Commit | Line | Data |
ba370e9b |
1 | =head1 NAME |
cc0fca54 |
2 | |
f39f21d8 |
3 | perldelta - what is new for perl v5.8.0 |
cc0fca54 |
4 | |
5 | =head1 DESCRIPTION |
6 | |
44da0e71 |
7 | This document describes differences between the 5.6.0 release |
8 | and the 5.8.0 release. |
f39f21d8 |
9 | |
44da0e71 |
10 | Many of the bug fixes in 5.8.0 were already seen in the 5.6.1 |
11 | maintenance release since the two releases were kept closely |
12 | coordinated. |
13 | |
4f8e5944 |
14 | If you are upgrading from Perl 5.005_03, you might also want |
15 | to read L<perl56delta>. |
16 | |
44da0e71 |
17 | =head1 Highlights In 5.8.0 |
76663d67 |
18 | |
19 | =over 4 |
20 | |
21 | =item * |
22 | |
23 | Better Unicode support |
24 | |
25 | =item * |
26 | |
27 | New Thread Implementation |
28 | |
29 | =item * |
30 | |
31 | Many New Modules |
32 | |
33 | =item * |
34 | |
35 | Better Numeric Accuracy |
36 | |
37 | =item * |
38 | |
39 | Safe Signals |
40 | |
41 | =item * |
42 | |
43 | More Extensive Regression Testing |
44 | |
45 | =back |
46 | |
f39f21d8 |
47 | =head1 Incompatible Changes |
48 | |
77c8cf41 |
49 | =head2 64-bit platforms and malloc |
50 | |
057b7f2b |
51 | If your pointers are 64 bits wide, the Perl malloc is no longer being |
c2e23569 |
52 | used because it does not work well with 8-byte pointers. Also, |
61947107 |
53 | usually the system mallocs on such platforms are much better optimized |
c2e23569 |
54 | for such large memory models than the Perl malloc. Some memory-hungry |
55 | Perl applications like the PDL don't work well with Perl's malloc. |
56 | Finally, other applications than Perl (like modperl) tend to prefer |
57 | the system malloc. Such platforms include Alpha and 64-bit HPPA, |
58 | MIPS, PPC, and Sparc. |
77c8cf41 |
59 | |
60 | =head2 AIX Dynaloading |
61 | |
62 | The AIX dynaloading now uses in AIX releases 4.3 and newer the native |
63 | dlopen interface of AIX instead of the old emulated interface. This |
64 | change will probably break backward compatibility with compiled |
65 | modules. The change was made to make Perl more compliant with other |
66 | applications like modperl which are using the AIX native interface. |
67 | |
95f0a2f1 |
68 | =head2 Attributes for C<my> variables now handled at run-time. |
69 | |
70 | The C<my EXPR : ATTRS> syntax now applies variable attributes at |
71 | run-time. (Subroutine and C<our> variables still get attributes applied |
72 | at compile-time.) See L<attributes> for additional details. In particular, |
73 | however, this allows variable attributes to be useful for C<tie> interfaces, |
c4f1ce08 |
74 | which was a deficiency of earlier releases. Note that the new semantics |
75 | doesn't work with the Attribute::Handlers module (as of version 0.76). |
95f0a2f1 |
76 | |
77c8cf41 |
77 | =head2 Socket Extension Dynamic in VMS |
78 | |
79 | The Socket extension is now dynamically loaded instead of being |
80 | statically built in. This may or may not be a problem with ancient |
81 | TCP/IP stacks of VMS: we do not know since we weren't able to test |
82 | Perl in such configurations. |
83 | |
00bb525a |
84 | =head2 IEEE-format Floating Point Default on OpenVMS Alpha |
85 | |
86 | Perl now uses IEEE format (T_FLOAT) as the default internal floating |
87 | point format on OpenVMS Alpha, potentially breaking binary compatibility |
88 | with external libraries or existing data. G_FLOAT is still available as |
89 | a configuration option. The default on VAX (D_FLOAT) has not changed. |
90 | |
77c8cf41 |
91 | =head2 Different Definition of the Unicode Character Classes \p{In...} |
92 | |
93 | As suggested by the Unicode consortium, the Unicode character classes |
94 | now prefer I<scripts> as opposed to I<blocks> (as defined by Unicode); |
95 | in Perl, when the C<\p{In....}> and the C<\p{In....}> regular expression |
96 | constructs are used. This has changed the definition of some of those |
97 | character classes. |
98 | |
99 | The difference between scripts and blocks is that scripts are the |
100 | glyphs used by a language or a group of languages, while the blocks |
101 | are more artificial groupings of 256 characters based on the Unicode |
102 | numbering. |
103 | |
104 | In general this change results in more inclusive Unicode character |
105 | classes, but changes to the other direction also do take place: |
106 | for example while the script C<Latin> includes all the Latin |
107 | characters and their various diacritic-adorned versions, it |
108 | does not include the various punctuation or digits (since they |
109 | are not solely C<Latin>). |
110 | |
111 | Changes in the character class semantics may have happened if a script |
112 | and a block happen to have the same name, for example C<Hebrew>. |
113 | In such cases the script wins and C<\p{InHebrew}> now means the script |
114 | definition of Hebrew. The block definition in still available, |
115 | though, by appending C<Block> to the name: C<\p{InHebrewBlock}> means |
116 | what C<\p{InHebrew}> meant in perl 5.6.0. For the full list |
117 | of affected character classes, see L<perlunicode/Blocks>. |
118 | |
61947107 |
119 | =head2 Perl Parser Stress Tested |
120 | |
121 | The Perl parser has been stress tested using both random input and |
122 | Markov chain input and the few found crashes and lockups have been |
123 | fixed. |
124 | |
c2e23569 |
125 | =head2 REF(...) Instead Of SCALAR(...) |
77c8cf41 |
126 | |
057b7f2b |
127 | A reference to a reference now stringifies as "REF(0x81485ec)" instead |
c2e23569 |
128 | of "SCALAR(0x81485ec)" in order to be more consistent with the return |
129 | value of ref(). |
77c8cf41 |
130 | |
c2e23569 |
131 | =head2 Deprecations |
77c8cf41 |
132 | |
61947107 |
133 | =over 4 |
77c8cf41 |
134 | |
61947107 |
135 | =item * |
f39f21d8 |
136 | |
61947107 |
137 | The semantics of bless(REF, REF) were unclear and until someone proves |
138 | it to make some sense, it is forbidden. |
f39f21d8 |
139 | |
140 | =item * |
141 | |
c2e23569 |
142 | The obsolete chat2 library that should never have been allowed |
143 | to escape the laboratory has been decommissioned. |
f39f21d8 |
144 | |
145 | =item * |
146 | |
61947107 |
147 | The very dusty examples in the eg/ directory have been removed. |
148 | Suggestions for new shiny examples welcome but the main issue is that |
149 | the examples need to be documented, tested and (most importantly) |
150 | maintained. |
f39f21d8 |
151 | |
152 | =item * |
153 | |
c2e23569 |
154 | The (bogus) escape sequences \8 and \9 now give an optional warning |
155 | ("Unrecognized escape passed through"). There is no need to \-escape |
156 | any C<\w> character. |
f39f21d8 |
157 | |
158 | =item * |
159 | |
c2e23569 |
160 | The list of filenames from glob() (or <...>) is now by default sorted |
44da0e71 |
161 | alphabetically to be csh-compliant (which is what happened before |
162 | in most UNIX platforms). (bsd_glob() does still sort platform |
c2e23569 |
163 | natively, ASCII or EBCDIC, unless GLOB_ALPHASORT is specified.) |
f39f21d8 |
164 | |
165 | =item * |
166 | |
44da0e71 |
167 | Spurious syntax errors generated in certain situations, when glob() |
168 | caused File::Glob to be loaded for the first time, have been fixed. |
169 | |
170 | =item * |
171 | |
c2e23569 |
172 | Although "you shouldn't do that", it was possible to write code that |
173 | depends on Perl's hashed key order (Data::Dumper does this). The new |
174 | algorithm "One-at-a-Time" produces a different hashed key order. |
175 | More details are in L</"Performance Enhancements">. |
f39f21d8 |
176 | |
177 | =item * |
178 | |
61947107 |
179 | lstat(FILEHANDLE) now gives a warning because the operation makes no sense. |
180 | In future releases this may become a fatal error. |
f39f21d8 |
181 | |
182 | =item * |
183 | |
057b7f2b |
184 | The C<package;> syntax (C<package> without an argument) has been |
c2e23569 |
185 | deprecated. Its semantics were never that clear and its |
186 | implementation even less so. If you have used that feature to |
187 | disallow all but fully qualified variables, C<use strict;> instead. |
61947107 |
188 | |
189 | =item * |
190 | |
c2e23569 |
191 | The unimplemented POSIX regex features [[.cc.]] and [[=c=]] are still |
192 | recognised but now cause fatal errors. The previous behaviour of |
193 | ignoring them by default and warning if requested was unacceptable |
194 | since it, in a way, falsely promised that the features could be used. |
61947107 |
195 | |
196 | =item * |
197 | |
c2e23569 |
198 | The current user-visible implementation of pseudo-hashes (the weird |
199 | use of the first array element) is deprecated starting from Perl 5.8.0 |
200 | and will be removed in Perl 5.10.0, and the feature will be |
201 | implemented differently. Not only is the current interface rather |
202 | ugly, but the current implementation slows down normal array and hash |
203 | use quite noticeably. The C<fields> pragma interface will remain |
204 | available. |
61947107 |
205 | |
206 | =item * |
207 | |
aecce728 |
208 | The syntaxes C<< @a->[...] >> and C<< %h->{...} >> have now been deprecated. |
61947107 |
209 | |
210 | =item * |
211 | |
c2e23569 |
212 | After years of trying the suidperl is considered to be too complex to |
213 | ever be considered truly secure. The suidperl functionality is likely |
214 | to be removed in a future release. |
215 | |
216 | =item * |
217 | |
218 | The long deprecated uppercase aliases for the string comparison |
219 | operators (EQ, NE, LT, LE, GE, GT) have now been removed. |
220 | |
221 | =item * |
222 | |
223 | The tr///C and tr///U features have been removed and will not return; |
224 | the interface was a mistake. Sorry about that. For similar |
225 | functionality, see pack('U0', ...) and pack('C0', ...). |
f39f21d8 |
226 | |
227 | =back |
228 | |
61947107 |
229 | =head1 Core Enhancements |
230 | |
77c8cf41 |
231 | =head2 PerlIO is Now The Default |
f39f21d8 |
232 | |
233 | =over 4 |
234 | |
235 | =item * |
236 | |
77c8cf41 |
237 | IO is now by default done via PerlIO rather than system's "stdio". |
238 | PerlIO allows "layers" to be "pushed" onto a file handle to alter the |
239 | handle's behaviour. Layers can be specified at open time via 3-arg |
240 | form of open: |
f39f21d8 |
241 | |
77c8cf41 |
242 | open($fh,'>:crlf :utf8', $path) || ... |
f39f21d8 |
243 | |
77c8cf41 |
244 | or on already opened handles via extended C<binmode>: |
f39f21d8 |
245 | |
77c8cf41 |
246 | binmode($fh,':encoding(iso-8859-7)'); |
f39f21d8 |
247 | |
77c8cf41 |
248 | The built-in layers are: unix (low level read/write), stdio (as in |
249 | previous Perls), perlio (re-implementation of stdio buffering in a |
250 | portable manner), crlf (does CRLF <=> "\n" translation as on Win32, |
251 | but available on any platform). A mmap layer may be available if |
252 | platform supports it (mostly UNIXes). |
f39f21d8 |
253 | |
77c8cf41 |
254 | Layers to be applied by default may be specified via the 'open' pragma. |
255 | |
256 | See L</"Installation and Configuration Improvements"> for the effects |
257 | of PerlIO on your architecture name. |
f39f21d8 |
258 | |
259 | =item * |
260 | |
77c8cf41 |
261 | File handles can be marked as accepting Perl's internal encoding of Unicode |
262 | (UTF-8 or UTF-EBCDIC depending on platform) by a pseudo layer ":utf8" : |
f39f21d8 |
263 | |
77c8cf41 |
264 | open($fh,">:utf8","Uni.txt"); |
f39f21d8 |
265 | |
77c8cf41 |
266 | Note for EBCDIC users: the pseudo layer ":utf8" is erroneously named |
267 | for you since it's not UTF-8 what you will be getting but instead |
268 | UTF-EBCDIC. See L<perlunicode>, L<utf8>, and |
269 | http://www.unicode.org/unicode/reports/tr16/ for more information. |
270 | In future releases this naming may change. |
f39f21d8 |
271 | |
272 | =item * |
273 | |
77c8cf41 |
274 | File handles can translate character encodings from/to Perl's internal |
275 | Unicode form on read/write via the ":encoding()" layer. |
f39f21d8 |
276 | |
277 | =item * |
278 | |
77c8cf41 |
279 | File handles can be opened to "in memory" files held in Perl scalars via: |
280 | |
281 | open($fh,'>', \$variable) || ... |
f39f21d8 |
282 | |
283 | =item * |
284 | |
77c8cf41 |
285 | Anonymous temporary files are available without need to |
286 | 'use FileHandle' or other module via |
f39f21d8 |
287 | |
77c8cf41 |
288 | open($fh,"+>", undef) || ... |
f39f21d8 |
289 | |
77c8cf41 |
290 | That is a literal undef, not an undefined value. |
f39f21d8 |
291 | |
292 | =item * |
293 | |
77c8cf41 |
294 | The list form of C<open> is now implemented for pipes (at least on UNIX): |
f39f21d8 |
295 | |
77c8cf41 |
296 | open($fh,"-|", 'cat', '/etc/motd') |
f39f21d8 |
297 | |
77c8cf41 |
298 | creates a pipe, and runs the equivalent of exec('cat', '/etc/motd') in |
299 | the child process. |
f39f21d8 |
300 | |
e1f170bd |
301 | =back |
f39f21d8 |
302 | |
3e33716f |
