This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Update to Unicode 3.1.1.
[perl5.git] / pod / perltodo.pod
CommitLineData
7711098a
GS
1=head1 NAME
2
3perltodo - Perl TO-DO List
4
5=head1 DESCRIPTION
e50bb9a1 6
722d2a37 7This is a list of wishes for Perl. Send updates to
e50bb9a1
GS
8I<perl5-porters@perl.org>. If you want to work on any of these
9projects, be sure to check the perl5-porters archives for past ideas,
10flames, and propaganda. This will save you time and also prevent you
11from implementing something that Larry has already vetoed. One set
12of archives may be found at:
13
14 http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/
15
722d2a37 16=head1 To do during 5.6.x
e50bb9a1 17
722d2a37 18=head2 Support for I/O disciplines
e50bb9a1 19
722d2a37
SC
20C<perlio> provides this, but the interface could be a lot more
21straightforward.
e50bb9a1 22
4b3b956a 23=head2 Autoload bytes.pm
e50bb9a1 24
4b3b956a
JH
25When the lexer sees, for instance, C<bytes::length>, it should
26automatically load the C<bytes> pragma.
27
28=head2 Make "\u{XXXX}" et al work
29
30Danger, Will Robinson! Discussing the semantics of C<"\x{F00}">,
31C<"\xF00"> and C<"\U{F00}"> on P5P I<will> lead to a long and boring
32flamewar.
e50bb9a1 33
c6287c21 34=head2 Create a char *sv_pvprintify(sv, STRLEN *lenp, UV flags)
0562c0e3
JH
35
36For displaying PVs with control characters, embedded nulls, and Unicode.
37This would be useful for printing warnings, or data and regex dumping,
38not_a_number(), and so on.
39
f35392ae
JH
40Requirements: should handle both byte and UTF8 strings. isPRINT()
41characters printed as-is, character less than 256 as \xHH, Unicode
0661e9a4
JH
42characters as \x{HHH}. Don't assume ASCII-like, either, get somebody
43on EBCDIC to test the output.
f35392ae
JH
44
45Possible options, controlled by the flags:
0661e9a4 46- whitespace (other than ' ' of isPRINT()) printed as-is
f35392ae
JH
47- use isPRINT_LC() instead of isPRINT()
48- print control characters like this: "\cA"
49- print control characters like this: "^A"
0661e9a4
JH
50- non-PRINTables printed as '.' instead of \xHH
51- use \OOO instead of \xHH
52- use the C/Perl-metacharacters like \n, \t
f35392ae
JH
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
0661e9a4 55- really fancy: print unicode characters as \N{...}
f35392ae 56
722d2a37 57=head2 Overloadable regex assertions
e50bb9a1 58
722d2a37
SC
59This may or may not be possible with the current regular expression
60engine. The idea is that, for instance, C<\b> needs to be
61algorithmically computed if you're dealing with Thai text. Hence, the
62B<\b> assertion wants to be overloaded by a function.
e50bb9a1 63
722d2a37 64=head2 Unicode collation and normalization
e50bb9a1 65
722d2a37 66Simon Cozens promises to work on this.
e50bb9a1 67
722d2a37
SC
68 Collation? http://www.unicode.org/unicode/reports/tr10/
69 Normalization? http://www.unicode.org/unicode/reports/tr15/
e50bb9a1 70
722d2a37 71=head2 Unicode case mappings
e50bb9a1 72
722d2a37 73 Case Mappings? http://www.unicode.org/unicode/reports/tr21/
e50bb9a1 74
722d2a37 75=head2 Unicode regular expression character classes
e50bb9a1 76
c6287c21
JH
77They have some tricks Perl doesn't yet implement like character
78class subtraction.
e50bb9a1 79
722d2a37 80 http://www.unicode.org/unicode/reports/tr18/
e50bb9a1 81
722d2a37 82=head2 use Thread for iThreads
e50bb9a1 83
722d2a37
SC
84Artur Bergman's C<iThreads> module is a start on this, but needs to
85be more mature.
e50bb9a1 86
dd0afe54
AB
87=head2 make perl_clone optionally clone ops
88
89So that pseudoforking, mod_perl, iThreads and nvi will work properly
90(but not as efficiently) until the regex engine is fixed to be threadsafe.
