3 perltodo - Perl TO-DO List
7 This is a list of wishes for Perl. Send updates to
8 I<perl5-porters@perl.org>. If you want to work on any of these
9 projects, be sure to check the perl5-porters archives for past ideas,
10 flames, and propaganda. This will save you time and also prevent you
11 from implementing something that Larry has already vetoed. One set
12 of archives may be found at:
14 http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/
16 =head1 To do during 5.6.x
18 =head2 Support for I/O disciplines
20 C<perlio> provides this, but the interface could be a lot more
23 =head2 Autoload bytes.pm
25 When the lexer sees, for instance, C<bytes::length>, it should
26 automatically load the C<bytes> pragma.
28 =head2 Make "\u{XXXX}" et al work
30 Danger, Will Robinson! Discussing the semantics of C<"\x{F00}">,
31 C<"\xF00"> and C<"\U{F00}"> on P5P I<will> lead to a long and boring
34 =head2 Create a char *sv_pvprintify(sv, STRLEN *lenp, UV flags)
36 For displaying PVs with control characters, embedded nulls, and Unicode.
37 This would be useful for printing warnings, or data and regex dumping,
38 not_a_number(), and so on.
40 Requirements: should handle both byte and UTF8 strings. isPRINT()
41 characters printed as-is, character less than 256 as \xHH, Unicode
42 characters as \x{HHH}. Don't assume ASCII-like, either, get somebody
43 on EBCDIC to test the output.
45 Possible options, controlled by the flags:
46 - whitespace (other than ' ' of isPRINT()) printed as-is
47 - use isPRINT_LC() instead of isPRINT()
48 - print control characters like this: "\cA"
49 - print control characters like this: "^A"
50 - non-PRINTables printed as '.' instead of \xHH
51 - use \OOO instead of \xHH
52 - use the C/Perl-metacharacters like \n, \t
53 - have a maximum length for the produced string (read it from *lenp)
54 - append a "..." to the produced string if the maximum length is exceeded
55 - really fancy: print unicode characters as \N{...}
57 NOTE: pv_display(), pv_uni_display(), sv_uni_display() are already
58 doing something like the above.
60 =head2 Overloadable regex assertions
62 This may or may not be possible with the current regular expression
63 engine. The idea is that, for instance, C<\b> needs to be
64 algorithmically computed if you're dealing with Thai text. Hence, the
65 B<\b> assertion wants to be overloaded by a function.
73 Allow for long form of the General Category Properties, e.g
74 C<\p{IsOpenPunctuation}>, not just the abbreviated form, e.g.
79 Allow for the metaproperties: C<XID Start>, C<XID Continue>,
80 C<NF*_NO>, C<NF*_MAYBE> (require the DerivedCoreProperties and
81 DerviceNormalizationProperties files).
83 There are also multiple value properties still unimplemented:
84 C<Numeric Type>, C<East Asian Width>.
88 Case Mappings? http://www.unicode.org/unicode/reports/tr21/
90 lc(), uc(), lcfirst(), and ucfirst() work only for some of the
91 simplest cases, where the mapping goes from a single Unicode character
92 to another single Unicode character. See lib/unicore/SpecCase.txt
97 UTF-8 identifier names should probably be canonicalized: NFC?
101 UTF-8 in package names and sub names? The first is problematic
102 because of the mapping to pathnames, ditto for the second one if
103 one does autosplitting, for example.
107 See L<perlunicode/UNICODE REGULAR EXPRESSION SUPPORT LEVEL> for what's
108 there and what's missing. Almost all of Levels 2 and 3 is missing,
109 and as of 5.8.0 not even all of Level 1 is there.
110 They have some tricks Perl doesn't yet implement, such as character
113 http://www.unicode.org/unicode/reports/tr18/
115 =head2 use Thread for iThreads
117 Artur Bergman's C<iThreads> module is a start on this, but needs to
120 =head2 make perl_clone optionally clone ops
122 So that pseudoforking, mod_perl, iThreads and nvi will work properly
123 (but not as efficiently) until the regex engine is fixed to be threadsafe.