303 | =head2 Safe Signals |
f39f21d8 |
304 | |
e1f170bd |
305 | Perl used to be fragile in that signals arriving at inopportune moments |
306 | could corrupt Perl's internal state. Now Perl postpones handling of |
3e33716f |
307 | signals until it's safe (between opcodes). |
308 | |
309 | This change may have surprising side effects because signals no more |
310 | interrupt Perl instantly. Perl will now first finish whatever it was |
311 | doing, like finishing an internal operation (like sort()) or an |
312 | external operation (like an I/O operation), and only then look at any |
313 | arrived signals (and before starting the next operation). No more corrupt |
314 | internal state since the current operation is always finished first, |
315 | but the signal may take more time to get heard. |
f39f21d8 |
316 | |
e1f170bd |
317 | =head2 Unicode Overhaul |
f39f21d8 |
318 | |
e1f170bd |
319 | Unicode in general should be now much more usable than in Perl 5.6.0 |
320 | (or even in 5.6.1). Unicode can be used in hash keys, Unicode in |
321 | regular expressions should work now, Unicode in tr/// should work now, |
322 | Unicode in I/O should work now. |
f39f21d8 |
323 | |
e1f170bd |
324 | =over 4 |
f39f21d8 |
325 | |
326 | =item * |
327 | |
e1f170bd |
328 | The Unicode Character Database coming with Perl has been upgraded |
329 | to Unicode 3.1.1. For more information, see http://www.unicode.org/. |
f39f21d8 |
330 | |
331 | =item * |
332 | |
77c8cf41 |
333 | For developers interested in enhancing Perl's Unicode capabilities: |
334 | almost all the UCD files are included with the Perl distribution in |
e1f170bd |
335 | the lib/unicore subdirectory. The most notable omission, for space |
77c8cf41 |
336 | considerations, is the Unihan database. |
f39f21d8 |
337 | |
338 | =item * |
339 | |
77c8cf41 |
340 | The Unicode character classes \p{Blank} and \p{SpacePerl} have been |
341 | added. "Blank" is like C isblank(), that is, it contains only |
342 | "horizontal whitespace" (the space character is, the newline isn't), |
343 | and the "SpacePerl" is the Unicode equivalent of C<\s> (\p{Space} |
344 | isn't, since that includes the vertical tabulator character, whereas |
345 | C<\s> doesn't.) |
f39f21d8 |
346 | |
347 | =back |
348 | |
77c8cf41 |
349 | =head2 Understanding of Numbers |
350 | |
351 | In general a lot of fixing has happened in the area of Perl's |
352 | understanding of numbers, both integer and floating point. Since in |
353 | many systems the standard number parsing functions like C<strtoul()> |
354 | and C<atof()> seem to have bugs, Perl tries to work around their |
355 | deficiencies. This results hopefully in more accurate numbers. |
f39f21d8 |
356 | |
e1f170bd |
357 | Perl now tries internally to use integer values in numeric conversions |
358 | and basic arithmetics (+ - * /) if the arguments are integers, and |
359 | tries also to keep the results stored internally as integers. |
057b7f2b |
360 | This change leads to often slightly faster and always less lossy |
e1f170bd |
361 | arithmetics. (Previously Perl always preferred floating point numbers |
362 | in its math.) |
363 | |
364 | =head2 Miscellaneous Enhancements |
365 | |
f39f21d8 |
366 | =over 4 |
367 | |
368 | =item * |
369 | |
e1f170bd |
370 | AUTOLOAD is now lvaluable, meaning that you can add the :lvalue attribute |
371 | to AUTOLOAD subroutines and you can assign to the AUTOLOAD return value. |
372 | |
373 | =item * |
374 | |
61947107 |
375 | C<perl -d:Module=arg,arg,arg> now works (previously one couldn't pass |
376 | in multiple arguments.) |
f39f21d8 |
377 | |
378 | =item * |
379 | |
61947107 |
380 | END blocks are now run even if you exit/die in a BEGIN block. |
381 | Internally, the execution of END blocks is now controlled by |
382 | PL_exit_flags & PERL_EXIT_DESTRUCT_END. This enables the new |
383 | behaviour for Perl embedders. This will default in 5.10. See |
384 | L<perlembed>. |
f39f21d8 |
385 | |
386 | =item * |
387 | |
e1f170bd |
388 | Formats now support zero-padded decimal fields. |
f39f21d8 |
389 | |
390 | =item * |
391 | |
77c8cf41 |
392 | Lvalue subroutines can now return C<undef> in list context. |
44da0e71 |
393 | However, the lvalue subroutine feature still remains experimental. |
f39f21d8 |
394 | |
395 | =item * |
396 | |
61947107 |
397 | A new special regular expression variable has been introduced: |
398 | C<$^N>, which contains the most-recently closed group (submatch). |
f39f21d8 |
399 | |
400 | =item * |
401 | |
61947107 |
402 | C<no Module;> now works even if there is no "sub unimport" in the Module. |
f39f21d8 |
403 | |
404 | =item * |
405 | |
61947107 |
406 | The numerical comparison operators return C<undef> if either operand |
407 | is a NaN. Previously the behaviour was unspecified. |
f39f21d8 |
408 | |
409 | =item * |
410 | |
e1f170bd |
411 | The following builtin functions are now overridable: each(), keys(), |
412 | pop(), push(), shift(), splice(), unshift(). |
413 | |
414 | =item * |
415 | |
61947107 |
416 | C<pack('U0a*', ...)> can now be used to force a string to UTF8. |
f39f21d8 |
417 | |
418 | =item * |
419 | |
61947107 |
420 | my __PACKAGE__ $obj now works. |
f39f21d8 |
421 | |
422 | =item * |
423 | |
e1f170bd |
424 | The printf() and sprintf() now support parameter reordering using the |
425 | C<%\d+\$> and C<*\d+\$> syntaxes. For example |
426 | |
427 | print "%2\$s %1\$s\n", "foo", "bar"; |
428 | |
da6838c8 |
429 | will print "bar foo\n". This feature helps in writing |
430 | internationalised software, and in general when the order |
431 | of the parameters can vary. |
f39f21d8 |
432 | |
433 | =item * |
434 | |
e1f170bd |
435 | prototype(\&) is now available. |
61947107 |
436 | |
437 | =item * |
438 | |
e1f170bd |
439 | prototype(\[$@%&]) is now available to implicitly create references |
440 | (useful for example if you want to emulate the tie() interface). |
61947107 |
441 | |
442 | =item * |
443 | |
44da0e71 |
444 | untie() will now call an UNTIE() hook if it exists. See L<perltie> |
445 | for details. |
61947107 |
446 | |
447 | =item * |
448 | |
449 | L<utime> now supports C<utime undef, undef, @files> to change the |
450 | file timestamps to the current time. |
451 | |
452 | =item * |
453 | |
e1f170bd |
454 | The rules for allowing underscores (underbars) in numeric constants |
455 | have been relaxed and simplified: now you can have an underscore |
456 | simply B<between digits>. |
f39f21d8 |
457 | |
458 | =back |
459 | |
77c8cf41 |
460 | =head1 Modules and Pragmata |
f39f21d8 |
461 | |
1e13d81f |
462 | =head2 New Modules and Pragmata |
f39f21d8 |
463 | |
464 | =over 4 |
465 | |
466 | =item * |
467 | |
0e9b9e0c |
468 | C<Attribute::Handlers> allows a class to define attribute handlers. |
469 | |
470 | package MyPack; |
471 | use Attribute::Handlers; |
472 | sub Wolf :ATTR(SCALAR) { print "howl!\n" } |
473 | |
474 | # later, in some package using or inheriting from MyPack... |
475 | |
476 | my MyPack $Fluffy : Wolf; # the attribute handler Wolf will be called |
477 | |
478 | Both variables and routines can have attribute handlers. Handlers can |
479 | be specific to type (SCALAR, ARRAY, HASH, or CODE), or specific to the |
480 | exact compilation phase (BEGIN, CHECK, INIT, or END). |
481 | |
482 | =item * |
483 | |
61947107 |
484 | B<B::Concise> is a new compiler backend for walking the Perl syntax |
485 | tree, printing concise info about ops, from Stephen McCamant. The |
486 | output is highly customisable. See L<B::Concise>. |
f39f21d8 |
487 | |
488 | =item * |
489 | |
61947107 |
490 | C<Class::ISA> for reporting the search path for a class's ISA tree, |
491 | by Sean Burke, has been added. See L<Class::ISA>. |
f39f21d8 |
492 | |
493 | =item * |
494 | |
61947107 |
495 | C<Cwd> has now a split personality: if possible, an XS extension is |
496 | used, (this will hopefully be faster, more secure, and more robust) |
497 | but if not possible, the familiar Perl implementation is used. |
f39f21d8 |
498 | |
499 | =item * |
500 | |
e1f170bd |
501 | C<Devel::PPPort>, originally from Kenneth Albanowski and now |
502 | maintained by Paul Marquess, has been added. It is primarily used |
503 | by C<h2xs> to enhance portability of of XS modules between different |
504 | versions of Perl. |
1e13d81f |
505 | |
506 | =item * |
507 | |
61947107 |
508 | C<Digest>, frontend module for calculating digests (checksums), from |
509 | Gisle Aas, has been added. See L<Digest>. |
f39f21d8 |
510 | |
511 | =item * |
512 | |
61947107 |
513 | C<Digest::MD5> for calculating MD5 digests (checksums) as defined in |
514 | RFC 1321, from Gisle Aas, has been added. See L<Digest::MD5>. |
f39f21d8 |
515 | |
516 | use Digest::MD5 'md5_hex'; |
517 | |
518 | $digest = md5_hex("Thirsty Camel"); |
519 | |
520 | print $digest, "\n"; # 01d19d9d2045e005c3f1b80e8b164de1 |
521 | |
61947107 |
522 | NOTE: the C<MD5> backward compatibility module is deliberately not |
e1f170bd |
523 | included since its further use is discouraged. |
f39f21d8 |
524 | |
f39f21d8 |
525 | =item * |
526 | |
61947107 |
527 | C<Encode>, by Nick Ing-Simmons, provides a mechanism to translate |
f39f21d8 |
528 | between different character encodings. Support for Unicode, |
529 | ISO-8859-*, ASCII, CP*, KOI8-R, and three variants of EBCDIC are |
530 | compiled in to the module. Several other encodings (like Japanese, |
531 | Chinese, and MacIntosh encodings) are included and will be loaded at |
61947107 |
532 | runtime. See L<Encode>. |
f39f21d8 |
533 | |
534 | Any encoding supported by Encode module is also available to the |
535 | ":encoding()" layer if PerlIO is used. |
536 | |
61947107 |
537 | =item * |
538 | |
539 | C<I18N::Langinfo> can be use to query locale information. |
540 | See L<I18N::Langinfo>. |
f39f21d8 |
541 | |
542 | =item * |
543 | |
61947107 |
544 | C<I18N::LangTags> has functions for dealing with RFC3066-style |
bea4d472 |
545 | language tags, by Sean Burke. See L<I18N::LangTags>. |
61947107 |
546 | |
547 | =item * |
548 | |
549 | C<ExtUtils::Constant> is a new tool for extension writers for |
550 | generating XS code to import C header constants, by Nicholas Clark. |
551 | See L<ExtUtils::Constant>. |
552 | |
553 | =item * |
554 | |
555 | C<Filter::Simple> is an easy-to-use frontend to Filter::Util::Call, |
556 | from Damian Conway. See L<Filter::Simple>. |
f39f21d8 |
557 | |
558 | # in MyFilter.pm: |
559 | |
560 | package MyFilter; |
561 | |
562 | use Filter::Simple sub { |
563 | while (my ($from, $to) = splice @_, 0, 2) { |
564 | s/$from/$to/g; |
565 | } |
566 | }; |
567 | |
568 | 1; |
569 | |
570 | # in user's code: |
571 | |
572 | use MyFilter qr/red/ => 'green'; |
573 | |
574 | print "red\n"; # this code is filtered, will print "green\n" |
575 | print "bored\n"; # this code is filtered, will print "bogreen\n" |
576 | |
577 | no MyFilter; |
578 | |
579 | print "red\n"; # this code is not filtered, will print "red\n" |
580 | |
61947107 |
581 | =item * |
582 | |
583 | C<File::Temp> allows one to create temporary files and directories in |
584 | an easy, portable, and secure way, by Tim Jenness. See L<File::Temp>. |
585 | |
586 | =item * |
587 | |
588 | C<Filter::Util::Call> provides you with the framework to write |
589 | I<Source Filters> in Perl, from Paul Marquess. For most uses the |
590 | frontend Filter::Simple is to be preferred. See L<Filter::Util::Call>. |
591 | |
592 | =item * |
593 | |
594 | L<libnet> is a collection of perl5 modules related to network |
595 | programming, from Graham Barr. See L<Net::FTP>, L<Net::NNTP>, |
596 | L<Net::Ping>, L<Net::POP3>, L<Net::SMTP>, and L<Net::Time>. |
597 | |
598 | Perl installation leaves libnet unconfigured, use F<libnetcfg> to configure. |
f39f21d8 |
599 | |
600 | =item * |
601 | |
61947107 |
602 | C<List::Util> is a selection of general-utility list subroutines, like |
bea4d472 |
603 | sum(), min(), first(), and shuffle(), by Graham Barr. See L<List::Util>. |
f39f21d8 |
604 | |
605 | =item * |
606 | |
61947107 |
607 | C<Locale::Constants>, C<Locale::Country>, C<Locale::Currency>, and |
608 | C<Locale::Language>, from Neil Bowers, have been added. They provide the |
609 | codes for various locale standards, such as "fr" for France, "usd" for |
610 | US Dollar, and "jp" for Japanese. |
f39f21d8 |
611 | |
612 | use Locale::Country; |
613 | |
614 | $country = code2country('jp'); # $country gets 'Japan' |
615 | $code = country2code('Norway'); # $code gets 'no' |
616 | |
617 | See L<Locale::Constants>, L<Locale::Country>, L<Locale::Currency>, |
61947107 |
618 | and L<Locale::Language>. |
619 | |
620 | =item * |
621 | |
622 | C<Locale::Maketext> is localization framework from Sean Burke. See |
623 | L<Locale::Maketext>, and L<Locale::Maketext::TPJ13>. The latter is an |
624 | article about software localization, originally published in The Perl |
625 | Journal #13, republished here with kind permission. |
626 | |
627 | =item * |
