This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Explain a bit the new more flexible \p\P syntax.
[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
776f8809
JH
64=head2 Unicode
65
66=over 4
67
68=item *
e50bb9a1 69
722d2a37 70 Case Mappings? http://www.unicode.org/unicode/reports/tr21/
e50bb9a1 71
6f16a292
JH
72lc(), uc(), lcfirst(), and ucfirst() work only for some of the
73simplest cases, where the mapping goes from a single Unicode character
74to another single Unicode character. See lib/unicore/SpecCase.txt
75(and CaseFold.txt).
ac1256e8 76
776f8809 77=item *
e50bb9a1 78
c6287c21
JH
79They have some tricks Perl doesn't yet implement like character
80class subtraction.
e50bb9a1 81
722d2a37 82 http://www.unicode.org/unicode/reports/tr18/
e50bb9a1 83
776f8809
JH
84=back
85
86See L<perlunicode/UNICODE REGULAR EXPRESSION SUPPORT LEVEL> for what's
87there and what's missing.
88
722d2a37 89=head2 use Thread for iThreads
e50bb9a1 90
722d2a37
SC
91Artur Bergman's C<iThreads> module is a start on this, but needs to
92be more mature.
e50bb9a1 93
dd0afe54
AB
94=head2 make perl_clone optionally clone ops
95
96So that pseudoforking, mod_perl, iThreads and nvi will work properly
97(but not as efficiently) until the regex engine is fixed to be threadsafe.
98
722d2a37 99=head2 Work out exit/die semantics for threads
e50bb9a1 100
722d2a37 101=head2 Typed lexicals for compiler
e50bb9a1 102
722d2a37 103=head2 Compiler workarounds for Win32
e50bb9a1 104
722d2a37 105=head2 AUTOLOADing in the compiler
e50bb9a1 106
722d2a37 107=head2 Fixing comppadlist when compiling
e50bb9a1 108
722d2a37 109=head2 Cleaning up exported namespace
e50bb9a1 110
722d2a37 111=head2 Complete signal handling
e50bb9a1 112
722d2a37
SC
113Add C<PERL_ASYNC_CHECK> to opcodes which loop; replace C<sigsetjmp> with
114C<sigjmp>; check C<wait> for signal safety.
e50bb9a1 115
722d2a37 116=head2 Out-of-source builds
e50bb9a1 117
722d2a37 118This was done for 5.6.0, but needs reworking for 5.7.x
e50bb9a1 119
722d2a37 120=head2 POSIX realtime support
e50bb9a1 121
722d2a37
SC
122POSIX 1003.1 1996 Edition support--realtime stuff: POSIX semaphores,
123message queues, shared memory, realtime clocks, timers, signals (the
124metaconfig units mostly already exist for these)
e50bb9a1 125
722d2a37 126=head2 UNIX98 support
e50bb9a1 127
722d2a37 128Reader-writer locks, realtime/asynchronous IO
e50bb9a1 129
722d2a37 130=head2 IPv6 Support
e50bb9a1 131
722d2a37
SC
132There are non-core modules, such as C<Net::IPv6>, but these will need
133integrating when IPv6 actually starts to really happen. See RFC 2292
134and RFC 2553.
e50bb9a1 135
722d2a37 136=head2 Long double conversion
e50bb9a1 137
722d2a37 138Floating point formatting is still causing some weird test failures.
e50bb9a1 139
722d2a37 140=head2 Locales
e50bb9a1 141
722d2a37
SC
142Locales and Unicode interact with each other in unpleasant ways.
143One possible solution would be to adopt/support ICU:
e50bb9a1 144
722d2a37 145 http://oss.software.ibm.com/developerworks/opensource/icu/project/
e50bb9a1 146
722d2a37 147=head2 Thread-safe regexes
e50bb9a1 148
722d2a37 149The regular expression engine is currently non-threadsafe.
e50bb9a1 150
722d2a37 151=head2 Arithmetic on non-Arabic numerals
e50bb9a1 152
722d2a37 153C<[1234567890]> aren't the only numerals any more.
e50bb9a1 154
722d2a37 155=head2 POSIX Unicode character classes
e50bb9a1 156
722d2a37
SC
157([=a=] for equivalance classes, [.ch.] for collation.)