125 =head2 Work out exit/die semantics for threads
127 =head2 Typed lexicals for compiler
129 =head2 Compiler workarounds for Win32
131 =head2 AUTOLOADing in the compiler
133 =head2 Fixing comppadlist when compiling
135 =head2 Cleaning up exported namespace
137 =head2 Complete signal handling
139 Add C<PERL_ASYNC_CHECK> to opcodes which loop; replace C<sigsetjmp> with
140 C<sigjmp>; check C<wait> for signal safety.
142 =head2 Out-of-source builds
144 This was done for 5.6.0, but needs reworking for 5.7.x
146 =head2 POSIX realtime support
148 POSIX 1003.1 1996 Edition support--realtime stuff: POSIX semaphores,
149 message queues, shared memory, realtime clocks, timers, signals (the
150 metaconfig units mostly already exist for these)
152 =head2 UNIX98 support
154 Reader-writer locks, realtime/asynchronous IO
158 There are non-core modules, such as C<Socket6>, but these will need
159 integrating when IPv6 actually starts to really happen. See RFC 2292
162 =head2 Long double conversion
164 Floating point formatting is still causing some weird test failures.
168 Locales and Unicode interact with each other in unpleasant ways.
169 One possible solution would be to adopt/support ICU:
171 http://oss.software.ibm.com/developerworks/opensource/icu/project/
173 =head2 Thread-safe regexes
175 The regular expression engine is currently non-threadsafe.
177 =head2 Arithmetic on non-Arabic numerals
179 C<[1234567890]> aren't the only numerals any more.
181 =head2 POSIX Unicode character classes
183 (C<[=a=]> for equivalance classes, C<[.ch.]> for collation.)
184 These are dependent on Unicode normalization and collation.
186 =head2 Factoring out common suffices/prefices in regexps (trie optimization)
188 Currently, the user has to optimize C<foo|far> and C<foo|goo> into
189 C<f(?:oo|ar)> and C<[fg]oo> by hand; this could be done automatically.
191 =head2 Security audit shipped utilities
193 All the code we ship with Perl needs to be sensible about temporary file
194 handling, locking, input validation, and so on.
196 =head2 Sort out the uid-setting mess
198 Currently there are several problems with the setting of uids ($<, $>
199 for the real and effective uids). Firstly, what exactly setuid() call
200 gets invoked in which platform is simply a big mess that needs to be
201 untangled. Secondly, the effects are apparently not standard across
202 platforms, (if you first set $< and then $>, or vice versa, being
203 uid == euid == zero, or just euid == zero, or as a normal user, what are
204 the results?). The test suite not (usually) being run as root means
205 that these things do not get much testing. Thirdly, there's quite
206 often a third uid called saved uid, and Perl has no knowledge of that
207 feature in any way. (If one has the saved uid of zero, one can get
208 back any real and effective uids.) As an example, to change also the
209 saved uid, one needs to set the real and effective uids B<twice>-- in
210 most systems, that is: in HP-UX that doesn't seem to work.
212 =head2 Custom opcodes
214 Have a way to introduce user-defined opcodes without the subroutine call
215 overhead of an XSUB; the user should be able to create PP code. Simon
216 Cozens has some ideas on this.
218 =head2 DLL Versioning
220 Windows needs a way to know what version of an XS or C<libperl> DLL it's
223 =head2 Introduce @( and @)
225 C<$(> may return "foo bar baz". Unfortunately, since groups can
226 theoretically have spaces in their names, this could be one, two or
229 =head2 Floating point handling
231 C<NaN> and C<inf> support is particularly troublesome.
232 (fp_classify(), fp_class(), fp_class_d(), class(), isinf(),
233 isfinite(), finite(), isnormal(), unordered(), <ieeefp.h>,
234 <fp_class.h> (there are metaconfig units for all these) (I think),
235 fp_setmask(), fp_getmask(), fp_setround(), fp_getround()
236 (no metaconfig units yet for these). Don't forget finitel(), fp_classl(),
237 fp_class_l(), (yes, both do, unfortunately, exist), and unorderedl().)
