This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Slight doc tweaks.
[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
1626a787
JH
57NOTE: pv_display(), pv_uni_display(), sv_uni_display() are already
58doing something like the above.
c5fc23ff 59
722d2a37 60=head2 Overloadable regex assertions
e50bb9a1 61
722d2a37
SC
62This may or may not be possible with the current regular expression
63engine. The idea is that, for instance, C<\b> needs to be
64algorithmically computed if you're dealing with Thai text. Hence, the
65B<\b> assertion wants to be overloaded by a function.
e50bb9a1 66
776f8809
JH
67=head2 Unicode
68
69=over 4
70
71=item *
e50bb9a1 72
f34dec15
JH
73Allow for long form of the General Category Properties, e.g
74C<\p{IsOpenPunctuation}>, not just the abbreviated form, e.g.
75C<\p{IsPs}>.
76
77=item *
78
1ac13f9a
JH
79Allow for the metaproperties: C<XID Start>, C<XID Continue>,
80C<NF*_NO>, C<NF*_MAYBE> (require the DerivedCoreProperties and
81DerviceNormalizationProperties files).
f34dec15 82
71d929cb
JH
83There are also multiple value properties still unimplemented:
84C<Numeric Type>, C<East Asian Width>.
f34dec15
JH
85
86=item *
87
722d2a37 88 Case Mappings? http://www.unicode.org/unicode/reports/tr21/
e50bb9a1 89
6f16a292
JH
90lc(), uc(), lcfirst(), and ucfirst() work only for some of the
91simplest cases, where the mapping goes from a single Unicode character
92to another single Unicode character. See lib/unicore/SpecCase.txt
93(and CaseFold.txt).
ac1256e8 94
776f8809 95=item *
e50bb9a1 96
8d3e8850 97UTF-8 identifier names should probably be canonicalized: NFC?
e50bb9a1 98
20eafb1c
JH
99=item *
100
101UTF-8 in package names and sub names? The first is problematic
8d3e8850 102because of the mapping to pathnames, ditto for the second one if
20eafb1c 103one does autosplitting, for example.
e50bb9a1 104
776f8809
JH
105=back
106
107See L<perlunicode/UNICODE REGULAR EXPRESSION SUPPORT LEVEL> for what's
f34dec15
JH
108there and what's missing. Almost all of Levels 2 and 3 is missing,
109and as of 5.8.0 not even all of Level 1 is there.
8d3e8850 110They have some tricks Perl doesn't yet implement, such as character
20eafb1c
JH
111class subtraction.
112
113 http://www.unicode.org/unicode/reports/tr18/
776f8809 114
56490ca2 115=head2 Work out exit/die semantics for threads
e50bb9a1 116
56490ca2
AB
117There are some suggestion for use a
118C<use threads wait => x> syntax.
e50bb9a1 119
56490ca2 120http://archive.develooper.com/perl5-porters@perl.org/msg79618.html
dd0afe54 121
56490ca2 122=head2 Better support for nonpreemptive threading systems like PTH
dd0afe54 123
56490ca2
AB
124Perhaps do yield() before blocking calls, exchange calls to non blocking
125calls specific for those libraries (for example GNUpth has a pth_sleep
126which is a non pth aware version of sleep).
e50bb9a1 127
722d2a37 128=head2 Typed lexicals for compiler
e50bb9a1 129
722d2a37 130=head2 Compiler workarounds for Win32
e50bb9a1 131
722d2a37 132=head2 AUTOLOADing in the compiler
e50bb9a1 133
722d2a37 134=head2 Fixing comppadlist when compiling
e50bb9a1 135
722d2a37 136=head2 Cleaning up exported namespace
e50bb9a1 137
722d2a37 138=head2 Complete signal handling
e50bb9a1 139
722d2a37
SC
140Add C<PERL_ASYNC_CHECK> to opcodes which loop; replace C<sigsetjmp> with
141C<sigjmp>; check C<wait> for signal safety.
e50bb9a1 142
722d2a37 143=head2 Out-of-source builds
e50bb9a1 144
722d2a37 145This was done for 5.6.0, but needs reworking for 5.7.x
e50bb9a1 146
722d2a37 147=head2 POSIX realtime support
e50bb9a1 148
722d2a37
SC
149POSIX 1003.1 1996 Edition support--realtime stuff: POSIX semaphores,
150message queues, shared memory, realtime clocks, timers, signals (the
151metaconfig units mostly already exist for these)
e50bb9a1 152
722d2a37 153=head2 UNIX98 support
e50bb9a1 154
722d2a37 155Reader-writer locks, realtime/asynchronous IO
e50bb9a1 156
722d2a37 157=head2 IPv6 Support
e50bb9a1 158
fe854a6f 159There are non-core modules, such as C<Socket6>, but these will need
722d2a37
SC
160integrating when IPv6 actually starts to really happen. See RFC 2292
161and RFC 2553.
e50bb9a1 162
722d2a37 163=head2 Long double conversion
e50bb9a1 164
722d2a37 165Floating point formatting is still causing some weird test failures.
e50bb9a1 166
722d2a37 167=head2 Locales
e50bb9a1 168
722d2a37
SC
169Locales and Unicode interact with each other in unpleasant ways.