158These are dependent on Unicode normalization and collation.
e50bb9a1 159
722d2a37 160=head2 Factoring out common suffices/prefices in regexps (trie optimization)
c47ff5f1 161
722d2a37
SC
162Currently, the user has to optimize C<foo|far> and C<foo|goo> into
163C<f(?:oo|ar)> and C<[fg]oo> by hand; this could be done automatically.
e50bb9a1 164
722d2a37 165=head2 Security audit shipped utilities
e50bb9a1 166
722d2a37
SC
167All the code we ship with Perl needs to be sensible about temporary file
168handling, locking, input validation, and so on.
e50bb9a1 169
722d2a37 170=head2 Custom opcodes
e50bb9a1 171
722d2a37
SC
172Have a way to introduce user-defined opcodes without the subroutine call
173overhead of an XSUB; the user should be able to create PP code. Simon
174Cozens has some ideas on this.
e50bb9a1 175
722d2a37 176=head2 spawnvp() on Win32
e50bb9a1 177
722d2a37
SC
178Win32 has problems spawning processes, particularly when the arguments
179to the child process contain spaces, quotes or tab characters.
e50bb9a1 180
722d2a37 181=head2 DLL Versioning
e50bb9a1 182
722d2a37
SC
183Windows needs a way to know what version of a XS or C<libperl> DLL it's
184loading.
e50bb9a1 185
722d2a37 186=head2 Introduce @( and @)
e50bb9a1 187
722d2a37
SC
188C<$(> may return "foo bar baz". Unfortunately, since groups can
189theoretically have spaces in their names, this could be one, two or
190three groups.
e50bb9a1 191
722d2a37 192=head2 Floating point handling
e50bb9a1 193
722d2a37
SC
194C<NaN> and C<inf> support is particularly troublesome.
195(fp_classify(), fp_class(), fp_class_d(), class(), isinf(),
196isfinite(), finite(), isnormal(), unordered(), <ieeefp.h>,
197<fp_class.h> (there are metaconfig units for all these) (I think),
198fp_setmask(), fp_getmask(), fp_setround(), fp_getround()
199(no metaconfig units yet for these). Don't forget finitel(), fp_classl(),
200fp_class_l(), (yes, both do, unfortunately, exist), and unorderedl().)
e50bb9a1 201
722d2a37 202As of Perl 5.6.1 is a Perl macro, Perl_isnan().
e50bb9a1 203
722d2a37 204=head2 IV/UV preservation
e50bb9a1 205
722d2a37
SC
206Nicholas Clark has done a lot of work on this, but work is continuing.
207C<+>, C<-> and C<*> work, but guards need to be in place for C<%>, C</>,
208C<&>, C<oct>, C<hex> and C<pack>.
e50bb9a1 209
722d2a37 210=head2 Replace pod2html with something using Pod::Parser
83df6a1d 211
722d2a37
SC
212The CPAN module C<Malik::Pod::Html> may be a more suitable basis for a
213C<pod2html> convertor; the current one duplicates the functionality
214abstracted in C<Pod::Parser>, which makes updating the POD language
215difficult.
e50bb9a1 216
722d2a37 217=head2 Automate module testing on CPAN
e50bb9a1 218
722d2a37
SC
219When a new Perl is being beta tested, porters have to manually grab
220their favourite CPAN modules and test them - this should be done
221automatically.
e50bb9a1 222
722d2a37 223=head2 sendmsg and recvmsg
83df6a1d 224
722d2a37
SC
225We have all the other BSD socket functions but these. There are
226metaconfig units for these functions which can be added. To avoid these
227being new opcodes, a solution similar to the way C<sockatmark> was added
228would be preferable. (Autoload the C<IO::whatever> module.)
e50bb9a1 229
722d2a37 230=head2 Rewrite perlre documentation
e50bb9a1 231
722d2a37
SC
232The new-style patterns need full documentation, and the whole document
233needs to be a lot clearer.
e50bb9a1 234
722d2a37 235=head2 Convert example code to IO::Handle filehandles
e50bb9a1 236
722d2a37 237=head2 Document Win32 choices
e50bb9a1 238
722d2a37 239=head2 Check new modules
e50bb9a1 240
722d2a37 241=head2 Make roffitall find pods and libs itself
e50bb9a1 242
722d2a37 243Simon Cozens has done some work on this but it needs a rethink.