239 As of Perl 5.6.1, there is a Perl macro, Perl_isnan().
241 =head2 IV/UV preservation
243 Nicholas Clark has done a lot of work on this, but work is continuing.
244 C<+>, C<-> and C<*> work, but guards need to be in place for C<%>, C</>,
245 C<&>, C<oct>, C<hex> and C<pack>.
247 =head2 Replace pod2html with something using Pod::Parser
249 The CPAN module C<Marek::Pod::Html> may be a more suitable basis for a
250 C<pod2html> convertor; the current one duplicates the functionality
251 abstracted in C<Pod::Parser>, which makes updating the POD language
254 =head2 Automate module testing on CPAN
256 When a new Perl is being beta tested, porters have to manually grab
257 their favourite CPAN modules and test them - this should be done
260 =head2 sendmsg and recvmsg
262 We have all the other BSD socket functions but these. There are
263 metaconfig units for these functions which can be added. To avoid these
264 being new opcodes, a solution similar to the way C<sockatmark> was added
265 would be preferable. (Autoload the C<IO::whatever> module.)
267 =head2 Rewrite perlre documentation
269 The new-style patterns need full documentation, and the whole document
270 needs to be a lot clearer.
272 =head2 Convert example code to IO::Handle filehandles
274 =head2 Document Win32 choices
276 =head2 Check new modules
278 =head2 Make roffitall find pods and libs itself
280 Simon Cozens has done some work on this but it needs a rethink.
282 =head1 To do at some point
284 These are ideas that have been regularly tossed around, that most
285 people believe should be done maybe during 5.8.x
287 =head2 Remove regular expression recursion
289 Because the regular expression engine is recursive, badly designed
290 expressions can lead to lots of recursion filling up the stack. Ilya
291 claims that it is easy to convert the engine to being iterative, but
292 this has still not yet been done. There may be a regular expression
293 engine hit squad meeting at TPC5.
295 =head2 Memory leaks after failed eval
297 Perl will leak memory if you C<eval "hlagh hlagh hlagh hlagh">. This is
298 partially because it attempts to build up an op tree for that code and
299 doesn't properly free it. The same goes for non-syntactically-correct
300 regular expressions. Hugo looked into this, but decided it needed a
301 mark-and-sweep GC implementation.
303 Alan notes that: The basic idea was to extend the parser token stack
304 (C<YYSTYPE>) to include a type field so we knew what sort of thing each
305 element of the stack was. The F<perly.c> code would then have to be
306 postprocessed to record the type of each entry on the stack as it was
307 created, and the parser patched so that it could unroll the stack
310 This is possible to do, but would be pretty messy to implement, as it
311 would rely on even more sed hackery in F<perly.fixer>.
313 =head2 bitfields in pack
315 =head2 Cross compilation
317 Make Perl buildable with a cross-compiler. This will play havoc with
318 Configure, which needs to know how the target system will respond to
319 its tests; maybe C<microperl> will be a good starting point here.
320 (Indeed, Bart Schuller reports that he compiled up C<microperl> for
321 the Agenda PDA and it works fine.) A really big spanner in the works
322 is the bootstrapping build process of Perl: if the filesystem the
323 target systems sees is not the same what the build host sees, various
324 input, output, and (Perl) library files need to be copied back and forth.
326 As of 5.8.0 Configure mostly works for cross-compilation
327 (used successfully for iPAQ Linux), miniperl gets built,
328 but then building DynaLoader (and other extensions) fails
329 since MakeMaker knows nothing of cross-compilation.
330 (See INSTALL/Cross-compilation for the state of things.)
332 =head2 Perl preprocessor / macros
334 Source filters help with this, but do not get us all the way. For
335 instance, it should be possible to implement the C<??> operator somehow;
336 source filters don't (quite) cut it.