170One possible solution would be to adopt/support ICU:
e50bb9a1 171
722d2a37 172 http://oss.software.ibm.com/developerworks/opensource/icu/project/
e50bb9a1 173
722d2a37 174=head2 Arithmetic on non-Arabic numerals
e50bb9a1 175
722d2a37 176C<[1234567890]> aren't the only numerals any more.
e50bb9a1 177
722d2a37 178=head2 POSIX Unicode character classes
e50bb9a1 179
210b36aa 180(C<[=a=]> for equivalance classes, C<[.ch.]> for collation.)
722d2a37 181These are dependent on Unicode normalization and collation.
e50bb9a1 182
722d2a37 183=head2 Factoring out common suffices/prefices in regexps (trie optimization)
c47ff5f1 184
722d2a37
SC
185Currently, the user has to optimize C<foo|far> and C<foo|goo> into
186C<f(?:oo|ar)> and C<[fg]oo> by hand; this could be done automatically.
e50bb9a1 187
722d2a37 188=head2 Security audit shipped utilities
e50bb9a1 189
722d2a37
SC
190All the code we ship with Perl needs to be sensible about temporary file
191handling, locking, input validation, and so on.
e50bb9a1 192
c8d2171d
JH
193=head2 Sort out the uid-setting mess
194
195Currently there are several problems with the setting of uids ($<, $>
196for the real and effective uids). Firstly, what exactly setuid() call
197gets invoked in which platform is simply a big mess that needs to be
198untangled. Secondly, the effects are apparently not standard across
199platforms, (if you first set $< and then $>, or vice versa, being
666f95b9 200uid == euid == zero, or just euid == zero, or as a normal user, what are
c8d2171d
JH
201the results?). The test suite not (usually) being run as root means
202that these things do not get much testing. Thirdly, there's quite
203often a third uid called saved uid, and Perl has no knowledge of that
204feature in any way. (If one has the saved uid of zero, one can get
205back any real and effective uids.) As an example, to change also the
206saved uid, one needs to set the real and effective uids B<twice>-- in
207most systems, that is: in HP-UX that doesn't seem to work.
666f95b9 208
722d2a37 209=head2 Custom opcodes
e50bb9a1 210
722d2a37
SC
211Have a way to introduce user-defined opcodes without the subroutine call
212overhead of an XSUB; the user should be able to create PP code. Simon
213Cozens has some ideas on this.
e50bb9a1 214
722d2a37 215=head2 DLL Versioning
e50bb9a1 216
d1be9408 217Windows needs a way to know what version of an XS or C<libperl> DLL it's
722d2a37 218loading.
e50bb9a1 219
722d2a37 220=head2 Introduce @( and @)
e50bb9a1 221
722d2a37
SC
222C<$(> may return "foo bar baz". Unfortunately, since groups can
223theoretically have spaces in their names, this could be one, two or
224three groups.
e50bb9a1 225
722d2a37 226=head2 Floating point handling
e50bb9a1 227
722d2a37
SC
228C<NaN> and C<inf> support is particularly troublesome.
229(fp_classify(), fp_class(), fp_class_d(), class(), isinf(),
230isfinite(), finite(), isnormal(), unordered(), <ieeefp.h>,
231<fp_class.h> (there are metaconfig units for all these) (I think),
232fp_setmask(), fp_getmask(), fp_setround(), fp_getround()
233(no metaconfig units yet for these). Don't forget finitel(), fp_classl(),
234fp_class_l(), (yes, both do, unfortunately, exist), and unorderedl().)
e50bb9a1 235
210b36aa 236As of Perl 5.6.1, there is a Perl macro, Perl_isnan().
e50bb9a1 237
722d2a37 238=head2 IV/UV preservation
e50bb9a1 239
722d2a37
SC
240Nicholas Clark has done a lot of work on this, but work is continuing.
241C<+>, C<-> and C<*> work, but guards need to be in place for C<%>, C</>,
242C<&>, C<oct>, C<hex> and C<pack>.
e50bb9a1 243
722d2a37 244=head2 Replace pod2html with something using Pod::Parser
83df6a1d 245
fe854a6f 246The CPAN module C<Marek::Pod::Html> may be a more suitable basis for a
722d2a37
SC
247C<pod2html> convertor; the current one duplicates the functionality
248abstracted in C<Pod::Parser>, which makes updating the POD language
249difficult.
e50bb9a1 250
722d2a37 251=head2 Automate module testing on CPAN
e50bb9a1 252
722d2a37
SC
253When a new Perl is being beta tested, porters have to manually grab
254their favourite CPAN modules and test them - this should be done
255automatically.
