |
1583 | =item Hexadecimal number > 0xffffffff non-portable |
1584 | |
1585 | (W) The hexadecimal number you specified is larger than 2**32-1 |
1586 | (4294967295) and therefore non-portable between systems. See |
1587 | L<perlport> for more on portability concerns. |
1588 | |
1589 | =item Ill-formed CRTL environ value "%s" |
1590 | |
1591 | (W) A warning peculiar to VMS. Perl tried to read the CRTL's internal |
1592 | environ array, and encountered an element without the C<=> delimiter |
1593 | used to spearate keys from values. The element is ignored. |
1594 | |
1595 | =item Ill-formed message in prime_env_iter: |%s| |
1596 | |
1597 | (W) A warning peculiar to VMS. Perl tried to read a logical name |
1598 | or CLI symbol definition when preparing to iterate over %ENV, and |
1599 | didn't see the expected delimiter between key and value, so the |
1600 | line was ignored. |
1601 | |
1602 | =item Illegal binary digit %s |
1603 | |
437784d6 |
1604 | (F) You used a digit other than 0 or 1 in a binary number. |
a99ba403 |
1605 | |
1606 | =item Illegal binary digit %s ignored |
1607 | |
1608 | (W) You may have tried to use a digit other than 0 or 1 in a binary number. |
1609 | Interpretation of the binary number stopped before the offending digit. |
1610 | |
1611 | =item Illegal number of bits in vec |
1612 | |
1613 | (F) The number of bits in vec() (the third argument) must be a power of |
1614 | two from 1 to 32 (or 64, if your platform supports that). |
1615 | |
1616 | =item Integer overflow in %s number |
1617 | |
1618 | (W) The hexadecimal, octal or binary number you have specified either |
c6edd1b7 |
1619 | as a literal or as an argument to hex() or oct() is too big for your |
a99ba403 |
1620 | architecture, and has been converted to a floating point number. On a |
1621 | 32-bit architecture the largest hexadecimal, octal or binary number |
1622 | representable without overflow is 0xFFFFFFFF, 037777777777, or |
1623 | 0b11111111111111111111111111111111 respectively. Note that Perl |
1624 | transparently promotes all numbers to a floating point representation |
1625 | internally--subject to loss of precision errors in subsequent |
1626 | operations. |
1627 | |
09bef843 |
1628 | =item Invalid %s attribute: %s |
1629 | |
1630 | The indicated attribute for a subroutine or variable was not recognized |
1631 | by Perl or by a user-supplied handler. See L<attributes>. |
1632 | |
1633 | =item Invalid %s attributes: %s |
1634 | |
1635 | The indicated attributes for a subroutine or variable were not recognized |
1636 | by Perl or by a user-supplied handler. See L<attributes>. |
1637 | |
1638 | =item Invalid separator character %s in attribute list |
1639 | |
1640 | (F) Something other than a comma or whitespace was seen between the |
1641 | elements of an attribute list. If the previous attribute |
1642 | had a parenthesised parameter list, perhaps that list was terminated |
1643 | too soon. See L<attributes>. |
1644 | |
a99ba403 |
1645 | =item Invalid separator character %s in subroutine attribute list |
1646 | |
1647 | (F) Something other than a comma or whitespace was seen between the |
1648 | elements of a subroutine attribute list. If the previous attribute |
1649 | had a parenthesised parameter list, perhaps that list was terminated |
1650 | too soon. |
1651 | |
1652 | =item leaving effective %s failed |
1653 | |
1654 | (F) While under the C<use filetest> pragma, switching the real and |
1655 | effective uids or gids failed. |
1656 | |
1657 | =item Lvalue subs returning %s not implemented yet |
1658 | |
1659 | (F) Due to limitations in the current implementation, array and hash |
1660 | values cannot be returned in subroutines used in lvalue context. |
1661 | See L<perlsub/"Lvalue subroutines">. |
1662 | |
1663 | =item Method %s not permitted |
1664 | |
1665 | See Server error. |
1666 | |
1667 | =item Missing %sbrace%s on \N{} |
1668 | |
1669 | (F) Wrong syntax of character name literal C<\N{charname}> within |
1670 | double-quotish context. |
1671 | |
06eaf0bc |
1672 | =item Missing command in piped open |
1673 | |
1674 | (W) You used the C<open(FH, "| command")> or C<open(FH, "command |")> |
1675 | construction, but the command was missing or blank. |
1676 | |
09bef843 |
1677 | =item Missing name in "my sub" |
1678 | |
1679 | (F) The reserved syntax for lexically scoped subroutines requires that they |
1680 | have a name with which they can be found. |
1681 | |
a99ba403 |
1682 | =item no UTC offset information; assuming local time is UTC |
1683 | |
1684 | (S) A warning peculiar to VMS. Perl was unable to find the local |
1685 | timezone offset, so it's assuming that local system time is equivalent |
1686 | to UTC. If it's not, define the logical name F<SYS$TIMEZONE_DIFFERENTIAL> |