338 =head2 Perl lexer in Perl
340 Damian Conway is planning to work on this, but it hasn't happened yet.
342 =head2 Using POSIX calls internally
344 When faced with a BSD vs. SysV -style interface to some library or
345 system function, perl's roots show in that it typically prefers the BSD
346 interface (but falls back to the SysV one). One example is getpgrp().
347 Other examples include C<memcpy> vs. C<bcopy>. There are others, mostly in
350 Mostly, this item is a suggestion for which way to start a journey into
351 an C<#ifdef> forest. It is not primarily a suggestion to eliminate any of
352 the C<#ifdef> forests.
354 POSIX calls are perhaps more likely to be portable to unexpected
355 architectures. They are also perhaps more likely to be actively
356 maintained by a current vendor. They are also perhaps more likely to be
357 available in thread-safe versions, if appropriate.
359 =head2 -i rename file when changed
361 It's only necessary to rename a file when inplace editing when the file
362 has changed. Detecting a change is perhaps the difficult bit.
364 =head2 All ARGV input should act like E<lt>E<gt>
366 eg C<read(ARGV, ...)> doesn't currently read across multiple files.
368 =head2 Support for rerunning debugger
370 There should be a way of restarting the debugger on demand.
372 =head2 Test Suite for the Debugger
374 The debugger is a complex piece of software and fixing something
375 here may inadvertently break something else over there. To tame
376 this chaotic behaviour, a test suite is necessary.
378 =head2 my sub foo { }
380 The basic principle is sound, but there are problems with the semantics
381 of self-referential and mutually referential lexical subs: how to
384 =head2 One-pass global destruction
386 Sweeping away all the allocated memory in one go is a laudable goal, but
387 it's difficult and in most cases, it's easier to let the memory get
390 =head2 Rewrite regexp parser
392 There has been talk recently of rewriting the regular expression parser
393 to produce an optree instead of a chain of opcodes; it's unclear whether
394 or not this would be a win.
396 =head2 Cache recently used regexps
400 for my $re (@regexps) {
404 C<qr//> already gives us a way of saving compiled regexps, but it should
405 be done automatically.
407 =head2 Re-entrant functions
409 Add configure probes for C<_r> forms of system calls and fit them to the
410 core. Unfortunately, calling conventions for these functions and not
413 =head2 Cross-compilation support
415 Bart Schuller reports that using C<microperl> and a cross-compiler, he
416 got Perl working on the Agenda PDA. However, one cannot build a full
417 Perl because Configure needs to get the results for the target platform,
420 =head2 Bit-shifting bitvectors
424 vec($v, 1000, 1) = 1;
426 One should be able to do
430 and have the 999'th bit set.
432 Currently if you try with shift bitvectors you shift the NV/UV, instead
433 of the bits in the PV. Not very logical.
435 =head2 debugger pragma
437 The debugger is implemented in Perl in F<perl5db.pl>; turning it into a
438 pragma should be easy, but making it work lexically might be more
439 difficult. Fiddling with C<$^P> would be necessary.
441 =head2 use less pragma
443 Identify areas where speed/memory tradeoffs can be made and have a hint
444 to switch between them.
446 =head2 switch structures
448 Although we have C<Switch.pm> in core, Larry points to the dormant
449 C<nswitch> and C<cswitch> ops in F<pp.c>; using these opcodes would be
452 =head2 Cache eval tree
456 =head2 Shrink opcode tables
458 =head2 Optimize away @_
460 Look at the "reification" code in C<av.c>
462 =head2 Prototypes versus indirect objects
464 Currently, indirect object syntax bypasses prototype checks.
468 HTML versions of the documentation need to be installed by default; a
469 call to C<installhtml> from C<installperl> may be all that's necessary.
471 =head2 Prototype method calls
473 =head2 Return context prototype declarations
477 =head2 Garbage collection
479 There have been persistent mumblings about putting a mark-and-sweep
480 garbage detector into Perl; Alan Burlison has some ideas about this.