e50bb9a1 256
722d2a37 257=head2 sendmsg and recvmsg
83df6a1d 258
722d2a37
SC
259We have all the other BSD socket functions but these. There are
260metaconfig units for these functions which can be added. To avoid these
261being new opcodes, a solution similar to the way C<sockatmark> was added
262would be preferable. (Autoload the C<IO::whatever> module.)
e50bb9a1 263
722d2a37 264=head2 Rewrite perlre documentation
e50bb9a1 265
722d2a37
SC
266The new-style patterns need full documentation, and the whole document
267needs to be a lot clearer.
e50bb9a1 268
722d2a37 269=head2 Convert example code to IO::Handle filehandles
e50bb9a1 270
722d2a37 271=head2 Document Win32 choices
e50bb9a1 272
722d2a37 273=head2 Check new modules
e50bb9a1 274
722d2a37 275=head2 Make roffitall find pods and libs itself
e50bb9a1 276
722d2a37 277Simon Cozens has done some work on this but it needs a rethink.
e50bb9a1 278
722d2a37 279=head1 To do at some point
e50bb9a1 280
722d2a37
SC
281These are ideas that have been regularly tossed around, that most
282people believe should be done maybe during 5.8.x
e50bb9a1 283
722d2a37 284=head2 Remove regular expression recursion
e50bb9a1 285
722d2a37
SC
286Because the regular expression engine is recursive, badly designed
287expressions can lead to lots of recursion filling up the stack. Ilya
288claims that it is easy to convert the engine to being iterative, but
289this has still not yet been done. There may be a regular expression
290engine hit squad meeting at TPC5.
e50bb9a1 291
722d2a37 292=head2 Memory leaks after failed eval
e50bb9a1 293
722d2a37
SC
294Perl will leak memory if you C<eval "hlagh hlagh hlagh hlagh">. This is
295partially because it attempts to build up an op tree for that code and
296doesn't properly free it. The same goes for non-syntactically-correct
297regular expressions. Hugo looked into this, but decided it needed a
298mark-and-sweep GC implementation.
e50bb9a1 299
722d2a37
SC
300Alan notes that: The basic idea was to extend the parser token stack
301(C<YYSTYPE>) to include a type field so we knew what sort of thing each
210b36aa 302element of the stack was. The F<perly.c> code would then have to be
722d2a37
SC
303postprocessed to record the type of each entry on the stack as it was
304created, and the parser patched so that it could unroll the stack
305properly on error.
e50bb9a1 306
722d2a37
SC
307This is possible to do, but would be pretty messy to implement, as it
308would rely on even more sed hackery in F<perly.fixer>.
e50bb9a1 309
722d2a37 310=head2 bitfields in pack
e50bb9a1 311
722d2a37 312=head2 Cross compilation
e50bb9a1 313
722d2a37 314Make Perl buildable with a cross-compiler. This will play havoc with
da75cd15 315Configure, which needs to know how the target system will respond to
722d2a37
SC
316its tests; maybe C<microperl> will be a good starting point here.
317(Indeed, Bart Schuller reports that he compiled up C<microperl> for
318the Agenda PDA and it works fine.) A really big spanner in the works
319is the bootstrapping build process of Perl: if the filesystem the
320target systems sees is not the same what the build host sees, various
321input, output, and (Perl) library files need to be copied back and forth.
e50bb9a1 322
f86a8bc5
JH
323As of 5.8.0 Configure mostly works for cross-compilation
324(used successfully for iPAQ Linux), miniperl gets built,
325but then building DynaLoader (and other extensions) fails
326since MakeMaker knows nothing of cross-compilation.
327(See INSTALL/Cross-compilation for the state of things.)
328
722d2a37 329=head2 Perl preprocessor / macros
e50bb9a1 330
722d2a37
SC
331Source filters help with this, but do not get us all the way. For
332instance, it should be possible to implement the C<??> operator somehow;
333source filters don't (quite) cut it.
e50bb9a1 334
722d2a37 335=head2 Perl lexer in Perl
a45bd81d 336
722d2a37 337Damian Conway is planning to work on this, but it hasn't happened yet.
e50bb9a1 338
722d2a37 339=head2 Using POSIX calls internally
e50bb9a1 340
210b36aa 341When faced with a BSD vs. SysV -style interface to some library or
722d2a37
SC
342system function, perl's roots show in that it typically prefers the BSD
343interface (but falls back to the SysV one). One example is getpgrp().