e50bb9a1 244
722d2a37 245=head1 To do at some point
e50bb9a1 246
722d2a37
SC
247These are ideas that have been regularly tossed around, that most
248people believe should be done maybe during 5.8.x
e50bb9a1 249
722d2a37 250=head2 Remove regular expression recursion
e50bb9a1 251
722d2a37
SC
252Because the regular expression engine is recursive, badly designed
253expressions can lead to lots of recursion filling up the stack. Ilya
254claims that it is easy to convert the engine to being iterative, but
255this has still not yet been done. There may be a regular expression
256engine hit squad meeting at TPC5.
e50bb9a1 257
722d2a37 258=head2 Memory leaks after failed eval
e50bb9a1 259
722d2a37
SC
260Perl will leak memory if you C<eval "hlagh hlagh hlagh hlagh">. This is
261partially because it attempts to build up an op tree for that code and
262doesn't properly free it. The same goes for non-syntactically-correct
263regular expressions. Hugo looked into this, but decided it needed a
264mark-and-sweep GC implementation.
e50bb9a1 265
722d2a37
SC
266Alan notes that: The basic idea was to extend the parser token stack
267(C<YYSTYPE>) to include a type field so we knew what sort of thing each
268element of the stack was. The F<<perly.c> code would then have to be
269postprocessed to record the type of each entry on the stack as it was
270created, and the parser patched so that it could unroll the stack
271properly on error.
e50bb9a1 272
722d2a37
SC
273This is possible to do, but would be pretty messy to implement, as it
274would rely on even more sed hackery in F<perly.fixer>.
e50bb9a1 275
722d2a37 276=head2 pack "(stuff)*"
e50bb9a1 277
722d2a37 278That's to say, C<pack "(sI)40"> would be the same as C<pack "sI"x40>
e50bb9a1 279
722d2a37 280=head2 bitfields in pack
e50bb9a1 281
722d2a37 282=head2 Cross compilation
e50bb9a1 283
722d2a37
SC
284Make Perl buildable with a cross-compiler. This will play havoc with
285Configure, which needs to how how the target system will respond to
286its tests; maybe C<microperl> will be a good starting point here.
287(Indeed, Bart Schuller reports that he compiled up C<microperl> for
288the Agenda PDA and it works fine.) A really big spanner in the works
289is the bootstrapping build process of Perl: if the filesystem the
290target systems sees is not the same what the build host sees, various
291input, output, and (Perl) library files need to be copied back and forth.
e50bb9a1 292
f86a8bc5
JH
293As of 5.8.0 Configure mostly works for cross-compilation
294(used successfully for iPAQ Linux), miniperl gets built,
295but then building DynaLoader (and other extensions) fails
296since MakeMaker knows nothing of cross-compilation.
297(See INSTALL/Cross-compilation for the state of things.)
298
722d2a37 299=head2 Perl preprocessor / macros
e50bb9a1 300
722d2a37
SC
301Source filters help with this, but do not get us all the way. For
302instance, it should be possible to implement the C<??> operator somehow;
303source filters don't (quite) cut it.
e50bb9a1 304
722d2a37 305=head2 Perl lexer in Perl
a45bd81d 306
722d2a37 307Damian Conway is planning to work on this, but it hasn't happened yet.
e50bb9a1 308
722d2a37 309=head2 Using POSIX calls internally
e50bb9a1 310
722d2a37
SC
311When faced with a BSD vs. SySV -style interface to some library or
312system function, perl's roots show in that it typically prefers the BSD
313interface (but falls back to the SysV one). One example is getpgrp().
314Other examples include C<memcpy> vs. C<bcopy>. There are others, mostly in
315F<<pp_sys.c>.
e50bb9a1 316
722d2a37
SC
317Mostly, this item is a suggestion for which way to start a journey into
318an C<#ifdef> forest. It is not primarily a suggestion to eliminate any of
319the C<#ifdef> forests.
e50bb9a1 320
722d2a37
SC
321POSIX calls are perhaps more likely to be portable to unexpected
322architectures. They are also perhaps more likely to be actively
323maintained by a current vendor. They are also perhaps more likely to be
324available in thread-safe versions, if appropriate.
e50bb9a1 325
722d2a37 326=head2 -i rename file when changed
e50bb9a1 327
722d2a37
SC
328It's only necessary to rename a file when inplace editing when the file
329has changed. Detecting a change is perhaps the difficult bit.
e50bb9a1 330
722d2a37 331=head2 All ARGV input should act like E<lt>E<gt>
e50bb9a1 332
2d84a16a
DM
333eg C<read(ARGV, ...)> doesn't currently read across multiple files.