91
722d2a37 92=head2 Work out exit/die semantics for threads
e50bb9a1 93
722d2a37 94=head2 Typed lexicals for compiler
e50bb9a1 95
722d2a37 96=head2 Compiler workarounds for Win32
e50bb9a1 97
722d2a37 98=head2 AUTOLOADing in the compiler
e50bb9a1 99
722d2a37 100=head2 Fixing comppadlist when compiling
e50bb9a1 101
722d2a37 102=head2 Cleaning up exported namespace
e50bb9a1 103
722d2a37 104=head2 Complete signal handling
e50bb9a1 105
722d2a37
SC
106Add C<PERL_ASYNC_CHECK> to opcodes which loop; replace C<sigsetjmp> with
107C<sigjmp>; check C<wait> for signal safety.
e50bb9a1 108
722d2a37 109=head2 Out-of-source builds
e50bb9a1 110
722d2a37 111This was done for 5.6.0, but needs reworking for 5.7.x
e50bb9a1 112
722d2a37 113=head2 POSIX realtime support
e50bb9a1 114
722d2a37
SC
115POSIX 1003.1 1996 Edition support--realtime stuff: POSIX semaphores,
116message queues, shared memory, realtime clocks, timers, signals (the
117metaconfig units mostly already exist for these)
e50bb9a1 118
722d2a37 119=head2 UNIX98 support
e50bb9a1 120
722d2a37 121Reader-writer locks, realtime/asynchronous IO
e50bb9a1 122
722d2a37 123=head2 IPv6 Support
e50bb9a1 124
722d2a37
SC
125There are non-core modules, such as C<Net::IPv6>, but these will need
126integrating when IPv6 actually starts to really happen. See RFC 2292
127and RFC 2553.
e50bb9a1 128
722d2a37 129=head2 Long double conversion
e50bb9a1 130
722d2a37 131Floating point formatting is still causing some weird test failures.
e50bb9a1 132
722d2a37 133=head2 Locales
e50bb9a1 134
722d2a37
SC
135Locales and Unicode interact with each other in unpleasant ways.
136One possible solution would be to adopt/support ICU:
e50bb9a1 137
722d2a37 138 http://oss.software.ibm.com/developerworks/opensource/icu/project/
e50bb9a1 139
722d2a37 140=head2 Thread-safe regexes
e50bb9a1 141
722d2a37 142The regular expression engine is currently non-threadsafe.
e50bb9a1 143
722d2a37 144=head2 Arithmetic on non-Arabic numerals
e50bb9a1 145
722d2a37 146C<[1234567890]> aren't the only numerals any more.
e50bb9a1 147
722d2a37 148=head2 POSIX Unicode character classes
e50bb9a1 149
722d2a37
SC
150([=a=] for equivalance classes, [.ch.] for collation.)
151These are dependent on Unicode normalization and collation.
e50bb9a1 152
722d2a37 153=head2 Factoring out common suffices/prefices in regexps (trie optimization)
c47ff5f1 154
722d2a37
SC
155Currently, the user has to optimize C<foo|far> and C<foo|goo> into
156C<f(?:oo|ar)> and C<[fg]oo> by hand; this could be done automatically.
e50bb9a1 157
722d2a37 158=head2 Security audit shipped utilities
e50bb9a1 159
722d2a37
SC
160All the code we ship with Perl needs to be sensible about temporary file
161handling, locking, input validation, and so on.
e50bb9a1 162
722d2a37 163=head2 Custom opcodes
e50bb9a1 164
722d2a37
SC
165Have a way to introduce user-defined opcodes without the subroutine call
166overhead of an XSUB; the user should be able to create PP code. Simon
167Cozens has some ideas on this.
e50bb9a1 168
722d2a37 169=head2 spawnvp() on Win32
e50bb9a1 170
722d2a37
SC
171Win32 has problems spawning processes, particularly when the arguments
172to the child process contain spaces, quotes or tab characters.
e50bb9a1 173
722d2a37 174=head2 DLL Versioning
e50bb9a1 175
722d2a37
SC
176Windows needs a way to know what version of a XS or C<libperl> DLL it's
177loading.
e50bb9a1 178
722d2a37 179=head2 Introduce @( and @)
e50bb9a1 180
722d2a37
SC
181C<$(> may return "foo bar baz". Unfortunately, since groups can
182theoretically have spaces in their names, this could be one, two or
183three groups.
e50bb9a1 184
722d2a37 185=head2 Floating point handling
e50bb9a1 186
722d2a37
SC
187C<NaN> and C<inf> support is particularly troublesome.