344Other examples include C<memcpy> vs. C<bcopy>. There are others, mostly in
210b36aa 345F<pp_sys.c>.
e50bb9a1 346
722d2a37
SC
347Mostly, this item is a suggestion for which way to start a journey into
348an C<#ifdef> forest. It is not primarily a suggestion to eliminate any of
349the C<#ifdef> forests.
e50bb9a1 350
722d2a37
SC
351POSIX calls are perhaps more likely to be portable to unexpected
352architectures. They are also perhaps more likely to be actively
353maintained by a current vendor. They are also perhaps more likely to be
354available in thread-safe versions, if appropriate.
e50bb9a1 355
722d2a37 356=head2 -i rename file when changed
e50bb9a1 357
722d2a37
SC
358It's only necessary to rename a file when inplace editing when the file
359has changed. Detecting a change is perhaps the difficult bit.
e50bb9a1 360
722d2a37 361=head2 All ARGV input should act like E<lt>E<gt>
e50bb9a1 362
2d84a16a
DM
363eg C<read(ARGV, ...)> doesn't currently read across multiple files.
364
722d2a37 365=head2 Support for rerunning debugger
e50bb9a1 366
722d2a37 367There should be a way of restarting the debugger on demand.
e50bb9a1 368
c6287c21
JH
369=head2 Test Suite for the Debugger
370
371The debugger is a complex piece of software and fixing something
372here may inadvertently break something else over there. To tame
373this chaotic behaviour, a test suite is necessary.
374
722d2a37 375=head2 my sub foo { }
c47ff5f1 376
722d2a37
SC
377The basic principle is sound, but there are problems with the semantics
378of self-referential and mutually referential lexical subs: how to
379declare the subs?
c47ff5f1 380
722d2a37 381=head2 One-pass global destruction
c47ff5f1 382
722d2a37
SC
383Sweeping away all the allocated memory in one go is a laudable goal, but
384it's difficult and in most cases, it's easier to let the memory get
385freed by exiting.
e50bb9a1 386
722d2a37 387=head2 Rewrite regexp parser
e50bb9a1 388
722d2a37
SC
389There has been talk recently of rewriting the regular expression parser
390to produce an optree instead of a chain of opcodes; it's unclear whether
391or not this would be a win.
e50bb9a1 392
722d2a37 393=head2 Cache recently used regexps
e50bb9a1 394
722d2a37 395This is to speed up
e50bb9a1 396
722d2a37
SC
397 for my $re (@regexps) {
398 $matched++ if /$re/
399 }
e50bb9a1 400
722d2a37
SC
401C<qr//> already gives us a way of saving compiled regexps, but it should
402be done automatically.
e50bb9a1 403
722d2a37 404=head2 Re-entrant functions
e50bb9a1 405
722d2a37
SC
406Add configure probes for C<_r> forms of system calls and fit them to the
407core. Unfortunately, calling conventions for these functions and not
408standardised.
04c70446 409
722d2a37 410=head2 Cross-compilation support
04c70446 411
722d2a37
SC
412Bart Schuller reports that using C<microperl> and a cross-compiler, he
413got Perl working on the Agenda PDA. However, one cannot build a full
414Perl because Configure needs to get the results for the target platform,
415for the host.
e50bb9a1 416
722d2a37 417=head2 Bit-shifting bitvectors
e50bb9a1 418
722d2a37 419Given:
e50bb9a1 420
722d2a37 421 vec($v, 1000, 1) = 1;
e50bb9a1 422
722d2a37 423One should be able to do
e50bb9a1 424
722d2a37 425 $v <<= 1;
e50bb9a1 426
722d2a37 427and have the 999'th bit set.
e50bb9a1 428
722d2a37
SC
429Currently if you try with shift bitvectors you shift the NV/UV, instead
430of the bits in the PV. Not very logical.
e50bb9a1 431
722d2a37 432=head2 debugger pragma
e50bb9a1 433
722d2a37
SC
434The debugger is implemented in Perl in F<perl5db.pl>; turning it into a
435pragma should be easy, but making it work lexically might be more
436difficult. Fiddling with C<$^P> would be necessary.
e50bb9a1 437
722d2a37 438=head2 use less pragma
e50bb9a1 439
722d2a37
SC
440Identify areas where speed/memory tradeoffs can be made and have a hint
441to switch between them.
e50bb9a1 442
722d2a37 443=head2 switch structures
e50bb9a1 444
722d2a37
SC
445Although we have C<Switch.pm> in core, Larry points to the dormant
446C<nswitch> and C<cswitch> ops in F<pp.c>; using these opcodes would be
447much faster.
e50bb9a1 448
722d2a37 449=head2 Cache eval tree
e50bb9a1 450
722d2a37 451=head2 rcatmaybe
e50bb9a1 452
722d2a37 453=head2 Shrink opcode tables
e50bb9a1 454
722d2a37 455=head2 Optimize away @_
e50bb9a1 456
722d2a37 457Look at the "reification" code in C<av.c>
e50bb9a1 458
722d2a37 459=head2 Prototypes versus indirect objects
e50bb9a1 460
722d2a37 461Currently, indirect object syntax bypasses prototype checks.
e50bb9a1 462
210b36aa 463=head2 Install HTML
e50bb9a1 464
722d2a37
SC
465HTML versions of the documentation need to be installed by default; a
466call to C<installhtml> from C<installperl> may be all that's necessary.