628 | |
629 | C<Memoize> can make your functions faster by trading space for time, |
630 | from Mark-Jason Dominus. See L<Memoize>. |
f39f21d8 |
631 | |
632 | =item * |
633 | |
61947107 |
634 | C<MIME::Base64> allows you to encode data in base64, from Gisle Aas, |
635 | as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail |
636 | Extensions)>. |
f39f21d8 |
637 | |
638 | use MIME::Base64; |
639 | |
640 | $encoded = encode_base64('Aladdin:open sesame'); |
641 | $decoded = decode_base64($encoded); |
642 | |
643 | print $encoded, "\n"; # "QWxhZGRpbjpvcGVuIHNlc2FtZQ==" |
644 | |
61947107 |
645 | See L<MIME::Base64>. |
f39f21d8 |
646 | |
647 | =item * |
648 | |
61947107 |
649 | C<MIME::QuotedPrint> allows you to encode data in quoted-printable |
650 | encoding, as defined in RFC 2045 - I<MIME (Multipurpose Internet Mail |
651 | Extensions)>, from Gisle Aas. |
f39f21d8 |
652 | |
653 | use MIME::QuotedPrint; |
654 | |
655 | $encoded = encode_qp("Smiley in Unicode: \x{263a}"); |
656 | $decoded = decode_qp($encoded); |
657 | |
658 | print $encoded, "\n"; # "Smiley in Unicode: =263A" |
659 | |
660 | MIME::QuotedPrint has been enhanced to provide the basic methods |
661 | necessary to use it with PerlIO::Via as in : |
662 | |
663 | use MIME::QuotedPrint; |
057b7f2b |
664 | open($fh,">Via(MIME::QuotedPrint)",$path); |
f39f21d8 |
665 | |
61947107 |
666 | See L<MIME::QuotedPrint>. |
f39f21d8 |
667 | |
668 | =item * |
669 | |
61947107 |
670 | C<NEXT> is pseudo-class for method redispatch, from Damian Conway. |
671 | See L<NEXT>. |
f39f21d8 |
672 | |
673 | =item * |
674 | |
1e13d81f |
675 | C<open> is a new pragma for setting the default I/O disciplines |
676 | for open(). |
677 | |
678 | =item * |
679 | |
61947107 |
680 | C<PerlIO::Scalar> provides the implementation of IO to "in memory" |
681 | Perl scalars as discussed above, from Nick Ing-Simmons. It also |
682 | serves as an example of a loadable PerlIO layer. Other future |
683 | possibilities include PerlIO::Array and PerlIO::Code. |
684 | See L<PerlIO::Scalar>. |
685 | |
686 | =item * |
687 | |
688 | C<PerlIO::Via> acts as a PerlIO layer and wraps PerlIO layer |
689 | functionality provided by a class (typically implemented in perl |
690 | code), from Nick Ing-Simmons. |
f39f21d8 |
691 | |
692 | use MIME::QuotedPrint; |
057b7f2b |
693 | open($fh,">Via(MIME::QuotedPrint)",$path); |
f39f21d8 |
694 | |
695 | This will automatically convert everything output to C<$fh> |
61947107 |
696 | to Quoted-Printable. See L<PerlIO::Via>. |
f39f21d8 |
697 | |
698 | =item * |
699 | |
1e13d81f |
700 | C<Pod::ParseLink>, by Russ Allbery, has been added, |
95f0a2f1 |
701 | to parse LZ<><> links in pods as described in the new |
1e13d81f |
702 | perlpodspec. |
703 | |
704 | =item * |
705 | |
61947107 |
706 | C<Pod::Text::Overstrike>, by Joe Smith, has been added. |
f39f21d8 |
707 | It converts POD data to formatted overstrike text. |
61947107 |
708 | See L<Pod::Text::Overstrike>. |
f39f21d8 |
709 | |
710 | =item * |
711 | |
61947107 |
712 | C<Scalar::Util> is a selection of general-utility scalar subroutines, |
713 | like blessed(), reftype(), and tainted(). See L<Scalar::Util>. |
714 | |
715 | =item * |
716 | |
1e13d81f |
717 | C<sort> is a new pragma for controlling the behaviour of sort(). |
718 | |
719 | =item * |
720 | |
61947107 |
721 | C<Storable> gives persistence to Perl data structures by allowing the |
722 | storage and retrieval of Perl data to and from files in a fast and |
723 | compact binary format, from Raphael Manfredi. See L<Storable>. |
724 | |
725 | =item * |
726 | |
727 | C<Switch>, from Damian Conway, has been added. Just by saying |
f39f21d8 |
728 | |
729 | use Switch; |
730 | |
731 | you have C<switch> and C<case> available in Perl. |
732 | |
733 | use Switch; |
734 | |
735 | switch ($val) { |
736 | |
737 | case 1 { print "number 1" } |
738 | case "a" { print "string a" } |
739 | case [1..10,42] { print "number in list" } |
740 | case (@array) { print "number in list" } |
741 | case /\w+/ { print "pattern" } |
742 | case qr/\w+/ { print "pattern" } |
743 | case (%hash) { print "entry in hash" } |
744 | case (\%hash) { print "entry in hash" } |
745 | case (\&sub) { print "arg to subroutine" } |
746 | else { print "previous case not true" } |
747 | } |
748 | |
61947107 |
749 | See L<Switch>. |
750 | |
751 | =item * |
752 | |
753 | C<Test::More> is yet another framework for writing test scripts, |
754 | more extensive than Test::Simple, by Michael Schwern. See L<Test::More>. |
755 | |
756 | =item * |
757 | |
aecce728 |
758 | C<Test::Simple> has basic utilities for writing tests, by Michael |
61947107 |
759 | Schwern. See L<Test::Simple>. |
77c8cf41 |
760 | |
761 | =item * |
762 | |
61947107 |
763 | C<Text::Balanced> has been added, for extracting delimited text |
764 | sequences from strings, from Damian Conway. |
77c8cf41 |
765 | |
766 | use Text::Balanced 'extract_delimited'; |
767 | |
768 | ($a, $b) = extract_delimited("'never say never', he never said", "'", ''); |
769 | |
770 | $a will be "'never say never'", $b will be ', he never said'. |
771 | |
772 | In addition to extract_delimited() there are also extract_bracketed(), |
773 | extract_quotelike(), extract_codeblock(), extract_variable(), |
774 | extract_tagged(), extract_multiple(), gen_delimited_pat(), and |
775 | gen_extract_tagged(). With these you can implement rather advanced |
61947107 |
776 | parsing algorithms. See L<Text::Balanced>. |
77c8cf41 |
777 | |
778 | =item * |
779 | |
c2e23569 |
780 | C<threads> is an interface to interpreter threads, by Arthur Bergman. |
61947107 |
781 | Interpreter threads (ithreads) is the new thread model introduced in |
c2e23569 |
782 | Perl 5.6 but only available as an internal interface for extension |
783 | writers (and for Win32 Perl for C<fork()> emulation). See L<threads>. |
77c8cf41 |
784 | |
785 | =item * |
786 | |
61947107 |
787 | C<threads::shared> allows data sharing for interpreter threads, from |
788 | Arthur Bergman. In the ithreads model any data sharing between |
789 | threads must be explicit, as opposed to the old 5.005 thread model |
790 | where data sharing was implicit. See L<threads::shared>. |
77c8cf41 |
791 | |
792 | =item * |
793 | |
61947107 |
794 | C<Tie::RefHash::Nestable>, by Edward Avis, allows storing hash |
ba370e9b |
795 | references (unlike the standard Tie::RefHash) The module is contained |
796 | within Tie::RefHash, see L<Tie::RefHash>. |
77c8cf41 |
797 | |
798 | =item * |
799 | |
61947107 |
800 | C<Time::HiRes> provides high resolution timing (ualarm, usleep, |
801 | and gettimeofday), from Douglas E. Wegscheid. See L<Time::HiRes>. |
77c8cf41 |
802 | |
803 | =item * |
804 | |
61947107 |
805 | C<Unicode::UCD> offers a querying interface to the Unicode Character |
806 | Database. See L<Unicode::UCD>. |
77c8cf41 |
807 | |
808 | =item * |
809 | |
61947107 |
810 | C<Unicode::Collate> implements the UCA (Unicode Collation Algorithm) |
811 | for sorting Unicode strings, by SADAHIRO Tomoyuki. See L<Unicode::Collate>. |
77c8cf41 |
812 | |
813 | =item * |
814 | |
61947107 |
815 | C<Unicode::Normalize> implements the various Unicode normalization |
816 | forms, by SADAHIRO Tomoyuki. See L<Unicode::Normalize>. |
77c8cf41 |
817 | |
818 | =item * |
819 | |
61947107 |
820 | C<XS::Typemap>, by Tim Jenness, is a test extension that exercises XS |
821 | typemaps. Nothing gets installed but for extension writers the code |
822 | is worth studying. |
77c8cf41 |
823 | |
824 | =back |
825 | |
826 | =head2 Updated And Improved Modules and Pragmata |
827 | |
828 | =over 4 |
829 | |
830 | =item * |
831 | |
61947107 |
832 | The following independently supported modules have been updated to the |
833 | newest versions from CPAN: CGI, CPAN, DB_File, File::Spec, File::Temp, |
834 | Getopt::Long, Math::BigFloat, Math::BigInt, the podlators bundle |
835 | (Pod::Man, Pod::Text), Pod::LaTeX, Pod::Parser, Storable, |
836 | Term::ANSIColor, Test, Text-Tabs+Wrap. |
77c8cf41 |
837 | |
838 | =item * |
839 | |
61947107 |
840 | The attributes::reftype() now works on tied arguments. |
77c8cf41 |
841 | |
842 | =item * |
843 | |
057b7f2b |
844 | AutoLoader can now be disabled with C<no AutoLoader;>. |
77c8cf41 |
845 | |
846 | =item * |
847 | |
1e13d81f |
848 | B::Deparse has been significantly enhanced. It now can deparse almost |
849 | all of the standard test suite (so that the tests still succeed). |
850 | There is a make target "test.deparse" for trying this out. |
77c8cf41 |
851 | |
852 | =item * |
853 | |
1e13d81f |
854 | Class::Struct can now define the classes in compile time. |
77c8cf41 |
855 | |
856 | =item * |
857 | |
1e13d81f |
858 | Class::Struct now assigns the array/hash element if the accessor |
859 | is called with an array/hash element as the B<sole> argument. |
77c8cf41 |
860 | |
861 | =item * |
862 | |
1e13d81f |
863 | Data::Dumper has now an option to sort hashes. |
77c8cf41 |
864 | |
865 | =item * |
866 | |
1e13d81f |
867 | Data::Dumper has now an option to dump code references |
868 | using B::Deparse. |
77c8cf41 |
869 | |
870 | =item * |
871 | |
44da0e71 |
872 | DB_File now supports newer Berkeley DB versions, among |
873 | other improvements. |
874 | |
875 | =item * |
876 | |
1e13d81f |
877 | The English module can now be used without the infamous performance |
878 | hit by saying |
77c8cf41 |
879 | |
1e13d81f |
880 | use English '-no_performance_hit'; |
77c8cf41 |
881 | |
1e13d81f |
882 | (Assuming, of course, that one doesn't need the troublesome variables |
883 | C<$`>, C<$&>, or C<$'>.) Also, introduced C<@LAST_MATCH_START> and |
884 | C<@LAST_MATCH_END> English aliases for C<@-> and C<@+>. |
77c8cf41 |
885 | |
886 | =item * |
887 | |
1e13d81f |
888 | Fcntl, Socket, and Sys::Syslog have been rewritten to use the |
889 | new-style constant dispatch section (see L<ExtUtils::Constant>). |
890 | This means that they will be more robust and hopefully faster. |
77c8cf41 |
891 | |
892 | =item * |
893 | |
44da0e71 |
894 | File::Find now chdir()s correctly when chasing symbolic links. |
895 | |
896 | =item * |
897 | |
1e13d81f |
898 | File::Find now has pre- and post-processing callbacks. It also |
899 | correctly changes directories when chasing symbolic links. Callbacks |
900 | (naughtily) exiting with "next;" instead of "return;" now work. |
61947107 |
901 | |
902 | =item * |
903 | |
1e13d81f |
904 | File::Find is now (again) reentrant. It also has been made |
905 | more portable. |
77c8cf41 |
906 | |
61947107 |
907 | =item * |
908 | |
1e13d81f |
909 | File::Glob::glob() renamed to File::Glob::bsd_glob() to avoid |
910 | prototype mismatch with CORE::glob(). |
61947107 |
911 | |
912 | =item * |
913 | |
914 | File::Glob now supports C<GLOB_LIMIT> constant to limit the size of |
915 | the returned list of filenames. |
77c8cf41 |
916 | |
917 | =item * |
918 | |
919 | Devel::Peek now has an interface for the Perl memory statistics |
920 | (this works only if you are using perl's malloc, and if you have |
921 | compiled with debugging). |
922 | |
923 | =item * |
924 | |
1e13d81f |
925 | IPC::Open3 now allows the use of numeric file descriptors. |
926 | |
927 | =item * |
928 | |
77c8cf41 |
929 | IO::Socket has now atmark() method, which returns true if the socket |
930 | is positioned at the out-of-band mark. The method is also exportable |
931 | as a sockatmark() function. |
932 | |
933 | =item * |
934 | |
935 | IO::Socket::INET has support for ReusePort option (if your platform |
936 | supports it). The Reuse option now has an alias, ReuseAddr. For clarity |
937 | you may want to prefer ReuseAddr. |
938 | |
939 | =item * |
940 | |
61947107 |
941 | IO::Socket::INET now supports C<LocalPort> of zero (usually meaning |
942 | that the operating system will make one up.) |
77c8cf41 |
943 | |
944 | =item * |
945 | |
1e13d81f |
946 | use lib now works identically to @INC. Removing directories |
947 | with 'no lib' now works. |
948 | |
949 | =item * |
950 | |
951 | Math::BigFloat and Math::BigInt have undergone a full rewrite. |
952 | They are now magnitudes faster, and they support various |
61947107 |
953 | bignum libraries such as GMP and PARI as their backends. |
f39f21d8 |
954 | |
955 | =item * |
956 | |
44da0e71 |
957 | Math::Complex handles inf, NaN etc., better. |
958 | |
959 | =item * |
960 | |
61947107 |
961 | Net::Ping has been enhanced. There is now "external" protocol which |
962 | uses Net::Ping::External module which runs external ping(1) and parses |
44da0e71 |
963 | the output. A version of Net::Ping::External is available in CPAN. |
f39f21d8 |
964 | |
77c8cf41 |
965 | =item * |
f39f21d8 |
966 | |
da6838c8 |
967 | POSIX::sigaction() is now much more flexible and robust. |
61947107 |
968 | You can now install coderef handlers, 'DEFAULT', and 'IGNORE' |
969 | handlers, installing new handlers was not atomic. |
f39f21d8 |
970 | |
971 | =item * |
972 | |
da6838c8 |