1687 | to translate to the number of seconds which need to be added to UTC to |
1688 | get local time. |
1689 | |
1690 | =item Octal number > 037777777777 non-portable |
1691 | |
1692 | (W) The octal number you specified is larger than 2**32-1 (4294967295) |
1693 | and therefore non-portable between systems. See L<perlport> for more |
1694 | on portability concerns. |
1695 | |
1696 | See also L<perlport> for writing portable code. |
1697 | |
1698 | =item panic: del_backref |
1699 | |
1700 | (P) Failed an internal consistency check while trying to reset a weak |
1701 | reference. |
1702 | |
1703 | =item panic: kid popen errno read |
1704 | |
1705 | (F) forked child returned an incomprehensible message about its errno. |
1706 | |
1707 | =item panic: magic_killbackrefs |
1708 | |
1709 | (P) Failed an internal consistency check while trying to reset all weak |
1710 | references to an object. |
1711 | |
1712 | =item Possible Y2K bug: %s |
1713 | |
1714 | (W) You are concatenating the number 19 with another number, which |
1715 | could be a potential Year 2000 problem. |
1716 | |
1717 | =item Premature end of script headers |
1718 | |
1719 | See Server error. |
1720 | |
1721 | =item realloc() of freed memory ignored |
1722 | |
1723 | (S) An internal routine called realloc() on something that had already |
1724 | been freed. |
1725 | |
1726 | =item Reference is already weak |
1727 | |
1728 | (W) You have attempted to weaken a reference that is already weak. |
1729 | Doing so has no effect. |
1730 | |
1731 | =item setpgrp can't take arguments |
1732 | |
1733 | (F) Your system has the setpgrp() from BSD 4.2, which takes no arguments, |
1734 | unlike POSIX setpgid(), which takes a process ID and process group ID. |
1735 | |
1736 | =item Strange *+?{} on zero-length expression |
1737 | |
1738 | (W) You applied a regular expression quantifier in a place where it |
1739 | makes no sense, such as on a zero-width assertion. |
1740 | Try putting the quantifier inside the assertion instead. For example, |
1741 | the way to match "abc" provided that it is followed by three |
1742 | repetitions of "xyz" is C</abc(?=(?:xyz){3})/>, not C</abc(?=xyz){3}/>. |
1743 | |
1744 | =item switching effective %s is not implemented |
1745 | |
1746 | (F) While under the C<use filetest> pragma, we cannot switch the |
1747 | real and effective uids or gids. |
1748 | |
437784d6 |
1749 | =item This Perl can't reset CRTL environ elements (%s) |
a99ba403 |
1750 | |
1751 | =item This Perl can't set CRTL environ elements (%s=%s) |
1752 | |
1753 | (W) Warnings peculiar to VMS. You tried to change or delete an element |
1754 | of the CRTL's internal environ array, but your copy of Perl wasn't |
1755 | built with a CRTL that contained the setenv() function. You'll need to |
1756 | rebuild Perl with a CRTL that does, or redefine F<PERL_ENV_TABLES> (see |
1757 | L<perlvms>) so that the environ array isn't the target of the change to |
1758 | %ENV which produced the warning. |
1759 | |
1760 | =item Unknown open() mode '%s' |
1761 | |
437784d6 |
1762 | (F) The second argument of 3-argument open() is not among the list |
1763 | of valid modes: C<L<lt>>, C<L<gt>>, C<E<gt>E<gt>>, C<+L<lt>>, |
1764 | C<+L<gt>>, C<+E<gt>E<gt>>, C<-|>, C<|->. |
a99ba403 |
1765 | |
1766 | =item Unknown process %x sent message to prime_env_iter: %s |
1767 | |
1768 | (P) An error peculiar to VMS. Perl was reading values for %ENV before |
1769 | iterating over it, and someone else stuck a message in the stream of |
1770 | data Perl expected. Someone's very confused, or perhaps trying to |
1771 | subvert Perl's population of %ENV for nefarious purposes. |
1772 | |
af8c498a |
1773 | =item Unrecognized escape \\%c passed through |
1774 | |
1775 | (W) You used a backslash-character combination which is not recognized |
1776 | by Perl. |
1777 | |
09bef843 |
1778 | =item Unterminated attribute parameter in attribute list |
1779 | |
1780 | (F) The lexer saw an opening (left) parenthesis character while parsing an |
1781 | attribute list, but the matching closing (right) parenthesis |
1782 | character was not found. You may need to add (or remove) a backslash |
1783 | character to get your parentheses to balance. See L<attributes>. |
1784 | |
1785 | =item Unterminated attribute list |
1786 | |
1787 | (F) The lexer found something other than a simple identifier at the start |
1788 | of an attribute, and it wasn't a semicolon or the start of a |
1789 | block. Perhaps you terminated the parameter list of the previous attribute |
1790 | too soon. See L<attributes>. |
1791 | |
09bef843 |
1792 | =item Unterminated attribute parameter in subroutine attribute list |
1793 | |
1794 | (F) The lexer saw an opening (left) parenthesis character while parsing a |
1795 | subroutine attribute list, but the matching closing (right) parenthesis |