e50bb9a1 467
722d2a37 468=head2 Prototype method calls
e50bb9a1 469
722d2a37 470=head2 Return context prototype declarations
e50bb9a1 471
722d2a37 472=head2 magic_setisa
e50bb9a1 473
722d2a37 474=head2 Garbage collection
e50bb9a1 475
722d2a37
SC
476There have been persistent mumblings about putting a mark-and-sweep
477garbage detector into Perl; Alan Burlison has some ideas about this.
e50bb9a1 478
722d2a37 479=head2 IO tutorial
e50bb9a1 480
722d2a37 481Mark-Jason Dominus has the beginnings of one of these.
e50bb9a1 482
722d2a37 483=head2 Rewrite perldoc
e50bb9a1 484
722d2a37
SC
485There are a few suggestions for what to do with C<perldoc>: maybe a
486full-text search, an index function, locating pages on a particular
487high-level subject, and so on.
e50bb9a1 488
3958b146 489=head2 Install .3p manpages
e50bb9a1 490
3958b146 491This is a bone of contention; we can create C<.3p> manpages for each
722d2a37
SC
492built-in function, but should we install them by default? Tcl does this,
493and it clutters up C<apropos>.
e50bb9a1 494
722d2a37 495=head2 Unicode tutorial
e50bb9a1 496
722d2a37 497Simon Cozens promises to do this before he gets old.
e50bb9a1 498
722d2a37 499=head2 Update POSIX.pm for 1003.1-2
3958b146 500
722d2a37 501=head2 Retargetable installation
e50bb9a1 502
722d2a37 503Allow C<@INC> to be changed after Perl is built.
e50bb9a1 504
722d2a37 505=head2 POSIX emulation on non-POSIX systems
e50bb9a1 506
722d2a37
SC
507Make C<POSIX.pm> behave as POSIXly as possible everywhere, meaning we
508have to implement POSIX equivalents for some functions if necessary.
e50bb9a1 509
722d2a37 510=head2 Rename Win32 headers
e50bb9a1 511
722d2a37
SC
512=head2 Finish off lvalue functions
513
514They don't work in the debugger, and they don't work for list or hash
515slices.
e50bb9a1 516
722d2a37 517=head2 Update sprintf documentation
e50bb9a1 518
722d2a37 519Hugo van der Sanden plans to look at this.
e50bb9a1 520
722d2a37 521=head2 Use fchown/fchmod internally
e50bb9a1 522
722d2a37
SC
523This has been done in places, but needs a thorough code review.
524Also fchdir is available in some platforms.
e50bb9a1 525
d45541b3 526=head2 Make v-strings overloaded objects
c5fc23ff 527
d45541b3
JH
528Instead of having to guess whether a string is a v-string and thus
529needs to be displayed with %vd, make v-strings (readonly) objects
530(class "vstring"?) with a stringify overload.
c5fc23ff 531
49293501
MS
532=head2 Allow restricted hash assignment
533
534Currently you're not allowed to assign to a restricted hash at all,
535even with the same keys.
536
537 %restricted = (foo => 42); # error
538
539This should be allowed if the new keyset is a subset of the old
540keyset. May require more extra code than we'd like in pp_aassign.
541
5387ccf1
JH
542=head2 Should overload be inheritable?
543
544Should overload be 'contagious' through @ISA so that derived classes
545would inherit their base classes' overload definitions? What to do
546in case of overload conflicts?
547
cbda53d5
JH
548=head2 Taint rethink
549
550Should taint be stopped from affecting control flow, if ($tainted)?
551Should tainted symbolic method calls and subref calls be stopped?
552(Look at Ruby's $SAFE levels for inspiration?)
553
722d2a37 554=head1 Vague ideas
e50bb9a1 555
722d2a37 556Ideas which have been discussed, and which may or may not happen.
e50bb9a1 557
722d2a37 558=head2 ref() in list context
e50bb9a1 559
722d2a37
SC
560It's unclear what this should do or how to do it without breaking old
561code.
e50bb9a1 562
f86a8bc5 563=head2 Make tr/// return histogram of characters in list context
e50bb9a1 564
722d2a37 565There is a patch for this, but it may require Unicodification.
e50bb9a1 566
722d2a37 567=head2 Compile to real threaded code
3958b146 568
722d2a37 569=head2 Structured types
3958b146 570
722d2a37 571=head2 Modifiable $1 et al.
e50bb9a1 572
722d2a37
SC
573 ($x = "elephant") =~ /e(ph)/;
574 $1 = "g"; # $x = "elegant"
e50bb9a1 575
722d2a37
SC
576What happens if there are multiple (nested?) brackets? What if the
577string changes between the match and the assignment?
e50bb9a1 578
722d2a37 579=head2 Procedural interfaces for IO::*, etc.
e50bb9a1 580
722d2a37
SC
581Some core modules have been accused of being overly-OO. Adding
582procedural interfaces could demystify them.