334
722d2a37 335=head2 Support for rerunning debugger
e50bb9a1 336
722d2a37 337There should be a way of restarting the debugger on demand.
e50bb9a1 338
c6287c21
JH
339=head2 Test Suite for the Debugger
340
341The debugger is a complex piece of software and fixing something
342here may inadvertently break something else over there. To tame
343this chaotic behaviour, a test suite is necessary.
344
722d2a37 345=head2 my sub foo { }
c47ff5f1 346
722d2a37
SC
347The basic principle is sound, but there are problems with the semantics
348of self-referential and mutually referential lexical subs: how to
349declare the subs?
c47ff5f1 350
722d2a37 351=head2 One-pass global destruction
c47ff5f1 352
722d2a37
SC
353Sweeping away all the allocated memory in one go is a laudable goal, but
354it's difficult and in most cases, it's easier to let the memory get
355freed by exiting.
e50bb9a1 356
722d2a37 357=head2 Rewrite regexp parser
e50bb9a1 358
722d2a37
SC
359There has been talk recently of rewriting the regular expression parser
360to produce an optree instead of a chain of opcodes; it's unclear whether
361or not this would be a win.
e50bb9a1 362
722d2a37 363=head2 Cache recently used regexps
e50bb9a1 364
722d2a37 365This is to speed up
e50bb9a1 366
722d2a37
SC
367 for my $re (@regexps) {
368 $matched++ if /$re/
369 }
e50bb9a1 370
722d2a37
SC
371C<qr//> already gives us a way of saving compiled regexps, but it should
372be done automatically.
e50bb9a1 373
722d2a37 374=head2 Re-entrant functions
e50bb9a1 375
722d2a37
SC
376Add configure probes for C<_r> forms of system calls and fit them to the
377core. Unfortunately, calling conventions for these functions and not
378standardised.
04c70446 379
722d2a37 380=head2 Cross-compilation support
04c70446 381
722d2a37
SC
382Bart Schuller reports that using C<microperl> and a cross-compiler, he
383got Perl working on the Agenda PDA. However, one cannot build a full
384Perl because Configure needs to get the results for the target platform,
385for the host.
e50bb9a1 386
722d2a37 387=head2 Bit-shifting bitvectors
e50bb9a1 388
722d2a37 389Given:
e50bb9a1 390
722d2a37 391 vec($v, 1000, 1) = 1;
e50bb9a1 392
722d2a37 393One should be able to do
e50bb9a1 394
722d2a37 395 $v <<= 1;
e50bb9a1 396
722d2a37 397and have the 999'th bit set.
e50bb9a1 398
722d2a37
SC
399Currently if you try with shift bitvectors you shift the NV/UV, instead
400of the bits in the PV. Not very logical.
e50bb9a1 401
722d2a37 402=head2 debugger pragma
e50bb9a1 403
722d2a37
SC
404The debugger is implemented in Perl in F<perl5db.pl>; turning it into a
405pragma should be easy, but making it work lexically might be more
406difficult. Fiddling with C<$^P> would be necessary.
e50bb9a1 407
722d2a37 408=head2 use less pragma
e50bb9a1 409
722d2a37
SC
410Identify areas where speed/memory tradeoffs can be made and have a hint
411to switch between them.
e50bb9a1 412
722d2a37 413=head2 switch structures
e50bb9a1 414
722d2a37
SC
415Although we have C<Switch.pm> in core, Larry points to the dormant
416C<nswitch> and C<cswitch> ops in F<pp.c>; using these opcodes would be
417much faster.
e50bb9a1 418
722d2a37 419=head2 Cache eval tree
e50bb9a1 420
722d2a37 421=head2 rcatmaybe
e50bb9a1 422
722d2a37 423=head2 Shrink opcode tables
e50bb9a1 424
722d2a37 425=head2 Optimize away @_
e50bb9a1 426
722d2a37 427Look at the "reification" code in C<av.c>
e50bb9a1 428
722d2a37 429=head2 Prototypes versus indirect objects
e50bb9a1 430
722d2a37 431Currently, indirect object syntax bypasses prototype checks.