188(fp_classify(), fp_class(), fp_class_d(), class(), isinf(),
189isfinite(), finite(), isnormal(), unordered(), <ieeefp.h>,
190<fp_class.h> (there are metaconfig units for all these) (I think),
191fp_setmask(), fp_getmask(), fp_setround(), fp_getround()
192(no metaconfig units yet for these). Don't forget finitel(), fp_classl(),
193fp_class_l(), (yes, both do, unfortunately, exist), and unorderedl().)
e50bb9a1 194
722d2a37 195As of Perl 5.6.1 is a Perl macro, Perl_isnan().
e50bb9a1 196
722d2a37 197=head2 IV/UV preservation
e50bb9a1 198
722d2a37
SC
199Nicholas Clark has done a lot of work on this, but work is continuing.
200C<+>, C<-> and C<*> work, but guards need to be in place for C<%>, C</>,
201C<&>, C<oct>, C<hex> and C<pack>.
e50bb9a1 202
722d2a37 203=head2 Replace pod2html with something using Pod::Parser
83df6a1d 204
722d2a37
SC
205The CPAN module C<Malik::Pod::Html> may be a more suitable basis for a
206C<pod2html> convertor; the current one duplicates the functionality
207abstracted in C<Pod::Parser>, which makes updating the POD language
208difficult.
e50bb9a1 209
722d2a37 210=head2 Automate module testing on CPAN
e50bb9a1 211
722d2a37
SC
212When a new Perl is being beta tested, porters have to manually grab
213their favourite CPAN modules and test them - this should be done
214automatically.
e50bb9a1 215
722d2a37 216=head2 sendmsg and recvmsg
83df6a1d 217
722d2a37
SC
218We have all the other BSD socket functions but these. There are
219metaconfig units for these functions which can be added. To avoid these
220being new opcodes, a solution similar to the way C<sockatmark> was added
221would be preferable. (Autoload the C<IO::whatever> module.)
e50bb9a1 222
722d2a37 223=head2 Rewrite perlre documentation
e50bb9a1 224
722d2a37
SC
225The new-style patterns need full documentation, and the whole document
226needs to be a lot clearer.
e50bb9a1 227
722d2a37 228=head2 Convert example code to IO::Handle filehandles
e50bb9a1 229
722d2a37 230=head2 Document Win32 choices
e50bb9a1 231
722d2a37 232=head2 Check new modules
e50bb9a1 233
722d2a37 234=head2 Make roffitall find pods and libs itself
e50bb9a1 235
722d2a37 236Simon Cozens has done some work on this but it needs a rethink.
e50bb9a1 237
722d2a37 238=head1 To do at some point
e50bb9a1 239
722d2a37
SC
240These are ideas that have been regularly tossed around, that most
241people believe should be done maybe during 5.8.x
e50bb9a1 242
722d2a37 243=head2 Remove regular expression recursion
e50bb9a1 244
722d2a37
SC
245Because the regular expression engine is recursive, badly designed
246expressions can lead to lots of recursion filling up the stack. Ilya
247claims that it is easy to convert the engine to being iterative, but
248this has still not yet been done. There may be a regular expression
249engine hit squad meeting at TPC5.
e50bb9a1 250
722d2a37 251=head2 Memory leaks after failed eval
e50bb9a1 252
722d2a37
SC
253Perl will leak memory if you C<eval "hlagh hlagh hlagh hlagh">. This is
254partially because it attempts to build up an op tree for that code and
255doesn't properly free it. The same goes for non-syntactically-correct
256regular expressions. Hugo looked into this, but decided it needed a
257mark-and-sweep GC implementation.
e50bb9a1 258
722d2a37
SC
259Alan notes that: The basic idea was to extend the parser token stack
260(C<YYSTYPE>) to include a type field so we knew what sort of thing each
261element of the stack was. The F<<perly.c> code would then have to be
262postprocessed to record the type of each entry on the stack as it was
263created, and the parser patched so that it could unroll the stack
264properly on error.
e50bb9a1 265
722d2a37
SC
266This is possible to do, but would be pretty messy to implement, as it
267would rely on even more sed hackery in F<perly.fixer>.
e50bb9a1 268
722d2a37 269=head2 pack "(stuff)*"
e50bb9a1 270
722d2a37 271That's to say, C<pack "(sI)40"> would be the same as C<pack "sI"x40>
e50bb9a1 272
722d2a37 273=head2 bitfields in pack
e50bb9a1 274
722d2a37 275=head2 Cross compilation
e50bb9a1 276
722d2a37
SC
277Make Perl buildable with a cross-compiler. This will play havoc with
278Configure, which needs to how how the target system will respond to
279its tests; maybe C<microperl> will be a good starting point here.