e50bb9a1 583
722d2a37 584=head2 RPC modules
e50bb9a1 585
722d2a37 586=head2 Attach/detach debugger from running program
e50bb9a1 587
722d2a37
SC
588With C<gdb>, you can attach the debugger to a running program if you
589pass the process ID. It would be good to do this with the Perl debugger
590on a running Perl program, although I'm not sure how it would be done.
e50bb9a1 591
722d2a37 592=head2 GUI::Native
e50bb9a1 593
722d2a37
SC
594A non-core module that would use "native" GUI to create graphical
595applications.
e50bb9a1 596
722d2a37 597=head2 foreach(reverse ...)
e50bb9a1 598
722d2a37 599Currently
e50bb9a1 600
722d2a37 601 foreach (reverse @_) { ... }
e50bb9a1 602
722d2a37
SC
603puts C<@_> on the stack, reverses it putting the reversed version on the
604stack, then iterates forwards. Instead, it could be special-cased to put
605C<@_> on the stack then iterate backwards.
e50bb9a1 606
722d2a37 607=head2 Constant function cache
e50bb9a1 608
722d2a37 609=head2 Approximate regular expression matching
e50bb9a1 610
722d2a37 611=head1 Ongoing
e50bb9a1 612
722d2a37 613These items B<always> need doing:
e50bb9a1 614
722d2a37 615=head2 Update guts documentation
e50bb9a1 616
722d2a37
SC
617Simon Cozens tries to do this when possible, and contributions to the
618C<perlapi> documentation is welcome.
e50bb9a1 619
722d2a37 620=head2 Add more tests
e50bb9a1 621
722d2a37
SC
622Michael Schwern will donate $500 to Yet Another Society when all core
623modules have tests.
e50bb9a1 624
722d2a37 625=head2 Update auxiliary tools
e50bb9a1 626
722d2a37 627The code we ship with Perl should look like good Perl 5.
e50bb9a1 628
1e278fd9
JH
629=head2 Create debugging macros
630
631Debugging macros (like printsv, dump) can make debugging perl inside a
632C debugger much easier. A good set for gdb comes with mod_perl.
633Something similar should be distributed with perl.
634
635The proper way to do this is to use and extend Devel::DebugInit.
636Devel::DebugInit also needs to be extended to support threads.
637
638See p5p archives for late May/early June 2001 for a recent discussion
639on this topic.
640
641=head2 truncate to the people
642
643One can emulate ftruncate() using F_FREESP and F_CHSIZ fcntls
644(see the UNIX FAQ for details). This needs to go somewhere near
645pp_sys.c:pp_truncate().
646
647One can emulate truncate() easily if one has ftruncate().
648This emulation should also go near pp_sys.pp_truncate().
649
650=head2 Unicode in Filenames
651
652chdir, chmod, chown, chroot, exec, glob, link, lstat, mkdir, open, qx,
653readdir, readlink, rename, rmdir, stat, symlink, sysopen, system,
654truncate, unlink, utime. All these could potentially accept Unicode
655filenames either as input or output (and in the case of system and qx
656Unicode in general, as input or output to/from the shell). Whether a
657filesystem - an operating system pair understands Unicode in filenames
658varies.
659
660Known combinations that have some level of understanding include
661Microsoft NTFS, Apple HFS+ (In Mac OS 9 and X) and Apple UFS (in Mac
662OS X), NFS v4 is rumored to be Unicode, and of course Plan 9. How to
663create Unicode filenames, what forms of Unicode are accepted and used
664(UCS-2, UTF-16, UTF-8), what (if any) is the normalization form used,
665and so on, varies. Finding the right level of interfacing to Perl
666requires some thought. Remember that an OS does not implicate a
667filesystem.
668
eb450546
JH
669Note that in Windows the -C command line flag already does quite
670a bit of the above (but even there the support is not complete:
671for example the exec/spawn are not Unicode-aware) by turning on
672the so-called "wide API support".
673
722d2a37 674=head1 Recently done things
e50bb9a1 675
722d2a37
SC
676These are things which have been on the todo lists in previous releases
677but have recently been completed.
e50bb9a1 678
b0b7f283 679=head2 Alternative RE syntax module
680
681The C<Regexp::English> module, available from the CPAN, provides this:
682
683 my $re = Regexp::English
684 -> start_of_line
685 -> literal('Flippers')
686 -> literal(':')
687 -> optional
688 -> whitespace_char
689 -> end
690 -> remember
691 -> multiple
692 -> digit;
693
694 /$re/;
695
722d2a37 696=head2 Safe signal handling
e50bb9a1 697
722d2a37
SC
698A new signal model went into 5.7.1 without much fanfare. Operations and
699C<malloc>s are no longer interrupted by signals, which are handled
700between opcodes. This means that C<PERL_ASYNC_CHECK> now actually does
701something. However, there are still a few things that need to be done.
e50bb9a1 702
722d2a37 703=head2 Tie Modules
e50bb9a1 704
722d2a37
SC
705Modules which implement arrays in terms of strings, substrings or files
706can be found on the CPAN.
e50bb9a1 707
722d2a37 708=head2 gettimeofday
e50bb9a1 709
210b36aa 710C<Time::HiRes> has been integrated into the core.
e50bb9a1 711
722d2a37 712=head2 setitimer and getimiter
e50bb9a1 713
210b36aa 714Adding C<Time::HiRes> got us this too.
e50bb9a1 715
722d2a37
SC
716=head2 Testing __DIE__ hook
717
718Tests have been added.