e50bb9a1 432
722d2a37 433=head2 Install HMTL
e50bb9a1 434
722d2a37
SC
435HTML versions of the documentation need to be installed by default; a
436call to C<installhtml> from C<installperl> may be all that's necessary.
e50bb9a1 437
722d2a37 438=head2 Prototype method calls
e50bb9a1 439
722d2a37 440=head2 Return context prototype declarations
e50bb9a1 441
722d2a37 442=head2 magic_setisa
e50bb9a1 443
722d2a37 444=head2 Garbage collection
e50bb9a1 445
722d2a37
SC
446There have been persistent mumblings about putting a mark-and-sweep
447garbage detector into Perl; Alan Burlison has some ideas about this.
e50bb9a1 448
722d2a37 449=head2 IO tutorial
e50bb9a1 450
722d2a37 451Mark-Jason Dominus has the beginnings of one of these.
e50bb9a1 452
722d2a37 453=head2 pack/unpack tutorial
e50bb9a1 454
722d2a37 455Simon Cozens has the beginnings of one of these.
e50bb9a1 456
722d2a37 457=head2 Rewrite perldoc
e50bb9a1 458
722d2a37
SC
459There are a few suggestions for what to do with C<perldoc>: maybe a
460full-text search, an index function, locating pages on a particular
461high-level subject, and so on.
e50bb9a1 462
3958b146 463=head2 Install .3p manpages
e50bb9a1 464
3958b146 465This is a bone of contention; we can create C<.3p> manpages for each
722d2a37
SC
466built-in function, but should we install them by default? Tcl does this,
467and it clutters up C<apropos>.
e50bb9a1 468
722d2a37 469=head2 Unicode tutorial
e50bb9a1 470
722d2a37 471Simon Cozens promises to do this before he gets old.
e50bb9a1 472
722d2a37 473=head2 Update POSIX.pm for 1003.1-2
3958b146 474
722d2a37 475=head2 Retargetable installation
e50bb9a1 476
722d2a37 477Allow C<@INC> to be changed after Perl is built.
e50bb9a1 478
722d2a37 479=head2 POSIX emulation on non-POSIX systems
e50bb9a1 480
722d2a37
SC
481Make C<POSIX.pm> behave as POSIXly as possible everywhere, meaning we
482have to implement POSIX equivalents for some functions if necessary.
e50bb9a1 483
722d2a37 484=head2 Rename Win32 headers
e50bb9a1 485
722d2a37
SC
486=head2 Finish off lvalue functions
487
488They don't work in the debugger, and they don't work for list or hash
489slices.
e50bb9a1 490
722d2a37 491=head2 Update sprintf documentation
e50bb9a1 492
722d2a37 493Hugo van der Sanden plans to look at this.
e50bb9a1 494
722d2a37 495=head2 Use fchown/fchmod internally
e50bb9a1 496
722d2a37
SC
497This has been done in places, but needs a thorough code review.
498Also fchdir is available in some platforms.
e50bb9a1 499
722d2a37 500=head1 Vague ideas
e50bb9a1 501
722d2a37 502Ideas which have been discussed, and which may or may not happen.
e50bb9a1 503
722d2a37 504=head2 ref() in list context
e50bb9a1 505
722d2a37
SC
506It's unclear what this should do or how to do it without breaking old
507code.
e50bb9a1 508
f86a8bc5 509=head2 Make tr/// return histogram of characters in list context
e50bb9a1 510
722d2a37 511There is a patch for this, but it may require Unicodification.
e50bb9a1 512
722d2a37 513=head2 Compile to real threaded code
3958b146 514
722d2a37 515=head2 Structured types
3958b146 516
722d2a37 517=head2 Modifiable $1 et al.
e50bb9a1 518
722d2a37
SC
519 ($x = "elephant") =~ /e(ph)/;
520 $1 = "g"; # $x = "elegant"
e50bb9a1 521
722d2a37
SC
522What happens if there are multiple (nested?) brackets? What if the
523string changes between the match and the assignment?
e50bb9a1 524
722d2a37 525=head2 Procedural interfaces for IO::*, etc.
e50bb9a1 526
722d2a37
SC
527Some core modules have been accused of being overly-OO. Adding
528procedural interfaces could demystify them.