280(Indeed, Bart Schuller reports that he compiled up C<microperl> for
281the Agenda PDA and it works fine.) A really big spanner in the works
282is the bootstrapping build process of Perl: if the filesystem the
283target systems sees is not the same what the build host sees, various
284input, output, and (Perl) library files need to be copied back and forth.
e50bb9a1 285
f86a8bc5
JH
286As of 5.8.0 Configure mostly works for cross-compilation
287(used successfully for iPAQ Linux), miniperl gets built,
288but then building DynaLoader (and other extensions) fails
289since MakeMaker knows nothing of cross-compilation.
290(See INSTALL/Cross-compilation for the state of things.)
291
722d2a37 292=head2 Perl preprocessor / macros
e50bb9a1 293
722d2a37
SC
294Source filters help with this, but do not get us all the way. For
295instance, it should be possible to implement the C<??> operator somehow;
296source filters don't (quite) cut it.
e50bb9a1 297
722d2a37 298=head2 Perl lexer in Perl
a45bd81d 299
722d2a37 300Damian Conway is planning to work on this, but it hasn't happened yet.
e50bb9a1 301
722d2a37 302=head2 Using POSIX calls internally
e50bb9a1 303
722d2a37
SC
304When faced with a BSD vs. SySV -style interface to some library or
305system function, perl's roots show in that it typically prefers the BSD
306interface (but falls back to the SysV one). One example is getpgrp().
307Other examples include C<memcpy> vs. C<bcopy>. There are others, mostly in
308F<<pp_sys.c>.
e50bb9a1 309
722d2a37
SC
310Mostly, this item is a suggestion for which way to start a journey into
311an C<#ifdef> forest. It is not primarily a suggestion to eliminate any of
312the C<#ifdef> forests.
e50bb9a1 313
722d2a37
SC
314POSIX calls are perhaps more likely to be portable to unexpected
315architectures. They are also perhaps more likely to be actively
316maintained by a current vendor. They are also perhaps more likely to be
317available in thread-safe versions, if appropriate.
e50bb9a1 318
722d2a37 319=head2 -i rename file when changed
e50bb9a1 320
722d2a37
SC
321It's only necessary to rename a file when inplace editing when the file
322has changed. Detecting a change is perhaps the difficult bit.
e50bb9a1 323
722d2a37 324=head2 All ARGV input should act like E<lt>E<gt>
e50bb9a1 325
722d2a37 326=head2 Support for rerunning debugger
e50bb9a1 327
722d2a37 328There should be a way of restarting the debugger on demand.
e50bb9a1 329
c6287c21
JH
330=head2 Test Suite for the Debugger
331
332The debugger is a complex piece of software and fixing something
333here may inadvertently break something else over there. To tame
334this chaotic behaviour, a test suite is necessary.
335
722d2a37 336=head2 my sub foo { }
c47ff5f1 337
722d2a37
SC
338The basic principle is sound, but there are problems with the semantics
339of self-referential and mutually referential lexical subs: how to
340declare the subs?
c47ff5f1 341
722d2a37 342=head2 One-pass global destruction
c47ff5f1 343
722d2a37
SC
344Sweeping away all the allocated memory in one go is a laudable goal, but
345it's difficult and in most cases, it's easier to let the memory get
346freed by exiting.
e50bb9a1 347
722d2a37 348=head2 Rewrite regexp parser
e50bb9a1 349
722d2a37
SC
350There has been talk recently of rewriting the regular expression parser
351to produce an optree instead of a chain of opcodes; it's unclear whether
352or not this would be a win.
e50bb9a1 353
722d2a37 354=head2 Cache recently used regexps
e50bb9a1 355
722d2a37 356This is to speed up
e50bb9a1 357
722d2a37
SC
358 for my $re (@regexps) {
359 $matched++ if /$re/
360 }
e50bb9a1 361
722d2a37
SC
362C<qr//> already gives us a way of saving compiled regexps, but it should
363be done automatically.
e50bb9a1 364
722d2a37 365=head2 Re-entrant functions
e50bb9a1 366
722d2a37
SC
367Add configure probes for C<_r> forms of system calls and fit them to the
368core. Unfortunately, calling conventions for these functions and not
369standardised.