973 | In Safe the C<%INC> now localised in a Safe compartment so that |
76663d67 |
974 | use/require work. |
975 | |
976 | =item * |
977 | |
44da0e71 |
978 | In SDBM_File on dosish platforms, some keys went missing because of |
979 | lack of support for files with "holes". A workaround for the problem |
980 | has been added. |
981 | |
982 | =item * |
983 | |
da6838c8 |
984 | In Search::Dict one can now have a pre-processing hook for the |
76663d67 |
985 | lines being searched. |
1e13d81f |
986 | |
987 | =item * |
988 | |
989 | The Shell module now has an OO interface. |
990 | |
991 | =item * |
992 | |
61947107 |
993 | The Test module has been significantly enhanced. |
f39f21d8 |
994 | |
995 | =item * |
996 | |
da6838c8 |
997 | The vars pragma now supports declaring fully qualified variables. |
77c8cf41 |
998 | (Something that C<our()> does not and will not support.) |
f39f21d8 |
999 | |
888aee59 |
1000 | =item * |
1001 | |
61947107 |
1002 | The utf8:: name space (as in the pragma) provides various |
1003 | Perl-callable functions to provide low level access to Perl's |
1004 | internal Unicode representation. At the moment only length() |
1005 | has been implemented. |
888aee59 |
1006 | |
f39f21d8 |
1007 | =back |
1008 | |
77c8cf41 |
1009 | =head1 Utility Changes |
f39f21d8 |
1010 | |
1011 | =over 4 |
1012 | |
1013 | =item * |
1014 | |
61947107 |
1015 | Emacs perl mode (emacs/cperl-mode.el) has been updated to version |
77c8cf41 |
1016 | 4.31. |
f39f21d8 |
1017 | |
1018 | =item * |
1019 | |
61947107 |
1020 | F<emacs/e2ctags.pl> is now much faster. |
f39f21d8 |
1021 | |
1022 | =item * |
1023 | |
1e13d81f |
1024 | C<h2ph> now supports C trigraphs. |
1025 | |
1026 | =item * |
1027 | |
1028 | C<h2xs> now produces a template README. |
f39f21d8 |
1029 | |
77c8cf41 |
1030 | =item * |
1031 | |
1e13d81f |
1032 | C<h2xs> now uses C<Devel::PPort> for better portability between |
1033 | different versions of Perl. |
f39f21d8 |
1034 | |
1035 | =item * |
1036 | |
1e13d81f |
1037 | C<h2xs> uses the new L<ExtUtils::Constant> module which will affect |
61947107 |
1038 | newly created extensions that define constants. Since the new code is |
1039 | more correct (if you have two constants where the first one is a |
1040 | prefix of the second one, the first constant B<never> gets defined), |
1041 | less lossy (it uses integers for integer constant, as opposed to the |
1042 | old code that used floating point numbers even for integer constants), |
1043 | and slightly faster, you might want to consider regenerating your |
1044 | extension code (the new scheme makes regenerating easy). |
1045 | L<h2xs> now also supports C trigraphs. |
f39f21d8 |
1046 | |
1047 | =item * |
1048 | |
1e13d81f |
1049 | C<libnetcfg> has been added to configure the libnet. |
f39f21d8 |
1050 | |
1051 | =item * |
1052 | |
1e13d81f |
1053 | C<perlbug> is now much more robust. It also sends the bug report to |
61947107 |
1054 | perl.org, not perl.com. |
f39f21d8 |
1055 | |
1056 | =item * |
1057 | |
1e13d81f |
1058 | C<perlcc> has been rewritten and its user interface (that is, |
61947107 |
1059 | command line) is much more like that of the UNIX C compiler, cc. |
44da0e71 |
1060 | (The perlbc tools has been removed. Use C<perlcc -B> instead.) |
f39f21d8 |
1061 | |
1062 | =item * |
1063 | |
aecce728 |
1064 | C<perlivp> is a new Installation Verification Procedure utility |
1065 | for running any time after installing Perl. |
f39f21d8 |
1066 | |
1067 | =item * |
1068 | |
1e13d81f |
1069 | C<pod2html> now allows specifying a cache directory. |
f39f21d8 |
1070 | |
1071 | =item * |
1072 | |
1e13d81f |
1073 | C<s2p> has been completely rewritten in Perl. (It is in fact a full |
1074 | implementation of sed in Perl: you can use the sed functionality by |
1075 | using the C<psed> utility.) |
61947107 |
1076 | |
1077 | =item * |
1078 | |
1e13d81f |
1079 | C<xsubpp> now understands POD documentation embedded in the *.xs files. |
f39f21d8 |
1080 | |
1081 | =item * |
1082 | |
1e13d81f |
1083 | C<xsubpp> now supports OUT keyword. |
f39f21d8 |
1084 | |
1085 | =back |
1086 | |
77c8cf41 |
1087 | =head1 New Documentation |
f39f21d8 |
1088 | |
1089 | =over 4 |
1090 | |
1091 | =item * |
1092 | |
77c8cf41 |
1093 | perl56delta details the changes between the 5.005 release and the |
1094 | 5.6.0 release. |
f39f21d8 |
1095 | |
1096 | =item * |
1097 | |
61947107 |
1098 | perlclib documents the internal replacements for standard C library |
1099 | functions. (Interesting only for extension writers and Perl core |
1100 | hackers.) |
1101 | |
1102 | =item * |
1103 | |
77c8cf41 |
1104 | perldebtut is a Perl debugging tutorial. |
f39f21d8 |
1105 | |
77c8cf41 |
1106 | =item * |
f39f21d8 |
1107 | |
77c8cf41 |
1108 | perlebcdic contains considerations for running Perl on EBCDIC platforms. |
f39f21d8 |
1109 | |
77c8cf41 |
1110 | =item * |
1111 | |
888aee59 |
1112 | perlintro is a gentle introduction to Perl. |
1113 | |
1114 | =item * |
1115 | |
61947107 |
1116 | perliol documents the internals of PerlIO with layers. |
1117 | |
1118 | =item * |
1119 | |
888aee59 |
1120 | perlmodstyle is a style guide for writing modules. |
1121 | |
1122 | =item * |
1123 | |
77c8cf41 |
1124 | perlnewmod tells about writing and submitting a new module. |
f39f21d8 |
1125 | |
1126 | =item * |
1127 | |
34babc16 |
1128 | perlpacktut is a pack() tutorial. |
1129 | |
1130 | =item * |
1131 | |
888aee59 |
1132 | perlpod has been rewritten to be clearer and to record the best |
1133 | practices gathered over the years. |
1134 | |
1135 | =item * |
1136 | |
057b7f2b |
1137 | perlpodspec is a more formal specification of the pod format, |
888aee59 |
1138 | mainly of interest for writers of pod applications, not to |
1139 | people writing in pod. |
1140 | |
1141 | =item * |
1142 | |
77c8cf41 |
1143 | perlretut is a regular expression tutorial. |
f39f21d8 |
1144 | |
1145 | =item * |
1146 | |
77c8cf41 |
1147 | perlrequick is a regular expressions quick-start guide. |
1148 | Yes, much quicker than perlretut. |
f39f21d8 |
1149 | |
77c8cf41 |
1150 | =item * |
f39f21d8 |
1151 | |
61947107 |
1152 | perltodo has been updated. |
1153 | |
1154 | =item * |
1155 | |
888aee59 |
1156 | perltootc has been renamed as perltooc (to not to conflict |
61947107 |
1157 | with perltoot in filesystems restricted to "8.3" names) |
888aee59 |
1158 | |
1159 | =item * |
1160 | |
1161 | perluniintro is an introduction to using Unicode in Perl |
1162 | (perlunicode is more of a reference) |
1163 | |
1164 | =item * |
1165 | |
77c8cf41 |
1166 | perlutil explains the command line utilities packaged with the Perl |
1167 | distribution. |
1168 | |
1169 | =back |
f39f21d8 |
1170 | |
61947107 |
1171 | The following platform-specific documents are available before |
1172 | the installation as README.I<platform>, and after the installation |
1173 | as perlI<platform>: |
f39f21d8 |
1174 | |
61947107 |
1175 | perlaix perlamiga perlapollo perlbeos perlbs2000 |
1176 | perlce perlcygwin perldgux perldos perlepoc perlhpux |
1177 | perlhurd perlmachten perlmacos perlmint perlmpeix |
1178 | perlnetware perlos2 perlos390 perlplan9 perlqnx perlsolaris |
1179 | perltru64 perluts perlvmesa perlvms perlvos perlwin32 |
77c8cf41 |
1180 | |
1181 | =over 4 |
1182 | |
1183 | =item * |
1184 | |
61947107 |
1185 | The documentation for the POSIX-BC platform is called "BS2000", to avoid |
1186 | confusion with the Perl POSIX module. |
77c8cf41 |
1187 | |
1188 | =item * |
1189 | |
61947107 |
1190 | The documentation for the WinCE platform is called "CE", to avoid |
1191 | confusion with the perlwin32 documentation on 8.3-restricted filesystems. |
77c8cf41 |
1192 | |
1193 | =back |
1194 | |
1195 | =head1 Performance Enhancements |
1196 | |
1197 | =over 4 |
1198 | |
1199 | =item * |
1200 | |
44da0e71 |
1201 | map() could get pathologically slow when the result list it generates |
1202 | is larger than the source list. The performance has been improved for |
1203 | common scenarios. |
77c8cf41 |
1204 | |
1205 | =item * |
1206 | |
e1f170bd |
1207 | sort() has been changed to use primarily mergesort internally as |
1208 | opposed to the earlier quicksort. For very small lists this may |
1209 | result in slightly slower sorting times, but in general the speedup |
1210 | should be at least 20%. Additional bonuses are that the worst case |
1211 | behaviour of sort() is now better (in computer science terms it now |
1212 | runs in time O(N log N), as opposed to quicksort's Theta(N**2) |
1213 | worst-case run time behaviour), and that sort() is now stable |
1214 | (meaning that elements with identical keys will stay ordered as they |
1215 | were before the sort). See the C<sort> pragma for information. |
77c8cf41 |
1216 | |
05e25c75 |
1217 | The story in more detail: suppose you want to serve yourself a little |
1218 | slice of Pi. |
1219 | |
1220 | @digits = ( 3,1,4,1,5,9 ); |
1221 | |
1222 | A numerical sort of the digits will yield (1,1,3,4,5,9), as expected. |
1223 | Which C<1> comes first is hard to know, since one C<1> looks pretty |
1224 | much like any other. You can regard this as totally trivial, |
1225 | or somewhat profound. However, if you just want to sort the even |
1226 | digits ahead of the odd ones, then what will |
1227 | |
1228 | sort { ($a % 2) <=> ($b % 2) } @digits; |
1229 | |
1230 | yield? The only even digit, C<4>, will come first. But how about |
1231 | the odd numbers, which all compare equal? With the quicksort algorithm |
1232 | used to implement Perl 5.6 and earlier, the order of ties is left up |
1233 | to the sort. So, as you add more and more digits of Pi, the order |
1234 | in which the sorted even and odd digits appear will change. |
1235 | and, for sufficiently large slices of Pi, the quicksort algorithm |
1236 | in Perl 5.8 won't return the same results even if reinvoked with the |
1237 | same input. The justification for this rests with quicksort's |
1238 | worst case behavior. If you run |
1239 | |
1240 | sort { $a <=> $b } ( 1 .. $N , 1 .. $N ); |
1241 | |
1242 | (something you might approximate if you wanted to merge two sorted |
1243 | arrays using sort), doubling $N doesn't just double the quicksort time, |
1244 | it I<quadruples> it. Quicksort has a worst case run time that can |
1245 | grow like N**2, so-called I<quadratic> behaviour, and it can happen |
1246 | on patterns that may well arise in normal use. You won't notice this |
1247 | for small arrays, but you I<will> notice it with larger arrays, |
1248 | and you may not live long enough for the sort to complete on arrays |
1249 | of a million elements. So the 5.8 quicksort scrambles large arrays |
1250 | before sorting them, as a statistical defence against quadratic behaviour. |
1251 | But that means if you sort the same large array twice, ties may be |
1252 | broken in different ways. |
1253 | |
1254 | Because of the unpredictability of tie-breaking order, and the quadratic |
1255 | worst-case behaviour, quicksort was I<almost> replaced completely with |
1256 | a stable mergesort. I<Stable> means that ties are broken to preserve |
1257 | the original order of appearance in the input array. So |
1258 | |
1259 | sort { ($a % 2) <=> ($b % 2) } (3,1,4,1,5,9); |
1260 | |
1261 | will yield (4,3,1,1,5,9), guaranteed. The even and odd numbers |
1262 | appear in the output in the same order they appeared in the input. |
1263 | Mergesort has worst case O(NlogN) behaviour, the best value |
1264 | attainable. And, ironically, this mergesort does particularly |
1265 | well where quicksort goes quadratic: mergesort sorts (1..$N, 1..$N) |
1266 | in O(N) time. But quicksort was rescued at the last moment because |
1267 | it is faster than mergesort on certain inputs and platforms. |
1268 | For example, if you really I<don't> care about the order of even |
1269 | and odd digits, quicksort will run in O(N) time; it's very good |
1270 | at sorting many repetitions of a small number of distinct elements. |
1271 | The quicksort divide and conquer strategy works well on platforms |
1272 | with relatively small, very fast, caches. Eventually, the problem gets |
1273 | whittled down to one that fits in the cache, from which point it |
1274 | benefits from the increased memory speed. |
1275 | |
1276 | Quicksort was rescued by implementing a sort pragma to control aspects |
1277 | of the sort. The B<stable> subpragma forces stable behaviour, |
1278 | regardless of algorithm. The B<_quicksort> and B<_mergesort> |
1279 | subpragmas are heavy-handed ways to select the underlying implementation. |
1280 | The leading C<_> is a reminder that these subpragmas may not survive |
1281 | beyond 5.8. More appropriate mechanisms for selecting the implementation |
1282 | exist, but they wouldn't have arrived in time to save quicksort. |
1283 | |
77c8cf41 |
1284 | =item * |
1285 | |
1286 | Hashes now use Bob Jenkins "One-at-a-Time" hashing key algorithm |