e50bb9a1 529
722d2a37 530=head2 RPC modules
e50bb9a1 531
722d2a37 532=head2 Attach/detach debugger from running program
e50bb9a1 533
722d2a37
SC
534With C<gdb>, you can attach the debugger to a running program if you
535pass the process ID. It would be good to do this with the Perl debugger
536on a running Perl program, although I'm not sure how it would be done.
e50bb9a1 537
722d2a37 538=head2 Alternative RE syntax module
e50bb9a1 539
722d2a37
SC
540 use Regex::Newbie;
541 $re = Regex::Newbie->new
542 ->start
543 ->match("foo")
544 ->repeat(Regex::Newbie->class("char"),3)
545 ->end;
546 /$re/;
e50bb9a1 547
722d2a37 548=head2 GUI::Native
e50bb9a1 549
722d2a37
SC
550A non-core module that would use "native" GUI to create graphical
551applications.
e50bb9a1 552
722d2a37 553=head2 foreach(reverse ...)
e50bb9a1 554
722d2a37 555Currently
e50bb9a1 556
722d2a37 557 foreach (reverse @_) { ... }
e50bb9a1 558
722d2a37
SC
559puts C<@_> on the stack, reverses it putting the reversed version on the
560stack, then iterates forwards. Instead, it could be special-cased to put
561C<@_> on the stack then iterate backwards.
e50bb9a1 562
722d2a37 563=head2 Constant function cache
e50bb9a1 564
722d2a37 565=head2 Approximate regular expression matching
e50bb9a1 566
722d2a37 567=head1 Ongoing
e50bb9a1 568
722d2a37 569These items B<always> need doing:
e50bb9a1 570
722d2a37 571=head2 Update guts documentation
e50bb9a1 572
722d2a37
SC
573Simon Cozens tries to do this when possible, and contributions to the
574C<perlapi> documentation is welcome.
e50bb9a1 575
722d2a37 576=head2 Add more tests
e50bb9a1 577
722d2a37
SC
578Michael Schwern will donate $500 to Yet Another Society when all core
579modules have tests.
e50bb9a1 580
722d2a37 581=head2 Update auxiliary tools
e50bb9a1 582
722d2a37 583The code we ship with Perl should look like good Perl 5.
e50bb9a1 584
722d2a37 585=head1 Recently done things
e50bb9a1 586
722d2a37
SC
587These are things which have been on the todo lists in previous releases
588but have recently been completed.
e50bb9a1 589
722d2a37 590=head2 Safe signal handling
e50bb9a1 591
722d2a37
SC
592A new signal model went into 5.7.1 without much fanfare. Operations and
593C<malloc>s are no longer interrupted by signals, which are handled
594between opcodes. This means that C<PERL_ASYNC_CHECK> now actually does
595something. However, there are still a few things that need to be done.
e50bb9a1 596
722d2a37 597=head2 Tie Modules
e50bb9a1 598
722d2a37
SC
599Modules which implement arrays in terms of strings, substrings or files
600can be found on the CPAN.
e50bb9a1 601
722d2a37 602=head2 gettimeofday
e50bb9a1 603
722d2a37 604C<Time::Hires> has been integrated into the core.
e50bb9a1 605
722d2a37 606=head2 setitimer and getimiter
e50bb9a1 607
722d2a37 608Adding C<Time::Hires> got us this too.
e50bb9a1 609
722d2a37
SC
610=head2 Testing __DIE__ hook
611
612Tests have been added.
613
614=head2 CPP equivalent in Perl
e50bb9a1 615
722d2a37
SC
616A C Yardley will probably have done this by the time you can read this.
617This allows for a generalization of the C constant detection used in
618building C<Errno.pm>.
e50bb9a1 619
722d2a37 620=head2 Explicit switch statements
e50bb9a1 621
722d2a37
SC
622C<Switch.pm> has been integrated into the core to give you all manner of
623C<switch...case> semantics.
e50bb9a1 624
722d2a37 625=head2 autocroak
e50bb9a1 626
722d2a37 627This is C<Fatal.pm>.
e50bb9a1 628
722d2a37 629=head2 UTF/EBCDIC
e50bb9a1 630
722d2a37 631Nick Ing-Simmons has made UTF-EBCDIC (UTR13) work with Perl.