04c70446 370
722d2a37 371=head2 Cross-compilation support
04c70446 372
722d2a37
SC
373Bart Schuller reports that using C<microperl> and a cross-compiler, he
374got Perl working on the Agenda PDA. However, one cannot build a full
375Perl because Configure needs to get the results for the target platform,
376for the host.
e50bb9a1 377
722d2a37 378=head2 Bit-shifting bitvectors
e50bb9a1 379
722d2a37 380Given:
e50bb9a1 381
722d2a37 382 vec($v, 1000, 1) = 1;
e50bb9a1 383
722d2a37 384One should be able to do
e50bb9a1 385
722d2a37 386 $v <<= 1;
e50bb9a1 387
722d2a37 388and have the 999'th bit set.
e50bb9a1 389
722d2a37
SC
390Currently if you try with shift bitvectors you shift the NV/UV, instead
391of the bits in the PV. Not very logical.
e50bb9a1 392
722d2a37 393=head2 debugger pragma
e50bb9a1 394
722d2a37
SC
395The debugger is implemented in Perl in F<perl5db.pl>; turning it into a
396pragma should be easy, but making it work lexically might be more
397difficult. Fiddling with C<$^P> would be necessary.
e50bb9a1 398
722d2a37 399=head2 use less pragma
e50bb9a1 400
722d2a37
SC
401Identify areas where speed/memory tradeoffs can be made and have a hint
402to switch between them.
e50bb9a1 403
722d2a37 404=head2 switch structures
e50bb9a1 405
722d2a37
SC
406Although we have C<Switch.pm> in core, Larry points to the dormant
407C<nswitch> and C<cswitch> ops in F<pp.c>; using these opcodes would be
408much faster.
e50bb9a1 409
722d2a37 410=head2 Cache eval tree
e50bb9a1 411
722d2a37 412=head2 rcatmaybe
e50bb9a1 413
722d2a37 414=head2 Shrink opcode tables
e50bb9a1 415
722d2a37 416=head2 Optimize away @_
e50bb9a1 417
722d2a37 418Look at the "reification" code in C<av.c>
e50bb9a1 419
722d2a37 420=head2 Prototypes versus indirect objects
e50bb9a1 421
722d2a37 422Currently, indirect object syntax bypasses prototype checks.
e50bb9a1 423
722d2a37 424=head2 Install HMTL
e50bb9a1 425
722d2a37
SC
426HTML versions of the documentation need to be installed by default; a
427call to C<installhtml> from C<installperl> may be all that's necessary.
e50bb9a1 428
722d2a37 429=head2 Prototype method calls
e50bb9a1 430
722d2a37 431=head2 Return context prototype declarations
e50bb9a1 432
722d2a37 433=head2 magic_setisa
e50bb9a1 434
722d2a37 435=head2 Garbage collection
e50bb9a1 436
722d2a37
SC
437There have been persistent mumblings about putting a mark-and-sweep
438garbage detector into Perl; Alan Burlison has some ideas about this.
e50bb9a1 439
722d2a37 440=head2 IO tutorial
e50bb9a1 441
722d2a37 442Mark-Jason Dominus has the beginnings of one of these.
e50bb9a1 443
722d2a37 444=head2 pack/unpack tutorial
e50bb9a1 445
722d2a37 446Simon Cozens has the beginnings of one of these.
e50bb9a1 447
722d2a37 448=head2 Rewrite perldoc
e50bb9a1 449
722d2a37
SC
450There are a few suggestions for what to do with C<perldoc>: maybe a
451full-text search, an index function, locating pages on a particular
452high-level subject, and so on.
e50bb9a1 453
3958b146 454=head2 Install .3p manpages
e50bb9a1 455
3958b146 456This is a bone of contention; we can create C<.3p> manpages for each
722d2a37
SC
457built-in function, but should we install them by default? Tcl does this,
458and it clutters up C<apropos>.
e50bb9a1 459
722d2a37 460=head2 Unicode tutorial
e50bb9a1 461
722d2a37 462Simon Cozens promises to do this before he gets old.
e50bb9a1 463
722d2a37 464=head2 Update POSIX.pm for 1003.1-2
3958b146 465
722d2a37 466=head2 Retargetable installation
e50bb9a1 467
722d2a37 468Allow C<@INC> to be changed after Perl is built.
e50bb9a1 469
722d2a37 470=head2 POSIX emulation on non-POSIX systems
e50bb9a1 471
722d2a37
SC
472Make C<POSIX.pm> behave as POSIXly as possible everywhere, meaning we
473have to implement POSIX equivalents for some functions if necessary.
e50bb9a1 474
722d2a37 475=head2 Rename Win32 headers
e50bb9a1 476
722d2a37
SC
477=head2 Finish off lvalue functions
478
479They don't work in the debugger, and they don't work for list or hash
480slices.
e50bb9a1 481
722d2a37 482=head2 Update sprintf documentation
e50bb9a1 483
722d2a37 484Hugo van der Sanden plans to look at this.
e50bb9a1 485
722d2a37 486=head2 Use fchown/fchmod internally
e50bb9a1 487
722d2a37
SC
488This has been done in places, but needs a thorough code review.