1287 | (http://burtleburtle.net/bob/hash/doobs.html). This algorithm is |
1288 | reasonably fast while producing a much better spread of values than |
1289 | the old hashing algorithm (originally by Chris Torek, later tweaked by |
1290 | Ilya Zakharevich). Hash values output from the algorithm on a hash of |
1291 | all 3-char printable ASCII keys comes much closer to passing the |
1292 | DIEHARD random number generation tests. According to perlbench, this |
1293 | change has not affected the overall speed of Perl. |
1294 | |
1295 | =item * |
1296 | |
1297 | unshift() should now be noticeably faster. |
1298 | |
1299 | =back |
1300 | |
1301 | =head1 Installation and Configuration Improvements |
1302 | |
1303 | =head2 Generic Improvements |
1304 | |
1305 | =over 4 |
1306 | |
1307 | =item * |
1308 | |
1309 | INSTALL now explains how you can configure Perl to use 64-bit |
1310 | integers even on non-64-bit platforms. |
1311 | |
1312 | =item * |
1313 | |
1314 | Policy.sh policy change: if you are reusing a Policy.sh file |
1315 | (see INSTALL) and you use Configure -Dprefix=/foo/bar and in the old |
1316 | Policy $prefix eq $siteprefix and $prefix eq $vendorprefix, all of |
1317 | them will now be changed to the new prefix, /foo/bar. (Previously |
1318 | only $prefix changed.) If you do not like this new behaviour, |
1319 | specify prefix, siteprefix, and vendorprefix explicitly. |
1320 | |
1321 | =item * |
1322 | |
1323 | A new optional location for Perl libraries, otherlibdirs, is available. |
1324 | It can be used for example for vendor add-ons without disturbing Perl's |
1325 | own library directories. |
1326 | |
1327 | =item * |
1328 | |
1329 | In many platforms the vendor-supplied 'cc' is too stripped-down to |
1330 | build Perl (basically, 'cc' doesn't do ANSI C). If this seems |
1331 | to be the case and 'cc' does not seem to be the GNU C compiler |
1332 | 'gcc', an automatic attempt is made to find and use 'gcc' instead. |
1333 | |
1334 | =item * |
1335 | |
1336 | gcc needs to closely track the operating system release to avoid |
1337 | build problems. If Configure finds that gcc was built for a different |
1338 | operating system release than is running, it now gives a clearly visible |
1339 | warning that there may be trouble ahead. |
1340 | |
1341 | =item * |
1342 | |
1343 | If binary compatibility with the 5.005 release is not wanted, Configure |
1344 | no longer suggests including the 5.005 modules in @INC. |
1345 | |
1346 | =item * |
1347 | |
1348 | Configure C<-S> can now run non-interactively. |
1349 | |
1350 | =item * |
1351 | |
44da0e71 |
1352 | Configure support for pdp11-style memory models has been removed due |
1353 | to obsolescence. |
1354 | |
1355 | =item * |
1356 | |
77c8cf41 |
1357 | configure.gnu now works with options with whitespace in them. |
f39f21d8 |
1358 | |
77c8cf41 |
1359 | =item * |
f39f21d8 |
1360 | |
77c8cf41 |
1361 | installperl now outputs everything to STDERR. |
f39f21d8 |
1362 | |
77c8cf41 |
1363 | =item * |
1364 | |
1365 | $Config{byteorder} is now computed dynamically (this is more robust |
1366 | with "fat binaries" where an executable image contains binaries for |
1367 | more than one binary platform.) |
f39f21d8 |
1368 | |
1369 | =item * |
1370 | |
1371 | Because PerlIO is now the default on most platforms, "-perlio" doesn't |
1372 | get appended to the $Config{archname} (also known as $^O) anymore. |
1373 | Instead, if you explicitly choose not to use perlio (Configure command |
1374 | line option -Uuseperlio), you will get "-stdio" appended. |
1375 | |
1376 | =item * |
1377 | |
1378 | Another change related to the architecture name is that "-64all" |
1379 | (-Duse64bitall, or "maximally 64-bit") is appended only if your |
1380 | pointers are 64 bits wide. (To be exact, the use64bitall is ignored.) |
1381 | |
1382 | =item * |
1383 | |
77c8cf41 |
1384 | In AFS installations one can configure the root of the AFS to be |
1385 | somewhere else than the default F</afs> by using the Configure |
1386 | parameter C<-Dafsroot=/some/where/else>. |
1387 | |
1388 | =item * |
1389 | |
61947107 |
1390 | APPLLIB_EXP, a less-know configuration-time definition, has been |
1391 | documented. It can be used to prepend site-specific directories |
1392 | to Perl's default search path (@INC), see INSTALL for information. |
1393 | |
1394 | =item * |
1395 | |
77c8cf41 |
1396 | The version of Berkeley DB used when the Perl (and, presumably, the |
1397 | DB_File extension) was built is now available as |
1398 | C<@Config{qw(db_version_major db_version_minor db_version_patch)}> |
1399 | from Perl and as C<DB_VERSION_MAJOR_CFG DB_VERSION_MINOR_CFG |
1400 | DB_VERSION_PATCH_CFG> from C. |
1401 | |
1402 | =item * |
1403 | |
61947107 |
1404 | Building Berkeley DB3 for compatibility modes for DB, NDBM, and ODBM |
1405 | has been documented in INSTALL. |
77c8cf41 |
1406 | |
1407 | =item * |
1408 | |
61947107 |
1409 | If you have CPAN access (either network or a local copy such as a |
1410 | CD-ROM) you can during specify extra modules to Configure to build and |
1411 | install with Perl using the -Dextras=... option. See INSTALL for |
1412 | more details. |
f39f21d8 |
1413 | |
61947107 |
1414 | =item * |
f39f21d8 |
1415 | |
61947107 |
1416 | In addition to config.over a new override file, config.arch, is |
1417 | available. That is supposed to be used by hints file writers for |
1418 | architecture-wide changes (as opposed to config.over which is for |
1419 | site-wide changes). |
f39f21d8 |
1420 | |
1421 | =item * |
1422 | |
e1f170bd |
1423 | If your file system supports symbolic links you can build Perl outside |
1424 | of the source directory by |
1425 | |
1426 | mkdir /tmp/perl/build/directory |
1427 | cd /tmp/perl/build/directory |
1428 | sh /path/to/perl/source/Configure -Dmksymlinks ... |
1429 | |
1430 | This will create in /tmp/perl/build/directory a tree of symbolic links |
1431 | pointing to files in /path/to/perl/source. The original files are left |
1432 | unaffected. After Configure has finished you can just say |
1433 | |
1434 | make all test |
1435 | |
1436 | and Perl will be built and tested, all in /tmp/perl/build/directory. |
1437 | |
1438 | =item * |
1439 | |
61947107 |
1440 | For Perl developers several new make targets for profiling |
1441 | and debugging have been added, see L<perlhack>. |
1442 | |
1443 | =over 8 |
f39f21d8 |
1444 | |
1445 | =item * |
1446 | |
61947107 |
1447 | Use of the F<gprof> tool to profile Perl has been documented in |
1448 | L<perlhack>. There is a make target called "perl.gprof" for |
1449 | generating a gprofiled Perl executable. |
f39f21d8 |
1450 | |
1451 | =item * |
1452 | |
61947107 |
1453 | If you have GCC 3, there is a make target called "perl.gcov" for |
1454 | creating a gcoved Perl executable for coverage analysis. See |
1455 | L<perlhack>. |
f39f21d8 |
1456 | |
1457 | =item * |
1458 | |
61947107 |
1459 | If you are on IRIX or Tru64 platforms, new profiling/debugging options |
1460 | have been added, see L<perlhack> for more information about pixie and |
1461 | Third Degree. |
1462 | |
1463 | =back |
f39f21d8 |
1464 | |
1465 | =item * |
1466 | |
61947107 |
1467 | Guidelines of how to construct minimal Perl installations have |
1468 | been added to INSTALL. |
f39f21d8 |
1469 | |
1470 | =item * |
1471 | |
61947107 |
1472 | The Thread extension is now not built at all under ithreads |
1473 | (C<Configure -Duseithreads>) because it wouldn't work anyway (the |
1474 | Thread extension requires being Configured with C<-Duse5005threads>). |
f39f21d8 |
1475 | |
61947107 |
1476 | But note that the Thread.pm interface is now shared by both |
1477 | thread models. |
f39f21d8 |
1478 | |
61947107 |
1479 | =back |
f39f21d8 |
1480 | |
61947107 |
1481 | =head2 New Or Improved Platforms |
f39f21d8 |
1482 | |
61947107 |
1483 | For the list of platforms known to support Perl, |
1484 | see L<perlport/"Supported Platforms">. |
1485 | |
1486 | =over 4 |
f39f21d8 |
1487 | |
1488 | =item * |
1489 | |
61947107 |
1490 | AIX dynamic loading should be now better supported. |
f39f21d8 |
1491 | |
f39f21d8 |
1492 | =item * |
1493 | |
77c8cf41 |
1494 | AIX should now work better with gcc, threads, and 64-bitness. Also the |
1495 | long doubles support in AIX should be better now. See L<perlaix>. |
f39f21d8 |
1496 | |
1497 | =item * |
1498 | |
61947107 |
1499 | After a long pause, AmigaOS has been verified to be happy with Perl. |
1500 | |
1501 | =item * |
1502 | |
77c8cf41 |
1503 | AtheOS (http://www.atheos.cx/) is a new platform. |
f39f21d8 |
1504 | |
77c8cf41 |
1505 | =item * |
f39f21d8 |
1506 | |
77c8cf41 |
1507 | DG/UX platform now supports the 5.005-style threads. See L<perldgux>. |
f39f21d8 |
1508 | |
1509 | =item * |
1510 | |
77c8cf41 |
1511 | DYNIX/ptx platform (a.k.a. dynixptx) is supported at or near osvers 4.5.2. |
f39f21d8 |
1512 | |
1513 | =item * |
1514 | |
61947107 |
1515 | EBCDIC platforms (z/OS, also known as OS/390, POSIX-BC, and VM/ESA) |
1516 | have been regained. Many test suite tests still fail and the |
1517 | co-existence of Unicode and EBCDIC isn't quite settled, but the |
1518 | situation is much better than with Perl 5.6. See L<perlos390>, |
1519 | L<perlbs2000> (for POSIX-BC), and L<perlvmesa> for more information. |
f39f21d8 |
1520 | |
1521 | =item * |
1522 | |
61947107 |
1523 | Building perl with -Duseithreads or -Duse5005threads now works under |
1524 | HP-UX 10.20 (previously it only worked under 10.30 or later). You will |
1525 | need a thread library package installed. See README.hpux. |
f39f21d8 |
1526 | |
77c8cf41 |
1527 | =item * |
f39f21d8 |
1528 | |
61947107 |
1529 | MacOS Classic (MacPerl has of course been available since |
1530 | perl 5.004 but now the source code bases of standard Perl |
1531 | and MacPerl have been synchronised) |
f39f21d8 |
1532 | |
77c8cf41 |
1533 | =item * |
f39f21d8 |
1534 | |
61947107 |
1535 | MacOS X (or Darwin) should now be able to build Perl even on HFS+ |
1536 | filesystems. (The case-insensitivity confused the Perl build process.) |
f39f21d8 |
1537 | |
888aee59 |
1538 | =item * |
1539 | |
61947107 |
1540 | NCR MP-RAS is now supported. |
888aee59 |
1541 | |
1542 | =item * |
1543 | |
61947107 |
1544 | NetWare from Novell is now supported. See L<perlnetware>. |
888aee59 |
1545 | |
1546 | =item * |
1547 | |
61947107 |
1548 | NonStop-UX is now supported. |
888aee59 |
1549 | |
1550 | =item * |
1551 | |
44da0e71 |
1552 | NEC SUPER-UX is now supported. |
1553 | |
1554 | =item * |
1555 | |
61947107 |
1556 | Amdahl UTS UNIX mainframe platform is now supported. |
888aee59 |
1557 | |
1558 | =item * |
1559 | |
61947107 |
1560 | WinCE is now supported. See L<perlce>. |
1561 | |
1562 | =item * |
1563 | |
1564 | z/OS (formerly known as OS/390, formerly known as MVS OE) has now |
1565 | support for dynamic loading. This is not selected by default, |
1566 | however, you must specify -Dusedl in the arguments of Configure. |
888aee59 |
1567 | |
f39f21d8 |
1568 | =back |
1569 | |
1570 | =head1 Selected Bug Fixes |
1571 | |
e1f170bd |
1572 | Numerous memory leaks and uninitialized memory accesses have been |
1573 | hunted down. Most importantly anonymous subs used to leak quite |
1574 | a bit. |
ba370e9b |
1575 | |
f39f21d8 |
1576 | =over 4 |
1577 | |
1578 | =item * |
1579 | |
e1f170bd |
1580 | The autouse pragma didn't work for Multi::Part::Function::Names. |
f39f21d8 |
1581 | |
1582 | =item * |
1583 | |
44da0e71 |
1584 | caller() could cause core dumps in certain situations. Carp was sometimes |
1585 | affected by this problem. |
1586 | |
1587 | =item * |
1588 | |
e1f170bd |
1589 | chop(@list) in list context returned the characters chopped in |
1590 | reverse order. This has been reversed to be in the right order. |
f39f21d8 |
1591 | |
1592 | =item * |
1593 | |
e1f170bd |
1594 | Configure no longer includes the DBM libraries (dbm, gdbm, db, ndbm) |
1595 | when building the Perl binary. The only exception to this is SunOS 4.x, |
1596 | which needs them. |
f39f21d8 |
1597 | |
1598 | =item * |
1599 | |
e1f170bd |
1600 | The behaviour of non-decimal but numeric string constants such as |
1601 | "0x23" was platform-dependent: in some platforms that was seen as 35, |
1602 | in some as 0, in some as a floating point number (don't ask). This |
1603 | was caused by Perl using the operating system libraries in a situation |
1604 | where the result of the string to number conversion is undefined: now |
1605 | Perl consistently handles such strings as zero in numeric contexts. |
f39f21d8 |
1606 | |
1607 | =item * |
1608 | |
e1f170bd |
1609 | The order of DESTROYs has been made more predictable. |
f39f21d8 |
1610 | |
1611 | =item * |
1612 | |
e1f170bd |
1613 | Several debugger fixes: exit code now reflects the script exit code, |
1614 | condition C<"0"> now treated correctly, the C<d> command now checks |
44da0e71 |
1615 | line number, the C<$.> no longer gets corrupted, all debugger output |