e50bb9a1 632
722d2a37 633 EBCDIC? http://www.unicode.org/unicode/reports/tr16/
e50bb9a1 634
722d2a37 635=head2 UTF Regexes
e50bb9a1 636
722d2a37
SC
637Although there are probably some small bugs to be rooted out, Jarkko
638Hietaniemi has made regular expressions polymorphic between bytes and
639characters.
e50bb9a1 640
722d2a37 641=head2 perlcc to produce executable
e50bb9a1 642
722d2a37
SC
643C<perlcc> was recently rewritten, and can now produce standalone
644executables.
e50bb9a1 645
722d2a37 646=head2 END blocks saved in compiled output
e50bb9a1 647
722d2a37 648=head2 Secure temporary file module
e50bb9a1 649
722d2a37 650Tim Jenness' C<File::Temp> is now in core.
e50bb9a1 651
722d2a37 652=head2 Integrate Time::HiRes
e50bb9a1 653
722d2a37 654This module is now part of core.
e50bb9a1 655
722d2a37 656=head2 Turn Cwd into XS
e50bb9a1 657
722d2a37 658Benjamin Sugars has done this.
e50bb9a1 659
722d2a37 660=head2 Mmap for input
e50bb9a1 661
722d2a37 662Nick Ing-Simmons' C<perlio> supports an C<mmap> IO method.
e50bb9a1 663
722d2a37 664=head2 Byte to/from UTF8 and UTF8 to/from local conversion
e50bb9a1 665
722d2a37 666C<Encode> provides this.
e50bb9a1 667
722d2a37 668=head2 Add sockatmark support
e50bb9a1 669
722d2a37 670Added in 5.7.1
e50bb9a1 671
722d2a37
SC
672=head2 Mailing list archives
673
674http://lists.perl.org/, http://archive.develooper.com/
675
676=head2 Bug tracking
677
678Richard Foley has written the bug tracking system at http://bugs.perl.org/
e50bb9a1 679
722d2a37 680=head2 Integrate MacPerl
e50bb9a1 681
722d2a37
SC
682Chris Nandor and Matthias Neeracher have integrated the MacPerl changes
683into 5.6.0.
e50bb9a1 684
722d2a37 685=head2 Web "nerve center" for Perl
e50bb9a1 686
722d2a37 687http://use.perl.org/ is what you're looking for.
e50bb9a1 688
722d2a37 689=head2 Regular expression tutorial
e50bb9a1 690
722d2a37 691C<perlretut>, provided by Mark Kvale.
e50bb9a1 692
722d2a37 693=head2 Debugging Tutorial
e50bb9a1 694
722d2a37 695C<perldebtut>, written by Richard Foley.
e50bb9a1 696
722d2a37 697=head2 Integrate new modules
e50bb9a1 698
722d2a37 699Jarkko has been integrating madly into 5.7.x
e50bb9a1 700
722d2a37 701=head2 Integrate profiler
e50bb9a1 702
722d2a37 703C<Devel::DProf> is now a core module.
e50bb9a1 704
722d2a37 705=head2 Y2K error detection
e50bb9a1 706
722d2a37
SC
707There's a configure option to detect unsafe concatenation with "19", and
708a CPAN module. (C<D'oh::Year>)
e50bb9a1 709
722d2a37 710=head2 Regular expression debugger
e50bb9a1 711
722d2a37
SC
712While not part of core, Mark-Jason Dominus has written C<Rx> and has
713also come up with a generalised strategy for regular expression
714debugging.
e50bb9a1 715
722d2a37 716=head2 POD checker
e50bb9a1 717
722d2a37 718That's, uh, F<podchecker>
e50bb9a1 719
722d2a37 720=head2 "Dynamic" lexicals
e50bb9a1 721
722d2a37 722=head2 Cache precompiled modules
e50bb9a1 723
722d2a37 724=head1 Deprecated Wishes
e50bb9a1 725
722d2a37
SC
726These are items which used to be in the todo file, but have been
727deprecated for some reason.
e50bb9a1 728
722d2a37 729=head2 Loop control on do{}
e50bb9a1 730
722d2a37 731This would break old code; use C<do{{ }}> instead.
e50bb9a1 732
722d2a37 733=head2 Lexically scoped typeglobs
e50bb9a1 734
722d2a37 735Not needed now we have lexical IO handles.
e50bb9a1 736
722d2a37 737=head2 format BOTTOM
3958b146 738
722d2a37 739=head2 report HANDLE
e50bb9a1 740
722d2a37 741Damian Conway's text formatting modules seem to be the Way To Go.