489Also fchdir is available in some platforms.
e50bb9a1 490
722d2a37 491=head1 Vague ideas
e50bb9a1 492
722d2a37 493Ideas which have been discussed, and which may or may not happen.
e50bb9a1 494
722d2a37 495=head2 ref() in list context
e50bb9a1 496
722d2a37
SC
497It's unclear what this should do or how to do it without breaking old
498code.
e50bb9a1 499
f86a8bc5 500=head2 Make tr/// return histogram of characters in list context
e50bb9a1 501
722d2a37 502There is a patch for this, but it may require Unicodification.
e50bb9a1 503
722d2a37 504=head2 Compile to real threaded code
3958b146 505
722d2a37 506=head2 Structured types
3958b146 507
722d2a37 508=head2 Modifiable $1 et al.
e50bb9a1 509
722d2a37
SC
510 ($x = "elephant") =~ /e(ph)/;
511 $1 = "g"; # $x = "elegant"
e50bb9a1 512
722d2a37
SC
513What happens if there are multiple (nested?) brackets? What if the
514string changes between the match and the assignment?
e50bb9a1 515
722d2a37 516=head2 Procedural interfaces for IO::*, etc.
e50bb9a1 517
722d2a37
SC
518Some core modules have been accused of being overly-OO. Adding
519procedural interfaces could demystify them.
e50bb9a1 520
722d2a37 521=head2 RPC modules
e50bb9a1 522
722d2a37 523=head2 Attach/detach debugger from running program
e50bb9a1 524
722d2a37
SC
525With C<gdb>, you can attach the debugger to a running program if you
526pass the process ID. It would be good to do this with the Perl debugger
527on a running Perl program, although I'm not sure how it would be done.
e50bb9a1 528
722d2a37 529=head2 Alternative RE syntax module
e50bb9a1 530
722d2a37
SC
531 use Regex::Newbie;
532 $re = Regex::Newbie->new
533 ->start
534 ->match("foo")
535 ->repeat(Regex::Newbie->class("char"),3)
536 ->end;
537 /$re/;
e50bb9a1 538
722d2a37 539=head2 GUI::Native
e50bb9a1 540
722d2a37
SC
541A non-core module that would use "native" GUI to create graphical
542applications.
e50bb9a1 543
722d2a37 544=head2 foreach(reverse ...)
e50bb9a1 545
722d2a37 546Currently
e50bb9a1 547
722d2a37 548 foreach (reverse @_) { ... }
e50bb9a1 549
722d2a37
SC
550puts C<@_> on the stack, reverses it putting the reversed version on the
551stack, then iterates forwards. Instead, it could be special-cased to put
552C<@_> on the stack then iterate backwards.
e50bb9a1 553
722d2a37 554=head2 Constant function cache
e50bb9a1 555
722d2a37 556=head2 Approximate regular expression matching
e50bb9a1 557
722d2a37 558=head1 Ongoing
e50bb9a1 559
722d2a37 560These items B<always> need doing:
e50bb9a1 561
722d2a37 562=head2 Update guts documentation
e50bb9a1 563
722d2a37
SC
564Simon Cozens tries to do this when possible, and contributions to the
565C<perlapi> documentation is welcome.
e50bb9a1 566
722d2a37 567=head2 Add more tests
e50bb9a1 568
722d2a37
SC
569Michael Schwern will donate $500 to Yet Another Society when all core
570modules have tests.
e50bb9a1 571
722d2a37 572=head2 Update auxiliary tools
e50bb9a1 573
722d2a37 574The code we ship with Perl should look like good Perl 5.
e50bb9a1 575
722d2a37 576=head1 Recently done things
e50bb9a1 577
722d2a37
SC
578These are things which have been on the todo lists in previous releases
579but have recently been completed.
e50bb9a1 580
722d2a37 581=head2 Safe signal handling
e50bb9a1 582
722d2a37
SC
583A new signal model went into 5.7.1 without much fanfare. Operations and
584C<malloc>s are no longer interrupted by signals, which are handled
585between opcodes. This means that C<PERL_ASYNC_CHECK> now actually does
586something. However, there are still a few things that need to be done.
e50bb9a1 587
722d2a37 588=head2 Tie Modules
e50bb9a1 589
722d2a37
SC
590Modules which implement arrays in terms of strings, substrings or files
591can be found on the CPAN.
e50bb9a1 592
722d2a37 593=head2 gettimeofday
e50bb9a1 594
722d2a37 595C<Time::Hires> has been integrated into the core.
e50bb9a1 596
722d2a37 597=head2 setitimer and getimiter
e50bb9a1 598
722d2a37 599Adding C<Time::Hires> got us this too.
e50bb9a1 600
722d2a37
SC
601=head2 Testing __DIE__ hook
602
603Tests have been added.