1796 | character was not found. You may need to add (or remove) a backslash |
1797 | character to get your parentheses to balance. |
1798 | |
1799 | =item Unterminated subroutine attribute list |
1800 | |
1801 | (F) The lexer found something other than a simple identifier at the start |
1802 | of a subroutine attribute, and it wasn't a semicolon or the start of a |
1803 | block. Perhaps you terminated the parameter list of the previous attribute |
1804 | too soon. |
1805 | |
a99ba403 |
1806 | =item Value of CLI symbol "%s" too long |
eb6e2d6f |
1807 | |
a99ba403 |
1808 | (W) A warning peculiar to VMS. Perl tried to read the value of an %ENV |
1809 | element from a CLI symbol table, and found a resultant string longer |
1810 | than 1024 characters. The return value has been truncated to 1024 |
1811 | characters. |
eb6e2d6f |
1812 | |
a99ba403 |
1813 | =item Version number must be a constant number |
ba8251e8 |
1814 | |
a99ba403 |
1815 | (P) The attempt to translate a C<use Module n.n LIST> statement into |
1816 | its equivalent C<BEGIN> block found an internal inconsistency with |
1817 | the version number. |
1818 | |
1819 | =back |
27806c82 |
1820 | |
a5222a85 |
1821 | =head1 Obsolete Diagnostics |
3175b8cd |
1822 | |
a99ba403 |
1823 | =over 4 |
1824 | |
1825 | =item Character class syntax [: :] is reserved for future extensions |
1826 | |
1827 | (W) Within regular expression character classes ([]) the syntax beginning |
1828 | with "[:" and ending with ":]" is reserved for future extensions. |
1829 | If you need to represent those character sequences inside a regular |
1830 | expression character class, just quote the square brackets with the |
1831 | backslash: "\[:" and ":\]". |
1832 | |
1833 | =item Ill-formed logical name |%s| in prime_env_iter |
1834 | |
1835 | (W) A warning peculiar to VMS. A logical name was encountered when preparing |
1836 | to iterate over %ENV which violates the syntactic rules governing logical |
1837 | names. Because it cannot be translated normally, it is skipped, and will not |
1838 | appear in %ENV. This may be a benign occurrence, as some software packages |
1839 | might directly modify logical name tables and introduce nonstandard names, |
1840 | or it may indicate that a logical name table has been corrupted. |
1841 | |
1842 | =item regexp too big |
1843 | |
1844 | (F) The current implementation of regular expressions uses shorts as |
1845 | address offsets within a string. Unfortunately this means that if |
1846 | the regular expression compiles to longer than 32767, it'll blow up. |
1847 | Usually when you want a regular expression this big, there is a better |
1848 | way to do it with multiple statements. See L<perlre>. |
1849 | |
1850 | =item Use of "$$<digit>" to mean "${$}<digit>" is deprecated |
1851 | |
1852 | (D) Perl versions before 5.004 misinterpreted any type marker followed |
1853 | by "$" and a digit. For example, "$$0" was incorrectly taken to mean |
1854 | "${$}0" instead of "${$0}". This bug is (mostly) fixed in Perl 5.004. |
1855 | |
1856 | However, the developers of Perl 5.004 could not fix this bug completely, |
1857 | because at least two widely-used modules depend on the old meaning of |
1858 | "$$0" in a string. So Perl 5.004 still interprets "$$<digit>" in the |
1859 | old (broken) way inside strings; but it generates this message as a |
1860 | warning. And in Perl 5.005, this special treatment will cease. |
1861 | |
1862 | =back |
3175b8cd |
1863 | |
ba8251e8 |
1864 | =head1 BUGS |
1865 | |
437784d6 |
1866 | If you find what you think is a bug, you might check the |
14218588 |
1867 | articles recently posted to the comp.lang.perl.misc newsgroup. |
ba8251e8 |
1868 | There may also be information at http://www.perl.com/perl/, the Perl |
1869 | Home Page. |
1870 | |
1871 | If you believe you have an unreported bug, please run the B<perlbug> |
14218588 |
1872 | program included with your release. Make sure to trim your bug down |
ba8251e8 |
1873 | to a tiny but sufficient test case. Your bug report, along with the |
14218588 |
1874 | output of C<perl -V>, will be sent off to perlbug@perl.com to be |
ba8251e8 |
1875 | analysed by the Perl porting team. |
1876 | |
1877 | =head1 SEE ALSO |
1878 | |
1879 | The F<Changes> file for exhaustive details on what changed. |
1880 | |
1881 | The F<INSTALL> file for how to build Perl. |
1882 | |
1883 | The F<README> file for general stuff. |
1884 | |
1885 | The F<Artistic> and F<Copying> files for copyright information. |
1886 | |
1887 | =head1 HISTORY |
1888 | |
a5222a85 |
1889 | Written by Gurusamy Sarathy <F<gsar@activestate.com>>, with many |
1890 | contributions from The Perl Porters. |
ba8251e8 |
1891 | |
1892 | Send omissions or corrections to <F<perlbug@perl.com>>. |
1893 | |
1894 | =cut |