484 Mark-Jason Dominus has the beginnings of one of these.
486 =head2 Rewrite perldoc
488 There are a few suggestions for what to do with C<perldoc>: maybe a
489 full-text search, an index function, locating pages on a particular
490 high-level subject, and so on.
492 =head2 Install .3p manpages
494 This is a bone of contention; we can create C<.3p> manpages for each
495 built-in function, but should we install them by default? Tcl does this,
496 and it clutters up C<apropos>.
498 =head2 Unicode tutorial
500 Simon Cozens promises to do this before he gets old.
502 =head2 Update POSIX.pm for 1003.1-2
504 =head2 Retargetable installation
506 Allow C<@INC> to be changed after Perl is built.
508 =head2 POSIX emulation on non-POSIX systems
510 Make C<POSIX.pm> behave as POSIXly as possible everywhere, meaning we
511 have to implement POSIX equivalents for some functions if necessary.
513 =head2 Rename Win32 headers
515 =head2 Finish off lvalue functions
517 They don't work in the debugger, and they don't work for list or hash
520 =head2 Update sprintf documentation
522 Hugo van der Sanden plans to look at this.
524 =head2 Use fchown/fchmod internally
526 This has been done in places, but needs a thorough code review.
527 Also fchdir is available in some platforms.
529 =head2 Make v-strings overloaded objects
531 Instead of having to guess whether a string is a v-string and thus
532 needs to be displayed with %vd, make v-strings (readonly) objects
533 (class "vstring"?) with a stringify overload.
535 =head2 Allow restricted hash assignment
537 Currently you're not allowed to assign to a restricted hash at all,
538 even with the same keys.
540 %restricted = (foo => 42); # error
542 This should be allowed if the new keyset is a subset of the old
543 keyset. May require more extra code than we'd like in pp_aassign.
545 =head2 Should overload be inheritable?
547 Should overload be 'contagious' through @ISA so that derived classes
548 would inherit their base classes' overload definitions? What to do
549 in case of overload conflicts?
553 Ideas which have been discussed, and which may or may not happen.
555 =head2 ref() in list context
557 It's unclear what this should do or how to do it without breaking old
560 =head2 Make tr/// return histogram of characters in list context
562 There is a patch for this, but it may require Unicodification.
564 =head2 Compile to real threaded code
566 =head2 Structured types
568 =head2 Modifiable $1 et al.
570 ($x = "elephant") =~ /e(ph)/;
571 $1 = "g"; # $x = "elegant"
573 What happens if there are multiple (nested?) brackets? What if the
574 string changes between the match and the assignment?
576 =head2 Procedural interfaces for IO::*, etc.
578 Some core modules have been accused of being overly-OO. Adding
579 procedural interfaces could demystify them.
583 =head2 Attach/detach debugger from running program
585 With C<gdb>, you can attach the debugger to a running program if you
586 pass the process ID. It would be good to do this with the Perl debugger
587 on a running Perl program, although I'm not sure how it would be done.
591 A non-core module that would use "native" GUI to create graphical
594 =head2 foreach(reverse ...)
598 foreach (reverse @_) { ... }
600 puts C<@_> on the stack, reverses it putting the reversed version on the
601 stack, then iterates forwards. Instead, it could be special-cased to put
602 C<@_> on the stack then iterate backwards.
604 =head2 Constant function cache
606 =head2 Approximate regular expression matching
610 These items B<always> need doing:
612 =head2 Update guts documentation
614 Simon Cozens tries to do this when possible, and contributions to the
615 C<perlapi> documentation is welcome.
617 =head2 Add more tests
619 Michael Schwern will donate $500 to Yet Another Society when all core
622 =head2 Update auxiliary tools
624 The code we ship with Perl should look like good Perl 5.
626 =head2 Create debugging macros
628 Debugging macros (like printsv, dump) can make debugging perl inside a
629 C debugger much easier. A good set for gdb comes with mod_perl.
630 Something similar should be distributed with perl.
632 The proper way to do this is to use and extend Devel::DebugInit.
633 Devel::DebugInit also needs to be extended to support threads.
635 See p5p archives for late May/early June 2001 for a recent discussion
638 =head2 truncate to the people
640 One can emulate ftruncate() using F_FREESP and F_CHSIZ fcntls
641 (see the UNIX FAQ for details). This needs to go somewhere near
642 pp_sys.c:pp_truncate().