719
720=head2 CPP equivalent in Perl
e50bb9a1 721
722d2a37
SC
722A C Yardley will probably have done this by the time you can read this.
723This allows for a generalization of the C constant detection used in
724building C<Errno.pm>.
e50bb9a1 725
722d2a37 726=head2 Explicit switch statements
e50bb9a1 727
722d2a37
SC
728C<Switch.pm> has been integrated into the core to give you all manner of
729C<switch...case> semantics.
e50bb9a1 730
722d2a37 731=head2 autocroak
e50bb9a1 732
722d2a37 733This is C<Fatal.pm>.
e50bb9a1 734
722d2a37 735=head2 UTF/EBCDIC
e50bb9a1 736
722d2a37 737Nick Ing-Simmons has made UTF-EBCDIC (UTR13) work with Perl.
e50bb9a1 738
722d2a37 739 EBCDIC? http://www.unicode.org/unicode/reports/tr16/
e50bb9a1 740
722d2a37 741=head2 UTF Regexes
e50bb9a1 742
722d2a37
SC
743Although there are probably some small bugs to be rooted out, Jarkko
744Hietaniemi has made regular expressions polymorphic between bytes and
745characters.
e50bb9a1 746
722d2a37 747=head2 perlcc to produce executable
e50bb9a1 748
722d2a37
SC
749C<perlcc> was recently rewritten, and can now produce standalone
750executables.
e50bb9a1 751
722d2a37 752=head2 END blocks saved in compiled output
e50bb9a1 753
722d2a37 754=head2 Secure temporary file module
e50bb9a1 755
722d2a37 756Tim Jenness' C<File::Temp> is now in core.
e50bb9a1 757
722d2a37 758=head2 Integrate Time::HiRes
e50bb9a1 759
722d2a37 760This module is now part of core.
e50bb9a1 761
722d2a37 762=head2 Turn Cwd into XS
e50bb9a1 763
722d2a37 764Benjamin Sugars has done this.
e50bb9a1 765
722d2a37 766=head2 Mmap for input
e50bb9a1 767
722d2a37 768Nick Ing-Simmons' C<perlio> supports an C<mmap> IO method.
e50bb9a1 769
722d2a37 770=head2 Byte to/from UTF8 and UTF8 to/from local conversion
e50bb9a1 771
722d2a37 772C<Encode> provides this.
e50bb9a1 773
722d2a37 774=head2 Add sockatmark support
e50bb9a1 775
722d2a37 776Added in 5.7.1
e50bb9a1 777
722d2a37
SC
778=head2 Mailing list archives
779
f224927c 780http://lists.perl.org/ , http://archive.develooper.com/
722d2a37
SC
781
782=head2 Bug tracking
783
784Richard Foley has written the bug tracking system at http://bugs.perl.org/
e50bb9a1 785
722d2a37 786=head2 Integrate MacPerl
e50bb9a1 787
722d2a37
SC
788Chris Nandor and Matthias Neeracher have integrated the MacPerl changes
789into 5.6.0.
e50bb9a1 790
722d2a37 791=head2 Web "nerve center" for Perl
e50bb9a1 792
722d2a37 793http://use.perl.org/ is what you're looking for.
e50bb9a1 794
722d2a37 795=head2 Regular expression tutorial
e50bb9a1 796
722d2a37 797C<perlretut>, provided by Mark Kvale.
e50bb9a1 798
722d2a37 799=head2 Debugging Tutorial
e50bb9a1 800
722d2a37 801C<perldebtut>, written by Richard Foley.
e50bb9a1 802
722d2a37 803=head2 Integrate new modules
e50bb9a1 804
722d2a37 805Jarkko has been integrating madly into 5.7.x
e50bb9a1 806
722d2a37 807=head2 Integrate profiler
e50bb9a1 808
722d2a37 809C<Devel::DProf> is now a core module.
e50bb9a1 810
722d2a37 811=head2 Y2K error detection
e50bb9a1 812
722d2a37
SC
813There's a configure option to detect unsafe concatenation with "19", and
814a CPAN module. (C<D'oh::Year>)
e50bb9a1 815
722d2a37 816=head2 Regular expression debugger
e50bb9a1 817
722d2a37
SC
818While not part of core, Mark-Jason Dominus has written C<Rx> and has
819also come up with a generalised strategy for regular expression
820debugging.