e50bb9a1 742
722d2a37 743=head2 Generalised want()/caller())
3958b146 744
722d2a37 745=head2 Named prototypes
e50bb9a1 746
722d2a37 747These both seem to be delayed until Perl 6.
e50bb9a1 748
722d2a37 749=head2 Built-in globbing
e50bb9a1 750
722d2a37 751The C<File::Glob> module has been used to replace the C<glob> function.
e50bb9a1 752
722d2a37 753=head2 Regression tests for suidperl
e50bb9a1 754
722d2a37 755C<suidperl> is deprecated in favour of common sense.
e50bb9a1 756
722d2a37 757=head2 Cached hash values
e50bb9a1 758
722d2a37 759We have shared hash keys, which perform the same job.
e50bb9a1 760
722d2a37 761=head2 Add compression modules
e50bb9a1 762
722d2a37
SC
763The compression modules are a little heavy; meanwhile, Nick Clark is
764working on experimental pragmata to do transparent decompression on
765input.
e50bb9a1 766
722d2a37 767=head2 Reorganise documentation into tutorials/references
e50bb9a1 768
722d2a37 769Could not get consensus on P5P about this.
e50bb9a1 770
722d2a37
SC
771=head2 Remove distinction between functions and operators
772
773Caution: highly flammable.
774
775=head2 Make XS easier to use
e50bb9a1 776
722d2a37 777Use C<Inline> instead, or SWIG.
e50bb9a1 778
722d2a37 779=head2 Make embedding easier to use
e50bb9a1 780
722d2a37 781Use C<Inline::CPR>.
e50bb9a1 782
722d2a37 783=head2 man for perl
04c70446 784
722d2a37 785See the Perl Power Tools. (http://language.perl.com/ppt/)
04c70446 786
722d2a37 787=head2 my $Package::variable
04c70446 788
722d2a37 789Use C<our> instead.
04c70446 790
722d2a37 791=head2 "or" tests defined, not truth
04c70446 792
722d2a37 793Suggesting this on P5P B<will> cause a boring and interminable flamewar.
04c70446 794
722d2a37 795=head2 "class"-based lexicals
04c70446 796
cbb3fa72 797Use flyweight objects, secure hashes or, dare I say it, pseudo-hashes instead.
f86a8bc5 798(Or whatever will replace pseudohashes in 5.10.)
04c70446 799
722d2a37 800=head2 byteperl
04c70446 801
722d2a37 802C<ByteLoader> covers this.
04c70446 803
722d2a37 804=head2 Lazy evaluation / tail recursion removal
04c70446 805
f86a8bc5
JH
806C<List::Util> gives first() (a short-circuiting grep); tail recursion
807removal is done manually, with C<goto &whoami;>. (However, MJD has
808found that C<goto &whoami> introduces a performance penalty, so maybe
809there should be a way to do this after all: C<sub foo {START: ... goto
810START;> is better.)
0562c0e3
JH
811
812=head2 Make "use utf8" the default
813
f86a8bc5
JH
814Because of backward compatibility this is difficult: scripts could not
815contain B<any legacy eight-bit data> (like Latin-1) anymore, even in
816string literals or pod. Also would introduce a measurable slowdown of
817at least few percentages since all regular expression operations would
818be done in full UTF-8. But if you want to try this, add
819-DUSE_UTF8_SCRIPTS to your compilation flags.
820
3298bd4d
JH
821=head2 Unicode collation and normalization
822
823The Unicode::Collate and Unicode::Normalize modules
824by SADAHIRO Tomoyuki have been included since 5.8.0.
825
826 Collation? http://www.unicode.org/unicode/reports/tr10/
827 Normalization? http://www.unicode.org/unicode/reports/tr15/
0562c0e3 828
825b3abc
RS
829=head2 Create debugging macros
830
831Debugging macros (like printsv, dump) can make debugging perl inside a
832C debugger much easier. A good set for gdb comes with mod_perl.
833Something similar should be distributed with perl.
834
835The proper way to do this is to use and extend Devel::DebugInit.
836Devel::DebugInit also needs to be extended to support threads.
837
838See p5p archives for late May/early June 2001 for a recent discussion
839on this topic.
840
3298bd4d 841=cut