604
605=head2 CPP equivalent in Perl
e50bb9a1 606
722d2a37
SC
607A C Yardley will probably have done this by the time you can read this.
608This allows for a generalization of the C constant detection used in
609building C<Errno.pm>.
e50bb9a1 610
722d2a37 611=head2 Explicit switch statements
e50bb9a1 612
722d2a37
SC
613C<Switch.pm> has been integrated into the core to give you all manner of
614C<switch...case> semantics.
e50bb9a1 615
722d2a37 616=head2 autocroak
e50bb9a1 617
722d2a37 618This is C<Fatal.pm>.
e50bb9a1 619
722d2a37 620=head2 UTF/EBCDIC
e50bb9a1 621
722d2a37 622Nick Ing-Simmons has made UTF-EBCDIC (UTR13) work with Perl.
e50bb9a1 623
722d2a37 624 EBCDIC? http://www.unicode.org/unicode/reports/tr16/
e50bb9a1 625
722d2a37 626=head2 UTF Regexes
e50bb9a1 627
722d2a37
SC
628Although there are probably some small bugs to be rooted out, Jarkko
629Hietaniemi has made regular expressions polymorphic between bytes and
630characters.
e50bb9a1 631
722d2a37 632=head2 perlcc to produce executable
e50bb9a1 633
722d2a37
SC
634C<perlcc> was recently rewritten, and can now produce standalone
635executables.
e50bb9a1 636
722d2a37 637=head2 END blocks saved in compiled output
e50bb9a1 638
722d2a37 639=head2 Secure temporary file module
e50bb9a1 640
722d2a37 641Tim Jenness' C<File::Temp> is now in core.
e50bb9a1 642
722d2a37 643=head2 Integrate Time::HiRes
e50bb9a1 644
722d2a37 645This module is now part of core.
e50bb9a1 646
722d2a37 647=head2 Turn Cwd into XS
e50bb9a1 648
722d2a37 649Benjamin Sugars has done this.
e50bb9a1 650
722d2a37 651=head2 Mmap for input
e50bb9a1 652
722d2a37 653Nick Ing-Simmons' C<perlio> supports an C<mmap> IO method.
e50bb9a1 654
722d2a37 655=head2 Byte to/from UTF8 and UTF8 to/from local conversion
e50bb9a1 656
722d2a37 657C<Encode> provides this.
e50bb9a1 658
722d2a37 659=head2 Add sockatmark support
e50bb9a1 660
722d2a37 661Added in 5.7.1
e50bb9a1 662
722d2a37
SC
663=head2 Mailing list archives
664
665http://lists.perl.org/, http://archive.develooper.com/
666
667=head2 Bug tracking
668
669Richard Foley has written the bug tracking system at http://bugs.perl.org/
e50bb9a1 670
722d2a37 671=head2 Integrate MacPerl
e50bb9a1 672
722d2a37
SC
673Chris Nandor and Matthias Neeracher have integrated the MacPerl changes
674into 5.6.0.
e50bb9a1 675
722d2a37 676=head2 Web "nerve center" for Perl
e50bb9a1 677
722d2a37 678http://use.perl.org/ is what you're looking for.
e50bb9a1 679
722d2a37 680=head2 Regular expression tutorial
e50bb9a1 681
722d2a37 682C<perlretut>, provided by Mark Kvale.
e50bb9a1 683
722d2a37 684=head2 Debugging Tutorial
e50bb9a1 685
722d2a37 686C<perldebtut>, written by Richard Foley.
e50bb9a1 687
722d2a37 688=head2 Integrate new modules
e50bb9a1 689
722d2a37 690Jarkko has been integrating madly into 5.7.x
e50bb9a1 691
722d2a37 692=head2 Integrate profiler
e50bb9a1 693
722d2a37 694C<Devel::DProf> is now a core module.
e50bb9a1 695
722d2a37 696=head2 Y2K error detection
e50bb9a1 697
722d2a37
SC
698There's a configure option to detect unsafe concatenation with "19", and
699a CPAN module. (C<D'oh::Year>)
e50bb9a1 700
722d2a37 701=head2 Regular expression debugger
e50bb9a1 702
722d2a37
SC
703While not part of core, Mark-Jason Dominus has written C<Rx> and has
704also come up with a generalised strategy for regular expression
705debugging.