1616 | now goes correctly to the socket if RemotePort is set. |
1617 | |
1618 | =item * |
1619 | |
1620 | Perl 5.6.0 could emit spurious warnings about redefinition of dl_error() |
1621 | when statically building extensions into perl. This has been corrected. |
f39f21d8 |
1622 | |
1623 | =item * |
1624 | |
e1f170bd |
1625 | L<dprofpp> -R didn't work. |
f39f21d8 |
1626 | |
1627 | =item * |
1628 | |
e1f170bd |
1629 | C<*foo{FORMAT}> now works. |
44da0e71 |
1630 | =item * |
1631 | |
1632 | Infinity is now recognized as a number. |
f39f21d8 |
1633 | |
1634 | =item * |
1635 | |
e1f170bd |
1636 | UNIVERSAL::isa no longer caches methods incorrectly. (This broke |
1637 | the Tk extension with 5.6.0.) |
f39f21d8 |
1638 | |
1639 | =item * |
1640 | |
e1f170bd |
1641 | Lexicals I: lexicals outside an eval "" weren't resolved |
1642 | correctly inside a subroutine definition inside the eval "" if they |
1643 | were not already referenced in the top level of the eval""ed code. |
f39f21d8 |
1644 | |
1645 | =item * |
1646 | |
e1f170bd |
1647 | Lexicals II: lexicals leaked at file scope into subroutines that |
1648 | were declared before the lexicals. |
f39f21d8 |
1649 | |
1650 | =item * |
1651 | |
44da0e71 |
1652 | Lexical warnings now propagating correctly between scopes |
1653 | and into C<eval "...">. |
1654 | |
1655 | =item * |
1656 | |
1657 | C<use warnings qw(FATAL all)> did not work as intended. This has been |
1658 | corrected. |
1659 | |
1660 | =item * |
1661 | |
1662 | warnings::enabled() now reports the state of $^W correctly if the caller |
1663 | isn't using lexical warnings. |
f39f21d8 |
1664 | |
1665 | =item * |
1666 | |
e1f170bd |
1667 | Line renumbering with eval and C<#line> now works. |
f39f21d8 |
1668 | |
1669 | =item * |
1670 | |
e1f170bd |
1671 | Fixed numerous memory leaks, especially in eval "". |
f39f21d8 |
1672 | |
1673 | =item * |
1674 | |
e1f170bd |
1675 | mkdir() now ignores trailing slashes in the directory name, |
1676 | as mandated by POSIX. |
f39f21d8 |
1677 | |
1678 | =item * |
1679 | |
e1f170bd |
1680 | Some versions of glibc have a broken modfl(). This affects builds |
1681 | with C<-Duselongdouble>. This version of Perl detects this brokenness |
1682 | and has a workaround for it. The glibc release 2.2.2 is known to have |
1683 | fixed the modfl() bug. |
f39f21d8 |
1684 | |
1685 | =item * |
1686 | |
e1f170bd |
1687 | Modulus of unsigned numbers now works (4063328477 % 65535 used to |
1688 | return 27406, instead of 27047). |
f39f21d8 |
1689 | |
1690 | =item * |
1691 | |
e1f170bd |
1692 | Some "not a number" warnings introduced in 5.6.0 eliminated to be |
1693 | more compatible with 5.005. Infinity is now recognised as a number. |
f39f21d8 |
1694 | |
77c8cf41 |
1695 | =item * |
f39f21d8 |
1696 | |
44da0e71 |
1697 | Numeric conversions did not recognize changes in the string value |
1698 | properly in certain circumstances. |
1699 | |
1700 | =item * |
1701 | |
e1f170bd |
1702 | Attributes (like :shared) didn't work with our(). |
f39f21d8 |
1703 | |
1704 | =item * |
1705 | |
e1f170bd |
1706 | our() variables will not cause "will not stay shared" warnings. |
f39f21d8 |
1707 | |
1708 | =item * |
1709 | |
44da0e71 |
1710 | "our" variables of the same name declared in two sibling blocks |
1711 | resulted in bogus warnings about "redeclaration" of the variables. |
1712 | The problem has been corrected. |
1713 | |
1714 | =item * |
1715 | |
e1f170bd |
1716 | pack "Z" now correctly terminates the string with "\0". |
f39f21d8 |
1717 | |
1718 | =item * |
1719 | |
e1f170bd |
1720 | Fix password routines which in some shadow password platforms |
1721 | (e.g. HP-UX) caused getpwent() to return every other entry. |
f39f21d8 |
1722 | |
77c8cf41 |
1723 | =item * |
f39f21d8 |
1724 | |
e1f170bd |
1725 | The PERL5OPT environment variable (for passing command line arguments |
1726 | to Perl) didn't work for more than a single group of options. |
f39f21d8 |
1727 | |
77c8cf41 |
1728 | =item * |
f39f21d8 |
1729 | |
e1f170bd |
1730 | PERL5OPT with embedded spaces didn't work. |
f39f21d8 |
1731 | |
77c8cf41 |
1732 | =item * |
f39f21d8 |
1733 | |
e1f170bd |
1734 | printf() no longer resets the numeric locale to "C". |
f39f21d8 |
1735 | |
77c8cf41 |
1736 | =item * |
f39f21d8 |
1737 | |
44da0e71 |
1738 | C<qw(a\\b)> now parses correctly as C<'a\\b'>. |
1739 | |
1740 | =item * |
1741 | |
1742 | pos() did not return the correct value within s///ge in earlier |
1743 | versions. This is now handled correctly. |
f39f21d8 |
1744 | |
77c8cf41 |
1745 | =item * |
f39f21d8 |
1746 | |
e1f170bd |
1747 | Printing quads (64-bit integers) with printf/sprintf now works |
1748 | without the q L ll prefixes (assuming you are on a quad-capable platform). |
f39f21d8 |
1749 | |
77c8cf41 |
1750 | =item * |
f39f21d8 |
1751 | |
e1f170bd |
1752 | Regular expressions on references and overloaded scalars now work. |
f39f21d8 |
1753 | |
ba370e9b |
1754 | =item * |
1755 | |
e1f170bd |
1756 | Right-hand side magic (GMAGIC) could in many cases such as string |
1757 | concatenation be invoked too many times. |
ba370e9b |
1758 | |
1759 | =item * |
1760 | |
e1f170bd |
1761 | scalar() now forces scalar context even when used in void context. |
ba370e9b |
1762 | |
1763 | =item * |
1764 | |
e1f170bd |
1765 | SOCKS support is now much more robust. |
ba370e9b |
1766 | |
1767 | =item * |
1768 | |
e1f170bd |
1769 | sort() arguments are now compiled in the right wantarray context |
1770 | (they were accidentally using the context of the sort() itself). |
44da0e71 |
1771 | The comparison block is now run in scalar context, and the arguments |
1772 | to be sorted are always provided list context. |
ba370e9b |
1773 | |
1774 | =item * |
1775 | |
e1f170bd |
1776 | Changed the POSIX character class C<[[:space:]]> to include the (very |
c2e23569 |
1777 | rarely used) vertical tab character. Added a new POSIX-ish character |
1778 | class C<[[:blank:]]> which stands for horizontal whitespace |
1779 | (currently, the space and the tab). |
ba370e9b |
1780 | |
1781 | =item * |
1782 | |
1783 | The tainting behaviour of sprintf() has been rationalized. It does |
1784 | not taint the result of floating point formats anymore, making the |
1785 | behaviour consistent with that of string interpolation. |
1786 | |
1787 | =item * |
1788 | |
44da0e71 |
1789 | Some cases of inconsistent taint propagation (such as within hash |
1790 | values) have been fixed. |
1791 | |
1792 | =item * |
1793 | |
1794 | The RE engine found in Perl 5.6.0 accidentally pessimised certain kinds |
1795 | of simple pattern matches. These are now handled better. |
1796 | |
1797 | =item * |
1798 | |
1799 | Regular expression debug output (whether through C<use re 'debug'> |
1800 | or via C<-Dr>) now looks better. |
1801 | |
1802 | =item * |
1803 | |
1804 | Multi-line matches like C<"a\nxb\n" =~ /(?!\A)x/m> were flawed. The |
1805 | bug has been fixed. |
1806 | |
1807 | =item * |
1808 | |
1809 | Use of $& could trigger a core dump under some situations. This |
1810 | is now avoided. |
1811 | |
1812 | =item * |
1813 | |
c2e23569 |
1814 | The regular expression captured submatches ($1, $2, ...) are now |
1815 | more consistently unset if the match fails, instead of leaving false |
1816 | data lying around in them. |
1817 | |
1818 | =item * |
1819 | |
44da0e71 |
1820 | readline() on files opened in "slurp" mode could return an extra "" at |
1821 | the end in certain situations. This has been corrected. |
1822 | |
1823 | =item * |
1824 | |
1825 | Autovivification of symbolic references of special variables described |
1826 | in L<perlvar> (as in C<${$num}>) was accidentally disabled. This works |
1827 | again now. |
1828 | |
1829 | =item * |
1830 | |
da6838c8 |
1831 | Sys::Syslog ignored the C<LOG_AUTH> constant. |
ba370e9b |
1832 | |
1833 | =item * |
1834 | |
e1f170bd |
1835 | All but the first argument of the IO syswrite() method are now optional. |
ba370e9b |
1836 | |
1837 | =item * |
1838 | |
e1f170bd |
1839 | $AUTOLOAD, sort(), lock(), and spawning subprocesses |
1840 | in multiple threads simultaneously are now thread-safe. |
ba370e9b |
1841 | |
1842 | =item * |
1843 | |
e1f170bd |
1844 | Tie::ARRAY SPLICE method was broken. |
ba370e9b |
1845 | |
1846 | =item * |
1847 | |
e1f170bd |
1848 | Allow read-only string on left hand side of non-modifying tr///. |
ba370e9b |
1849 | |
1850 | =item * |
1851 | |
e1f170bd |
1852 | Several Unicode fixes. |
ba370e9b |
1853 | |
1854 | =over 8 |
1855 | |
1856 | =item * |
1857 | |
e1f170bd |
1858 | BOMs (byte order marks) in the beginning of Perl files |
1859 | (scripts, modules) should now be transparently skipped. |
1860 | UTF-16 (UCS-2) encoded Perl files should now be read correctly. |
ba370e9b |
1861 | |
1862 | =item * |
1863 | |
e1f170bd |
1864 | The character tables have been updated to Unicode 3.1.1. |
ba370e9b |
1865 | |
1866 | =item * |
1867 | |
e1f170bd |
1868 | Comparing with utf8 data does not magically upgrade non-utf8 data |
1869 | into utf8. |
ba370e9b |
1870 | |
1871 | =item * |
1872 | |
e1f170bd |
1873 | C<IsAlnum>, C<IsAlpha>, and C<IsWord> now match titlecase. |
f39f21d8 |
1874 | |
77c8cf41 |
1875 | =item * |
f39f21d8 |
1876 | |
e1f170bd |
1877 | Concatenation with the C<.> operator or via variable interpolation, |
1878 | C<eq>, C<substr>, C<reverse>, C<quotemeta>, the C<x> operator, |
1879 | substitution with C<s///>, single-quoted UTF8, should now work. |
f39f21d8 |
1880 | |
77c8cf41 |
1881 | =item * |
f39f21d8 |
1882 | |
e1f170bd |
1883 | The C<tr///> operator now works. Note that the C<tr///CU> |
1884 | functionality has been removed (but see pack('U0', ...)). |
f39f21d8 |
1885 | |
77c8cf41 |
1886 | =item * |
f39f21d8 |
1887 | |
e1f170bd |
1888 | C<eval "v200"> now works. |
f39f21d8 |
1889 | |
77c8cf41 |
1890 | =item * |
f39f21d8 |
1891 | |
44da0e71 |
1892 | Perl 5.6.0 parsed m/\x{ab}/ incorrectly, leading to spurious warnings. |
1893 | This has been corrected. |
1894 | |
1895 | =item * |
1896 | |
e1f170bd |
1897 | Zero entries were missing from the Unicode classes like C<IsDigit>. |
f39f21d8 |
1898 | |
e1f170bd |
1899 | =back |
f39f21d8 |
1900 | |
44da0e71 |
1901 | =item * |
1902 | |
1903 | Large unsigned numbers (those above 2**31) could sometimes lose their |
1904 | unsignedness, causing bogus results in arithmetic operations. |
1905 | |
77c8cf41 |
1906 | =back |
f39f21d8 |
1907 | |
77c8cf41 |
1908 | =head2 Platform Specific Changes and Fixes |
f39f21d8 |
1909 | |
1910 | =over 4 |
1911 | |
1912 | =item * |
1913 | |
77c8cf41 |
1914 | BSDI 4.* |
f39f21d8 |
1915 | |
77c8cf41 |
1916 | Perl now works on post-4.0 BSD/OSes. |
f39f21d8 |
1917 | |
1918 | =item * |
1919 | |
77c8cf41 |
1920 | All BSDs |
f39f21d8 |
1921 | |
057b7f2b |
1922 | Setting C<$0> now works (as much as possible; see L<perlvar> for details). |
f39f21d8 |
1923 | |
1924 | =item * |
1925 | |
77c8cf41 |
1926 | Cygwin |
f39f21d8 |
1927 | |
77c8cf41 |
1928 | Numerous updates; currently synchronised with Cygwin 1.1.4. |
f39f21d8 |
1929 | |
1930 | =item * |
1931 | |
e1f170bd |
1932 | Previously DYNIX/ptx had problems in its Configure probe for non-blocking I/O. |
1933 | |
1934 | =item * |
1935 | |
77c8cf41 |
1936 | EPOC |
f39f21d8 |
1937 | |
77c8cf41 |
1938 | EPOC update after Perl 5.6.0. See README.epoc. |
f39f21d8 |
1939 | |
1940 | =item * |
1941 | |
77c8cf41 |
1942 | FreeBSD 3.* |
f39f21d8 |
1943 | |
77c8cf41 |
1944 | Perl now works on post-3.0 FreeBSDs. |
f39f21d8 |
1945 | |
1946 | =item * |
1947 | |
77c8cf41 |
1948 | HP-UX |
1949 | |
1950 | README.hpux updated; C<Configure -Duse64bitall> now almost works. |
f39f21d8 |
1951 | |
1952 | =item * |
1953 | |
77c8cf41 |
1954 | IRIX |
f39f21d8 |
1955 | |
77c8cf41 |
1956 | Numerous compilation flag and hint enhancements; accidental mixing |
1957 | of 32-bit and 64-bit libraries (a doomed attempt) made much harder. |
f39f21d8 |
1958 | |
77c8cf41 |
1959 | =item * |
f39f21d8 |
1960 | |
77c8cf41 |
1961 | Linux |
f39f21d8 |
1962 | |
e1f170bd |
1963 | =over 8 |
1964 | |
1965 | =item * |
1966 | |
77c8cf41 |
1967 | Long doubles should now work (see INSTALL). |
f39f21d8 |
1968 | |
1969 | =item * |
1970 | |
e1f170bd |
1971 | Linux previously had problems related to sockaddrlen when using |
1972 | accept(), revcfrom() (in Perl: recv()), getpeername(), and getsockname(). |
1973 | |
1974 | =back |
1975 | |
1976 | =item * |
1977 | |
77c8cf41 |
1978 | MacOS Classic |
f39f21d8 |
1979 | |
77c8cf41 |
1980 | Compilation of the standard Perl distribution in MacOS Classic should |
1981 | now work if you have the Metrowerks development environment and |
1982 | the missing Mac-specific toolkit bits. Contact the macperl mailing |
1983 | list for details. |
f39f21d8 |
1984 | |
1985 | =item * |
1986 | |
77c8cf41 |
1987 | MPE/iX |
f39f21d8 |
1988 | |
77c8cf41 |
1989 | MPE/iX update after Perl 5.6.0. See README.mpeix. |
f39f21d8 |
1990 | |
1991 | =item * |
1992 | |
77c8cf41 |
1993 | NetBSD/sparc |
f39f21d8 |
1994 | |