644 One can emulate truncate() easily if one has ftruncate().
645 This emulation should also go near pp_sys.pp_truncate().
647 =head2 Unicode in Filenames
649 chdir, chmod, chown, chroot, exec, glob, link, lstat, mkdir, open, qx,
650 readdir, readlink, rename, rmdir, stat, symlink, sysopen, system,
651 truncate, unlink, utime. All these could potentially accept Unicode
652 filenames either as input or output (and in the case of system and qx
653 Unicode in general, as input or output to/from the shell). Whether a
654 filesystem - an operating system pair understands Unicode in filenames
657 Known combinations that have some level of understanding include
658 Microsoft NTFS, Apple HFS+ (In Mac OS 9 and X) and Apple UFS (in Mac
659 OS X), NFS v4 is rumored to be Unicode, and of course Plan 9. How to
660 create Unicode filenames, what forms of Unicode are accepted and used
661 (UCS-2, UTF-16, UTF-8), what (if any) is the normalization form used,
662 and so on, varies. Finding the right level of interfacing to Perl
663 requires some thought. Remember that an OS does not implicate a
666 Note that in Windows the -C command line flag already does quite
667 a bit of the above (but even there the support is not complete:
668 for example the exec/spawn are not Unicode-aware) by turning on
669 the so-called "wide API support".
671 =head1 Recently done things
673 These are things which have been on the todo lists in previous releases
674 but have recently been completed.
676 =head2 Alternative RE syntax module
678 The C<Regexp::English> module, available from the CPAN, provides this:
680 my $re = Regexp::English
682 -> literal('Flippers')
693 =head2 Safe signal handling
695 A new signal model went into 5.7.1 without much fanfare. Operations and
696 C<malloc>s are no longer interrupted by signals, which are handled
697 between opcodes. This means that C<PERL_ASYNC_CHECK> now actually does
698 something. However, there are still a few things that need to be done.
702 Modules which implement arrays in terms of strings, substrings or files
703 can be found on the CPAN.
707 C<Time::HiRes> has been integrated into the core.
709 =head2 setitimer and getimiter
711 Adding C<Time::HiRes> got us this too.
713 =head2 Testing __DIE__ hook
715 Tests have been added.
717 =head2 CPP equivalent in Perl
719 A C Yardley will probably have done this by the time you can read this.
720 This allows for a generalization of the C constant detection used in
721 building C<Errno.pm>.
723 =head2 Explicit switch statements
725 C<Switch.pm> has been integrated into the core to give you all manner of
726 C<switch...case> semantics.
734 Nick Ing-Simmons has made UTF-EBCDIC (UTR13) work with Perl.
736 EBCDIC? http://www.unicode.org/unicode/reports/tr16/
740 Although there are probably some small bugs to be rooted out, Jarkko
741 Hietaniemi has made regular expressions polymorphic between bytes and
744 =head2 perlcc to produce executable
746 C<perlcc> was recently rewritten, and can now produce standalone
749 =head2 END blocks saved in compiled output
751 =head2 Secure temporary file module
753 Tim Jenness' C<File::Temp> is now in core.
755 =head2 Integrate Time::HiRes
757 This module is now part of core.
759 =head2 Turn Cwd into XS
761 Benjamin Sugars has done this.
763 =head2 Mmap for input
765 Nick Ing-Simmons' C<perlio> supports an C<mmap> IO method.
767 =head2 Byte to/from UTF8 and UTF8 to/from local conversion
769 C<Encode> provides this.
771 =head2 Add sockatmark support
775 =head2 Mailing list archives
777 http://lists.perl.org/ , http://archive.develooper.com/
781 Richard Foley has written the bug tracking system at http://bugs.perl.org/
783 =head2 Integrate MacPerl
785 Chris Nandor and Matthias Neeracher have integrated the MacPerl changes
788 =head2 Web "nerve center" for Perl
790 http://use.perl.org/ is what you're looking for.