e50bb9a1 821
722d2a37 822=head2 POD checker
e50bb9a1 823
722d2a37 824That's, uh, F<podchecker>
e50bb9a1 825
722d2a37 826=head2 "Dynamic" lexicals
e50bb9a1 827
722d2a37 828=head2 Cache precompiled modules
e50bb9a1 829
722d2a37 830=head1 Deprecated Wishes
e50bb9a1 831
722d2a37
SC
832These are items which used to be in the todo file, but have been
833deprecated for some reason.
e50bb9a1 834
722d2a37 835=head2 Loop control on do{}
e50bb9a1 836
722d2a37 837This would break old code; use C<do{{ }}> instead.
e50bb9a1 838
722d2a37 839=head2 Lexically scoped typeglobs
e50bb9a1 840
722d2a37 841Not needed now we have lexical IO handles.
e50bb9a1 842
722d2a37 843=head2 format BOTTOM
3958b146 844
722d2a37 845=head2 report HANDLE
e50bb9a1 846
722d2a37 847Damian Conway's text formatting modules seem to be the Way To Go.
e50bb9a1 848
722d2a37 849=head2 Generalised want()/caller())
3958b146 850
638ae6a9
MJD
851Robin Houston's C<Want> module does this.
852
722d2a37 853=head2 Named prototypes
e50bb9a1 854
638ae6a9 855This seems to be delayed until Perl 6.
e50bb9a1 856
722d2a37 857=head2 Built-in globbing
e50bb9a1 858
722d2a37 859The C<File::Glob> module has been used to replace the C<glob> function.
e50bb9a1 860
722d2a37 861=head2 Regression tests for suidperl
e50bb9a1 862
722d2a37 863C<suidperl> is deprecated in favour of common sense.
e50bb9a1 864
722d2a37 865=head2 Cached hash values
e50bb9a1 866
722d2a37 867We have shared hash keys, which perform the same job.
e50bb9a1 868
722d2a37 869=head2 Add compression modules
e50bb9a1 870
722d2a37
SC
871The compression modules are a little heavy; meanwhile, Nick Clark is
872working on experimental pragmata to do transparent decompression on
873input.
e50bb9a1 874
722d2a37 875=head2 Reorganise documentation into tutorials/references
e50bb9a1 876
722d2a37 877Could not get consensus on P5P about this.
e50bb9a1 878
722d2a37
SC
879=head2 Remove distinction between functions and operators
880
881Caution: highly flammable.
882
883=head2 Make XS easier to use
e50bb9a1 884
722d2a37 885Use C<Inline> instead, or SWIG.
e50bb9a1 886
722d2a37 887=head2 Make embedding easier to use
e50bb9a1 888
722d2a37 889Use C<Inline::CPR>.
e50bb9a1 890
722d2a37 891=head2 man for perl
04c70446 892
1577cd80 893See the Perl Power Tools. ( http://language.perl.com/ppt/ )
04c70446 894
722d2a37 895=head2 my $Package::variable
04c70446 896
722d2a37 897Use C<our> instead.
04c70446 898
722d2a37 899=head2 "or" tests defined, not truth
04c70446 900
722d2a37 901Suggesting this on P5P B<will> cause a boring and interminable flamewar.
04c70446 902
722d2a37 903=head2 "class"-based lexicals
04c70446 904
cbb3fa72 905Use flyweight objects, secure hashes or, dare I say it, pseudo-hashes instead.
f86a8bc5 906(Or whatever will replace pseudohashes in 5.10.)
04c70446 907
722d2a37 908=head2 byteperl
04c70446 909
722d2a37 910C<ByteLoader> covers this.
04c70446 911
722d2a37 912=head2 Lazy evaluation / tail recursion removal
04c70446 913
f86a8bc5
JH
914C<List::Util> gives first() (a short-circuiting grep); tail recursion
915removal is done manually, with C<goto &whoami;>. (However, MJD has
916found that C<goto &whoami> introduces a performance penalty, so maybe
917there should be a way to do this after all: C<sub foo {START: ... goto
918START;> is better.)
0562c0e3
JH
919
920=head2 Make "use utf8" the default
921
f86a8bc5
JH
922Because of backward compatibility this is difficult: scripts could not
923contain B<any legacy eight-bit data> (like Latin-1) anymore, even in
924string literals or pod. Also would introduce a measurable slowdown of
925at least few percentages since all regular expression operations would
926be done in full UTF-8. But if you want to try this, add
927-DUSE_UTF8_SCRIPTS to your compilation flags.
928
3298bd4d
JH
929=head2 Unicode collation and normalization
930
931The Unicode::Collate and Unicode::Normalize modules
932by SADAHIRO Tomoyuki have been included since 5.8.0.
933
934 Collation? http://www.unicode.org/unicode/reports/tr10/
935 Normalization? http://www.unicode.org/unicode/reports/tr15/
0562c0e3 936
1626a787
JH
937=head2 pack/unpack tutorial
938
939Wolfgang Laun finished what Simon Cozens started.
940
3298bd4d 941=cut