77c8cf41 |
1995 | Perl now works on NetBSD/sparc. |
f39f21d8 |
1996 | |
1997 | =item * |
1998 | |
77c8cf41 |
1999 | OS/2 |
f39f21d8 |
2000 | |
77c8cf41 |
2001 | Now works with usethreads (see INSTALL). |
f39f21d8 |
2002 | |
2003 | =item * |
2004 | |
77c8cf41 |
2005 | Solaris |
f39f21d8 |
2006 | |
77c8cf41 |
2007 | 64-bitness using the Sun Workshop compiler now works. |
f39f21d8 |
2008 | |
2009 | =item * |
2010 | |
77c8cf41 |
2011 | Tru64 (aka Digital UNIX, aka DEC OSF/1) |
f39f21d8 |
2012 | |
77c8cf41 |
2013 | The operating system version letter now recorded in $Config{osvers}. |
2014 | Allow compiling with gcc (previously explicitly forbidden). Compiling |
2015 | with gcc still not recommended because buggy code results, even with |
2016 | gcc 2.95.2. |
f39f21d8 |
2017 | |
2018 | =item * |
2019 | |
77c8cf41 |
2020 | Unicos |
2021 | |
2022 | Fixed various alignment problems that lead into core dumps either |
2023 | during build or later; no longer dies on math errors at runtime; |
2024 | now using full quad integers (64 bits), previously was using |
2025 | only 46 bit integers for speed. |
f39f21d8 |
2026 | |
2027 | =item * |
2028 | |
77c8cf41 |
2029 | VMS |
2030 | |
2031 | chdir() now works better despite a CRT bug; now works with MULTIPLICITY |
2032 | (see INSTALL); now works with Perl's malloc. |
f39f21d8 |
2033 | |
00bb525a |
2034 | The tainting of C<%ENV> elements via C<keys> or C<values> was previously |
2035 | unimplemented. It now works as documented. |
2036 | |
2037 | The C<waitpid> emulation has been improved. The worst bug (now fixed) |
2038 | was that a pid of -1 would cause a wildcard search of all processes on |
2039 | the system. The most significant enhancement is that we can now |
2040 | usually get the completion status of a terminated process. |
2041 | |
2042 | POSIX-style signals are now emulated much better on VMS versions prior |
2043 | to 7.0. |
2044 | |
2045 | The C<system> function and backticks operator have improved |
2046 | functionality and better error handling. |
2047 | |
f39f21d8 |
2048 | =item * |
2049 | |
77c8cf41 |
2050 | Windows |
f39f21d8 |
2051 | |
77c8cf41 |
2052 | =over 8 |
f39f21d8 |
2053 | |
2054 | =item * |
2055 | |
77c8cf41 |
2056 | accept() no longer leaks memory. |
f39f21d8 |
2057 | |
2058 | =item * |
2059 | |
e1f170bd |
2060 | Borland C++ v5.5 is now a supported compiler that can build Perl. |
2061 | However, the generated binaries continue to be incompatible with those |
2062 | generated by the other supported compilers (GCC and Visual C++). |
2063 | |
2064 | =item * |
2065 | |
77c8cf41 |
2066 | Better chdir() return value for a non-existent directory. |
f39f21d8 |
2067 | |
77c8cf41 |
2068 | =item * |
f39f21d8 |
2069 | |
e1f170bd |
2070 | Duping socket handles with open(F, ">&MYSOCK") now works under Windows 9x. |
2071 | |
2072 | =item * |
2073 | |
77c8cf41 |
2074 | New %ENV entries now propagate to subprocesses. |
f39f21d8 |
2075 | |
2076 | =item * |
2077 | |
44da0e71 |
2078 | Current directory entries in %ENV are now correctly propagated to child |
2079 | processes. |
2080 | |
2081 | =item * |
2082 | |
77c8cf41 |
2083 | $ENV{LIB} now used to search for libs under Visual C. |
2084 | |
2085 | =item * |
2086 | |
44da0e71 |
2087 | fork() emulation has been improved in various ways, but still continues |
2088 | to be experimental. See L<perlfork> for known bugs and caveats. |
e1f170bd |
2089 | |
2090 | =item * |
2091 | |
77c8cf41 |
2092 | A failed (pseudo)fork now returns undef and sets errno to EAGAIN. |
f39f21d8 |
2093 | |
2094 | =item * |
2095 | |
44da0e71 |
2096 | Win32::GetCwd() correctly returns C:\ instead of C: when at the drive root. |
2097 | Other bugs in chdir() and Cwd::cwd() have also been fixed. |
2098 | |
2099 | =item * |
2100 | |
e1f170bd |
2101 | HTML files will be installed in c:\perl\html instead of c:\perl\lib\pod\html |
2102 | |
2103 | =item * |
2104 | |
2105 | The makefiles now provide a single switch to bulk-enable all the features |
2106 | enabled in ActiveState ActivePerl (a popular Win32 binary distribution). |
2107 | |
2108 | =item * |
2109 | |
77c8cf41 |
2110 | Allow REG_EXPAND_SZ keys in the registry. |
f39f21d8 |
2111 | |
2112 | =item * |
2113 | |
77c8cf41 |
2114 | Can now send() from all threads, not just the first one. |
f39f21d8 |
2115 | |
2116 | =item * |
2117 | |
77c8cf41 |
2118 | Fake signal handling reenabled, bugs and all. |
f39f21d8 |
2119 | |
2120 | =item * |
2121 | |
44da0e71 |
2122 | %SIG has been enabled under USE_ITHREADS, but its use is completely |
2123 | unsupported under all configurations. |
2124 | |
2125 | =item * |
2126 | |
77c8cf41 |
2127 | Less stack reserved per thread so that more threads can run |
2128 | concurrently. (Still 16M per thread.) |
f39f21d8 |
2129 | |
2130 | =item * |
2131 | |
c2e23569 |
2132 | C<File::Spec->tmpdir()> now prefers C:/temp over /tmp |
77c8cf41 |
2133 | (works better when perl is running as service). |
f39f21d8 |
2134 | |
2135 | =item * |
2136 | |
77c8cf41 |
2137 | Better UNC path handling under ithreads. |
f39f21d8 |
2138 | |
2139 | =item * |
2140 | |
44da0e71 |
2141 | wait(), waitpid() and backticks now return the correct exit status under |
2142 | Windows 9x. |
f39f21d8 |
2143 | |
2144 | =item * |
2145 | |
77c8cf41 |
2146 | winsock handle leak fixed. |
f39f21d8 |
2147 | |
2148 | =back |
2149 | |
77c8cf41 |
2150 | =back |
f39f21d8 |
2151 | |
77c8cf41 |
2152 | =head1 New or Changed Diagnostics |
f39f21d8 |
2153 | |
ba370e9b |
2154 | =over 4 |
2155 | |
2156 | =item * |
2157 | |
77c8cf41 |
2158 | All regular expression compilation error messages are now hopefully |
2159 | easier to understand both because the error message now comes before |
2160 | the failed regex and because the point of failure is now clearly |
ba370e9b |
2161 | marked by a C<E<lt>-- HERE> marker. |
2162 | |
2163 | =item * |
f39f21d8 |
2164 | |
77c8cf41 |
2165 | The various "opened only for", "on closed", "never opened" warnings |
2166 | drop the C<main::> prefix for filehandles in the C<main> package, |
bea4d472 |
2167 | for example C<STDIN> instead of C<main::STDIN>. |
f39f21d8 |
2168 | |
ba370e9b |
2169 | =item * |
2170 | |
77c8cf41 |
2171 | The "Unrecognized escape" warning has been extended to include C<\8>, |
2172 | C<\9>, and C<\_>. There is no need to escape any of the C<\w> characters. |
f39f21d8 |
2173 | |
ba370e9b |
2174 | =item * |
f39f21d8 |
2175 | |
77c8cf41 |
2176 | Two new debugging options have been added: if you have compiled your |
2177 | Perl with debugging, you can use the -DT and -DR options to trace |
2178 | tokenising and to add reference counts to displaying variables, |
2179 | respectively. |
f39f21d8 |
2180 | |
2181 | =item * |
2182 | |
77c8cf41 |
2183 | If an attempt to use a (non-blessed) reference as an array index |
2184 | is made, a warning is given. |
f39f21d8 |
2185 | |
2186 | =item * |
2187 | |
77c8cf41 |
2188 | C<push @a;> and C<unshift @a;> (with no values to push or unshift) |
2189 | now give a warning. This may be a problem for generated and evaled |
2190 | code. |
f39f21d8 |
2191 | |
ba370e9b |
2192 | =item * |
2193 | |
2194 | If you try to L<perlfunc/pack> a number less than 0 or larger than 255 |
2195 | using the C<"C"> format you will get an optional warning. Similarly |
2196 | for the C<"c"> format and a number less than -128 or more than 127. |
2197 | |
2198 | =item * |
2199 | |
2200 | Certain regex modifiers such as C<(?o)> make sense only if applied to |
2201 | the entire regex. You will an optional warning if you try to do otherwise. |
2202 | |
2203 | =item * |
2204 | |
c2e23569 |
2205 | Using arrays or hashes as references (e.g. C<< %foo->{bar} >> |
2206 | has been deprecated for a while. Now you will get an optional warning. |
ba370e9b |
2207 | |
f39f21d8 |
2208 | =back |
2209 | |
77c8cf41 |
2210 | =head1 Changed Internals |
f39f21d8 |
2211 | |
2212 | =over 4 |
2213 | |
2214 | =item * |
2215 | |
77c8cf41 |
2216 | perlapi.pod (a companion to perlguts) now attempts to document the |
2217 | internal API. |
f39f21d8 |
2218 | |
2219 | =item * |
2220 | |
77c8cf41 |
2221 | You can now build a really minimal perl called microperl. |
2222 | Building microperl does not require even running Configure; |
2223 | C<make -f Makefile.micro> should be enough. Beware: microperl makes |
2224 | many assumptions, some of which may be too bold; the resulting |
2225 | executable may crash or otherwise misbehave in wondrous ways. |
2226 | For careful hackers only. |
f39f21d8 |
2227 | |
2228 | =item * |
2229 | |
c2e23569 |
2230 | Added rsignal(), whichsig(), do_join(), op_clear, op_null, |
2231 | ptr_table_clear(), ptr_table_free(), sv_setref_uv(), and several UTF-8 |
2232 | interfaces to the publicised API. For the full list of the available |
2233 | APIs see L<perlapi>. |
f39f21d8 |
2234 | |
2235 | =item * |
2236 | |
77c8cf41 |
2237 | Made possible to propagate customised exceptions via croak()ing. |
f39f21d8 |
2238 | |
77c8cf41 |
2239 | =item * |
f39f21d8 |
2240 | |
95f0a2f1 |
2241 | Now xsubs can have attributes just like subs. (Well, at least the |
2242 | built-in attributes.) |
f39f21d8 |
2243 | |
2244 | =item * |
2245 | |
77c8cf41 |
2246 | dTHR and djSP have been obsoleted; the former removed (because it's |
2247 | a no-op) and the latter replaced with dSP. |
f39f21d8 |
2248 | |
2249 | =item * |
2250 | |
61947107 |
2251 | PERL_OBJECT has been completely removed. |
2252 | |
2253 | =item * |
2254 | |
ba370e9b |
2255 | The MAGIC constants (e.g. C<'P'>) have been macrofied |
2256 | (e.g. C<PERL_MAGIC_TIED>) for better source code readability |
2257 | and maintainability. |
2258 | |
2259 | =item * |
2260 | |
2261 | The regex compiler now maintains a structure that identifies nodes in |
2262 | the compiled bytecode with the corresponding syntactic features of the |
2263 | original regex expression. The information is attached to the new |
2264 | C<offsets> member of the C<struct regexp>. See L<perldebguts> for more |
2265 | complete information. |
2266 | |
2267 | =item * |
2268 | |
2269 | The C code has been made much more C<gcc -Wall> clean. Some warning |
2270 | messages still remain in some platforms, so if you are compiling with |
2271 | gcc you may see some warnings about dubious practices. The warnings |
2272 | are being worked on. |
2273 | |
2274 | =item * |
2275 | |
2276 | F<perly.c>, F<sv.c>, and F<sv.h> have now been extensively commented. |
2277 | |
2278 | =item * |
2279 | |
61947107 |
2280 | Documentation on how to use the Perl source repository has been added |
2281 | to F<Porting/repository.pod>. |
f39f21d8 |
2282 | |
888aee59 |
2283 | =item * |
2284 | |
c2e23569 |
2285 | There are now several profiling make targets. |
888aee59 |
2286 | |
77c8cf41 |
2287 | =back |
f39f21d8 |
2288 | |
77c8cf41 |
2289 | =head1 Security Vulnerability Closed |
f39f21d8 |
2290 | |
77c8cf41 |
2291 | (This change was already made in 5.7.0 but bears repeating here.) |
f39f21d8 |
2292 | |
77c8cf41 |
2293 | A potential security vulnerability in the optional suidperl component |
2294 | of Perl was identified in August 2000. suidperl is neither built nor |
2295 | installed by default. As of November 2001 the only known vulnerable |
2296 | platform is Linux, most likely all Linux distributions. CERT and |
2297 | various vendors and distributors have been alerted about the vulnerability. |
2298 | See http://www.cpan.org/src/5.0/sperl-2000-08-05/sperl-2000-08-05.txt |
2299 | for more information. |
f39f21d8 |
2300 | |
77c8cf41 |
2301 | The problem was caused by Perl trying to report a suspected security |
2302 | exploit attempt using an external program, /bin/mail. On Linux |
2303 | platforms the /bin/mail program had an undocumented feature which |
2304 | when combined with suidperl gave access to a root shell, resulting in |
2305 | a serious compromise instead of reporting the exploit attempt. If you |
2306 | don't have /bin/mail, or if you have 'safe setuid scripts', or if |
2307 | suidperl is not installed, you are safe. |
f39f21d8 |
2308 | |
77c8cf41 |
2309 | The exploit attempt reporting feature has been completely removed from |
2310 | Perl 5.8.0 (and the maintenance release 5.6.1, and it was removed also |
2311 | from all the Perl 5.7 releases), so that particular vulnerability |
2312 | isn't there anymore. However, further security vulnerabilities are, |
ba370e9b |
2313 | unfortunately, always possible. The suidperl functionality is most |
2314 | probably going to be removed in Perl 5.10. In any case, suidperl |
2315 | should only be used by security experts who know exactly what they are |
2316 | doing and why they are using suidperl instead of some other solution |
2317 | such as sudo (see http://www.courtesan.com/sudo/). |
77c8cf41 |
2318 | |
2319 | =head1 New Tests |
2320 | |
76663d67 |
2321 | Several new tests have been added, especially for the F<lib> |
2322 | subsection. There are now about 34 000 individual tests (spread over |