792 =head2 Regular expression tutorial
794 C<perlretut>, provided by Mark Kvale.
796 =head2 Debugging Tutorial
798 C<perldebtut>, written by Richard Foley.
800 =head2 Integrate new modules
802 Jarkko has been integrating madly into 5.7.x
804 =head2 Integrate profiler
806 C<Devel::DProf> is now a core module.
808 =head2 Y2K error detection
810 There's a configure option to detect unsafe concatenation with "19", and
811 a CPAN module. (C<D'oh::Year>)
813 =head2 Regular expression debugger
815 While not part of core, Mark-Jason Dominus has written C<Rx> and has
816 also come up with a generalised strategy for regular expression
821 That's, uh, F<podchecker>
823 =head2 "Dynamic" lexicals
825 =head2 Cache precompiled modules
827 =head1 Deprecated Wishes
829 These are items which used to be in the todo file, but have been
830 deprecated for some reason.
832 =head2 Loop control on do{}
834 This would break old code; use C<do{{ }}> instead.
836 =head2 Lexically scoped typeglobs
838 Not needed now we have lexical IO handles.
844 Damian Conway's text formatting modules seem to be the Way To Go.
846 =head2 Generalised want()/caller())
848 Robin Houston's C<Want> module does this.
850 =head2 Named prototypes
852 This seems to be delayed until Perl 6.
854 =head2 Built-in globbing
856 The C<File::Glob> module has been used to replace the C<glob> function.
858 =head2 Regression tests for suidperl
860 C<suidperl> is deprecated in favour of common sense.
862 =head2 Cached hash values
864 We have shared hash keys, which perform the same job.
866 =head2 Add compression modules
868 The compression modules are a little heavy; meanwhile, Nick Clark is
869 working on experimental pragmata to do transparent decompression on
872 =head2 Reorganise documentation into tutorials/references
874 Could not get consensus on P5P about this.
876 =head2 Remove distinction between functions and operators
878 Caution: highly flammable.
880 =head2 Make XS easier to use
882 Use C<Inline> instead, or SWIG.
884 =head2 Make embedding easier to use
890 See the Perl Power Tools. ( http://language.perl.com/ppt/ )
892 =head2 my $Package::variable
896 =head2 "or" tests defined, not truth
898 Suggesting this on P5P B<will> cause a boring and interminable flamewar.
900 =head2 "class"-based lexicals
902 Use flyweight objects, secure hashes or, dare I say it, pseudo-hashes instead.
903 (Or whatever will replace pseudohashes in 5.10.)
907 C<ByteLoader> covers this.
909 =head2 Lazy evaluation / tail recursion removal
911 C<List::Util> gives first() (a short-circuiting grep); tail recursion
912 removal is done manually, with C<goto &whoami;>. (However, MJD has
913 found that C<goto &whoami> introduces a performance penalty, so maybe
914 there should be a way to do this after all: C<sub foo {START: ... goto
917 =head2 Make "use utf8" the default
919 Because of backward compatibility this is difficult: scripts could not
920 contain B<any legacy eight-bit data> (like Latin-1) anymore, even in
921 string literals or pod. Also would introduce a measurable slowdown of
922 at least few percentages since all regular expression operations would
923 be done in full UTF-8. But if you want to try this, add
924 -DUSE_UTF8_SCRIPTS to your compilation flags.
926 =head2 Unicode collation and normalization
928 The Unicode::Collate and Unicode::Normalize modules
929 by SADAHIRO Tomoyuki have been included since 5.8.0.
931 Collation? http://www.unicode.org/unicode/reports/tr10/
932 Normalization? http://www.unicode.org/unicode/reports/tr15/
934 =head2 pack/unpack tutorial
936 Wolfgang Laun finished what Simon Cozens started.