e50bb9a1 706
722d2a37 707=head2 POD checker
e50bb9a1 708
722d2a37 709That's, uh, F<podchecker>
e50bb9a1 710
722d2a37 711=head2 "Dynamic" lexicals
e50bb9a1 712
722d2a37 713=head2 Cache precompiled modules
e50bb9a1 714
722d2a37 715=head1 Deprecated Wishes
e50bb9a1 716
722d2a37
SC
717These are items which used to be in the todo file, but have been
718deprecated for some reason.
e50bb9a1 719
722d2a37 720=head2 Loop control on do{}
e50bb9a1 721
722d2a37 722This would break old code; use C<do{{ }}> instead.
e50bb9a1 723
722d2a37 724=head2 Lexically scoped typeglobs
e50bb9a1 725
722d2a37 726Not needed now we have lexical IO handles.
e50bb9a1 727
722d2a37 728=head2 format BOTTOM
3958b146 729
722d2a37 730=head2 report HANDLE
e50bb9a1 731
722d2a37 732Damian Conway's text formatting modules seem to be the Way To Go.
e50bb9a1 733
722d2a37 734=head2 Generalised want()/caller())
3958b146 735
722d2a37 736=head2 Named prototypes
e50bb9a1 737
722d2a37 738These both seem to be delayed until Perl 6.
e50bb9a1 739
722d2a37 740=head2 Built-in globbing
e50bb9a1 741
722d2a37 742The C<File::Glob> module has been used to replace the C<glob> function.
e50bb9a1 743
722d2a37 744=head2 Regression tests for suidperl
e50bb9a1 745
722d2a37 746C<suidperl> is deprecated in favour of common sense.
e50bb9a1 747
722d2a37 748=head2 Cached hash values
e50bb9a1 749
722d2a37 750We have shared hash keys, which perform the same job.
e50bb9a1 751
722d2a37 752=head2 Add compression modules
e50bb9a1 753
722d2a37
SC
754The compression modules are a little heavy; meanwhile, Nick Clark is
755working on experimental pragmata to do transparent decompression on
756input.
e50bb9a1 757
722d2a37 758=head2 Reorganise documentation into tutorials/references
e50bb9a1 759
722d2a37 760Could not get consensus on P5P about this.
e50bb9a1 761
722d2a37
SC
762=head2 Remove distinction between functions and operators
763
764Caution: highly flammable.
765
766=head2 Make XS easier to use
e50bb9a1 767
722d2a37 768Use C<Inline> instead, or SWIG.
e50bb9a1 769
722d2a37 770=head2 Make embedding easier to use
e50bb9a1 771
722d2a37 772Use C<Inline::CPR>.
e50bb9a1 773
722d2a37 774=head2 man for perl
04c70446 775
722d2a37 776See the Perl Power Tools. (http://language.perl.com/ppt/)
04c70446 777
722d2a37 778=head2 my $Package::variable
04c70446 779
722d2a37 780Use C<our> instead.
04c70446 781
722d2a37 782=head2 "or" tests defined, not truth
04c70446 783
722d2a37 784Suggesting this on P5P B<will> cause a boring and interminable flamewar.
04c70446 785
722d2a37 786=head2 "class"-based lexicals
04c70446 787
cbb3fa72 788Use flyweight objects, secure hashes or, dare I say it, pseudo-hashes instead.
f86a8bc5 789(Or whatever will replace pseudohashes in 5.10.)
04c70446 790
722d2a37 791=head2 byteperl
04c70446 792
722d2a37 793C<ByteLoader> covers this.
04c70446 794
722d2a37 795=head2 Lazy evaluation / tail recursion removal
04c70446 796
f86a8bc5
JH
797C<List::Util> gives first() (a short-circuiting grep); tail recursion
798removal is done manually, with C<goto &whoami;>. (However, MJD has
799found that C<goto &whoami> introduces a performance penalty, so maybe
800there should be a way to do this after all: C<sub foo {START: ... goto
801START;> is better.)
0562c0e3
JH
802
803=head2 Make "use utf8" the default
804
f86a8bc5
JH
805Because of backward compatibility this is difficult: scripts could not
806contain B<any legacy eight-bit data> (like Latin-1) anymore, even in
807string literals or pod. Also would introduce a measurable slowdown of
808at least few percentages since all regular expression operations would
809be done in full UTF-8. But if you want to try this, add
810-DUSE_UTF8_SCRIPTS to your compilation flags.
811
0562c0e3 812