2323 | about 530 test scripts), in the regression suite (5.6.1 has about |
2324 | 11700 tests, in 258 test scripts) Many of the new tests are introduced |
2325 | by the new modules, but still in general Perl is now more thoroughly |
2326 | tested. |
2327 | |
2328 | Because of the large number of tests, running the regression suite |
2329 | will take considerably longer time than it used to: expect the suite |
2330 | to take up to 4-5 times longer to run than in perl 5.6. In a really |
2331 | fast machine you can hope to finish the suite in about 5 minutes |
2332 | (wallclock time). |
77c8cf41 |
2333 | |
2334 | The tests are now reported in a different order than in earlier Perls. |
2335 | (This happens because the test scripts from under t/lib have been moved |
2336 | to be closer to the library/extension they are testing.) |
2337 | |
f39f21d8 |
2338 | =head1 Known Problems |
2339 | |
f39f21d8 |
2340 | =head2 AIX |
2341 | |
2342 | =over 4 |
2343 | |
2344 | =item * |
2345 | |
2346 | In AIX 4.2 Perl extensions that use C++ functions that use statics |
2347 | may have problems in that the statics are not getting initialized. |
2348 | In newer AIX releases this has been solved by linking Perl with |
2349 | the libC_r library, but unfortunately in AIX 4.2 the said library |
2350 | has an obscure bug where the various functions related to time |
2351 | (such as time() and gettimeofday()) return broken values, and |
2352 | therefore in AIX 4.2 Perl is not linked against the libC_r. |
2353 | |
2354 | =item * |
2355 | |
2356 | vac 5.0.0.0 May Produce Buggy Code For Perl |
2357 | |
2358 | The AIX C compiler vac version 5.0.0.0 may produce buggy code, |
2359 | resulting in few random tests failing, but when the failing tests |
2360 | are run by hand, they succeed. We suggest upgrading to at least |
2361 | vac version 5.0.1.0, that has been known to compile Perl correctly. |
2362 | "lslpp -L|grep vac.C" will tell you the vac version. |
2363 | |
2364 | =back |
2365 | |
2366 | =head2 Amiga Perl Invoking Mystery |
2367 | |
2368 | One cannot call Perl using the C<volume:> syntax, that is, C<perl -v> |
057b7f2b |
2369 | works, but for example C<bin:perl -v> doesn't. The exact reason isn't |
f39f21d8 |
2370 | known but the current suspect is the F<ixemul> library. |
2371 | |
2372 | =head2 lib/ftmp-security tests warn 'system possibly insecure' |
2373 | |
2374 | Don't panic. Read INSTALL 'make test' section instead. |
2375 | |
2376 | =head2 Cygwin intermittent failures of lib/Memoize/t/expire_file 11 and 12 |
2377 | |
2378 | The subtests 11 and 12 sometimes fail and sometimes work. |
2379 | |
2380 | =head2 HP-UX lib/io_multihomed Fails When LP64-Configured |
2381 | |
2382 | The lib/io_multihomed test may hang in HP-UX if Perl has been |
2383 | configured to be 64-bit. Because other 64-bit platforms do not hang in |
2384 | this test, HP-UX is suspect. All other tests pass in 64-bit HP-UX. The |
2385 | test attempts to create and connect to "multihomed" sockets (sockets |
2386 | which have multiple IP addresses). |
2387 | |
2388 | =head2 HP-UX lib/posix Subtest 9 Fails When LP64-Configured |
2389 | |
2390 | If perl is configured with -Duse64bitall, the successful result of the |
2391 | subtest 10 of lib/posix may arrive before the successful result of the |
2392 | subtest 9, which confuses the test harness so much that it thinks the |
2393 | subtest 9 failed. |
2394 | |
2395 | =head2 Linux With Sfio Fails op/misc Test 48 |
2396 | |
2397 | No known fix. |
2398 | |
a0aae13b |
2399 | =head2 Mac OS X |
2400 | |
2401 | The following tests are known to fail: |
2402 | |
2403 | Failed Test Stat Wstat Total Fail Failed List of Failed |
2404 | ------------------------------------------------------------------------- |
2405 | ../ext/DB_File/t/db-btree.t 0 11 ?? ?? % ?? |
2406 | ../ext/DB_File/t/db-recno.t 149 3 2.01% 61 63 65 |
2407 | ../ext/POSIX/t/posix.t 31 1 3.23% 10 |
2408 | ../lib/warnings.t 450 1 0.22% 316 |
2409 | |
f39f21d8 |
2410 | =head2 OS/390 |
2411 | |
2412 | OS/390 has rather many test failures but the situation is actually |
2413 | better than it was in 5.6.0, it's just that so many new modules and |
2414 | tests have been added. |
2415 | |
2416 | Failed Test Stat Wstat Total Fail Failed List of Failed |
2417 | ----------------------------------------------------------------------------- |
2418 | ../ext/B/Deparse.t 14 1 7.14% 14 |
2419 | ../ext/B/Showlex.t 1 1 100.00% 1 |
2420 | ../ext/Encode/Encode/Tcl.t 610 13 2.13% 592 594 596 598 |
2421 | 600 602 604-610 |
2422 | ../ext/IO/lib/IO/t/io_unix.t 113 28928 5 3 60.00% 3-5 |
2423 | ../ext/POSIX/POSIX.t 29 1 3.45% 14 |
2424 | ../ext/Storable/t/lock.t 255 65280 5 3 60.00% 3-5 |
2425 | ../lib/locale.t 129 33024 117 19 16.24% 99-117 |
2426 | ../lib/warnings.t 434 1 0.23% 75 |
2427 | ../lib/ExtUtils.t 27 1 3.70% 25 |
2428 | ../lib/Math/BigInt/t/bigintpm.t 1190 1 0.08% 1145 |
2429 | ../lib/Unicode/UCD.t 81 48 59.26% 1-16 49-64 66-81 |
2430 | ../lib/User/pwent.t 9 1 11.11% 4 |
2431 | op/pat.t 660 6 0.91% 242-243 424-425 |
2432 | 626-627 |
2433 | op/split.t 0 9 ?? ?? % ?? |
2434 | op/taint.t 174 3 1.72% 156 162 168 |
2435 | op/tr.t 70 3 4.29% 50 58-59 |
2436 | Failed 16/422 test scripts, 96.21% okay. 105/23251 subtests failed, 99.55% okay. |
2437 | |
2438 | =head2 op/sprintf tests 129 and 130 |
2439 | |
2440 | The op/sprintf tests 129 and 130 are known to fail on some platforms. |
2441 | Examples include any platform using sfio, and Compaq/Tandem's NonStop-UX. |
2442 | The failing platforms do not comply with the ANSI C Standard, line |
2443 | 19ff on page 134 of ANSI X3.159 1989 to be exact. (They produce |
2444 | something other than "1" and "-1" when formatting 0.6 and -0.6 using |
2445 | the printf format "%.0f", most often they produce "0" and "-0".) |
2446 | |
2447 | =head2 Failure of Thread tests |
2448 | |
fedd8cf1 |
2449 | B<Note that support for 5.005-style threading remains experimental |
2450 | and practically unsupported.> |
f39f21d8 |
2451 | |
2452 | The following tests are known to fail due to fundamental problems in |
2453 | the 5.005 threading implementation. These are not new failures--Perl |
2454 | 5.005_0x has the same bugs, but didn't have these tests. |
2455 | |
fedd8cf1 |
2456 | ext/List/Util/t/first 2 |
2457 | lib/autouse 4 |
2458 | ext/Thread/thr5005 19-20 |
2459 | |
2460 | These failures are unlikely to get fixed. |
f39f21d8 |
2461 | |
2462 | =head2 UNICOS |
2463 | |
2464 | =over 4 |
2465 | |
2466 | =item * |
2467 | |
2468 | ext/POSIX/sigaction subtests 6 and 13 may fail. |
2469 | |
2470 | =item * |
2471 | |
2472 | lib/ExtUtils may spuriously claim that subtest 28 failed, |
2473 | which is interesting since the test only has 27 tests. |
2474 | |
2475 | =item * |
2476 | |
2477 | Numerous numerical test failures |
2478 | |
2479 | op/numconvert 209,210,217,218 |
2480 | op/override 7 |
2481 | ext/Time/HiRes/HiRes 9 |
2482 | lib/Math/BigInt/t/bigintpm 1145 |
2483 | lib/Math/Trig 25 |
2484 | |
2485 | These tests fail because of yet unresolved floating point inaccuracies. |
2486 | |
2487 | =back |
2488 | |
2489 | =head2 UTS |
2490 | |
2491 | There are a few known test failures, see L<perluts>. |
2492 | |
2493 | =head2 VMS |
2494 | |
00bb525a |
2495 | There is one known test failure with a default configuration: |
aecce728 |
2496 | |
2497 | [.run]switches..........................FAILED on test 1 |
aecce728 |
2498 | |
f39f21d8 |
2499 | =head2 Win32 |
2500 | |
2501 | In multi-CPU boxes there are some problems with the I/O buffering: |
2502 | some output may appear twice. |
2503 | |
2504 | =head2 Localising a Tied Variable Leaks Memory |
2505 | |
2506 | use Tie::Hash; |
2507 | tie my %tie_hash => 'Tie::StdHash'; |
2508 | |
2509 | ... |
2510 | |
2511 | local($tie_hash{Foo}) = 1; # leaks |
2512 | |
2513 | Code like the above is known to leak memory every time the local() |
2514 | is executed. |
2515 | |
aecce728 |
2516 | =head2 Localising Tied Arrays and Hashes Is Broken |
2517 | |
2518 | local %tied_array; |
2519 | |
2520 | doesn't work as one would expect: the old value is restored |
2521 | incorrectly. |
2522 | |
f39f21d8 |
2523 | =head2 Self-tying of Arrays and Hashes Is Forbidden |
2524 | |
2525 | Self-tying of arrays and hashes is broken in rather deep and |
2526 | hard-to-fix ways. As a stop-gap measure to avoid people from getting |
2527 | frustrated at the mysterious results (core dumps, most often) it is |
2528 | for now forbidden (you will get a fatal error even from an attempt). |
2529 | |
f39f21d8 |
2530 | =head2 Building Extensions Can Fail Because Of Largefiles |
2531 | |
2532 | Some extensions like mod_perl are known to have issues with |
2533 | `largefiles', a change brought by Perl 5.6.0 in which file offsets |
2534 | default to 64 bits wide, where supported. Modules may fail to compile |
2535 | at all or compile and work incorrectly. Currently there is no good |
2536 | solution for the problem, but Configure now provides appropriate |
2537 | non-largefile ccflags, ldflags, libswanted, and libs in the %Config |
2538 | hash (e.g., $Config{ccflags_nolargefiles}) so the extensions that are |
2539 | having problems can try configuring themselves without the |
2540 | largefileness. This is admittedly not a clean solution, and the |
2541 | solution may not even work at all. One potential failure is whether |
2542 | one can (or, if one can, whether it's a good idea) link together at |
2543 | all binaries with different ideas about file offsets, all this is |
2544 | platform-dependent. |
2545 | |
aecce728 |
2546 | =head2 Unicode Support on EBCDIC Still Spotty |
2547 | |
2548 | Though mostly working, Unicode support still has problem spots on |
2549 | EBCDIC platforms. One such known spot are the C<\p{}> and C<\P{}> |
2550 | regular expression constructs for code points less than 256: the |
2551 | pP are testing for Unicode code points, not knowing about EBCDIC. |
2552 | |
f39f21d8 |
2553 | =head2 The Compiler Suite Is Still Experimental |
2554 | |
44da0e71 |
2555 | The compiler suite is slowly getting better but it continues to be |
2556 | highly experimental. Use in production environments is discouraged. |
f39f21d8 |
2557 | |
c4f1ce08 |
2558 | =head2 The Long Double Support Is Still Experimental |
f39f21d8 |
2559 | |
2560 | The ability to configure Perl's numbers to use "long doubles", |
2561 | floating point numbers of hopefully better accuracy, is still |
2562 | experimental. The implementations of long doubles are not yet |
2563 | widespread and the existing implementations are not quite mature |
2564 | or standardised, therefore trying to support them is a rare |
2565 | and moving target. The gain of more precision may also be offset |
2566 | by slowdown in computations (more bits to move around, and the |
2567 | operations are more likely to be executed by less optimised |
2568 | libraries). |
33a87e58 |
2569 | |
c4f1ce08 |
2570 | =head2 Seen In Perl 5.7 But Gone Now |
2571 | |
2572 | Some modules were seen in the Perl 5.7 development releases |
2573 | but are not present in 5.8.0. |
2574 | |
2575 | =over 4 |
2576 | |
2577 | =item * |
2578 | |
2579 | C<Attribute::Handlers> was removed because the implementation of C<my> |
2580 | variable attributes changed so much that the Attribute::Handlers will |
2581 | require a major rewrite. (This means that you can't use |
2582 | Attribute::Handler 0.76 with Perl 5.8.0.) |
2583 | |
2584 | =item * |
2585 | |
2586 | C<Time::Piece> (previously known as C<Time::Object>) was removed |
2587 | because it was felt that it didn't have enough value in it to be a |
2588 | core module. It is still a useful module, though, and is available |
2589 | from the CPAN. |
2590 | |
2591 | =back |
2592 | |
cc0fca54 |
2593 | =head1 Reporting Bugs |
2594 | |
d4ad863d |
2595 | If you find what you think is a bug, you might check the articles |
2596 | recently posted to the comp.lang.perl.misc newsgroup and the perl |
2597 | bug database at http://bugs.perl.org. There may also be |
2598 | information at http://www.perl.com/perl/, the Perl Home Page. |
cc0fca54 |
2599 | |
2600 | If you believe you have an unreported bug, please run the B<perlbug> |
2601 | program included with your release. Be sure to trim your bug down |
2602 | to a tiny but sufficient test case. Your bug report, along with the |
d4ad863d |
2603 | output of C<perl -V>, will be sent off to perlbug@perl.org to be |
cc0fca54 |
2604 | analysed by the Perl porting team. |
2605 | |
2606 | =head1 SEE ALSO |
2607 | |
2608 | The F<Changes> file for exhaustive details on what changed. |
2609 | |
2610 | The F<INSTALL> file for how to build Perl. |
2611 | |
2612 | The F<README> file for general stuff. |
2613 | |
2614 | The F<Artistic> and F<Copying> files for copyright information. |
2615 | |
2616 | =head1 HISTORY |
2617 | |
d468ca04 |
2618 | Written by Jarkko Hietaniemi <F<jhi@iki.fi>>. |
cc0fca54 |
2619 | |
2620 | =cut |