This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Propagating const outwards from Perl_moreswitches() is to be done.
[perl5.git] / pod / perltodo.pod
CommitLineData
7711098a
GS
1=head1 NAME
2
3perltodo - Perl TO-DO List
4
5=head1 DESCRIPTION
e50bb9a1 6
0bdfc961
NC
7This is a list of wishes for Perl. The tasks we think are smaller or easier
8are listed first. Anyone is welcome to work on any of these, but it's a good
9idea to first contact I<perl5-porters@perl.org> to avoid duplication of
10effort. By all means contact a pumpking privately first if you prefer.
e50bb9a1 11
0bdfc961
NC
12Whilst patches to make the list shorter are most welcome, ideas to add to
13the list are also encouraged. Check the perl5-porters archives for past
14ideas, and any discussion about them. One set of archives may be found at:
e50bb9a1 15
0bdfc961 16 http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/
938c8732 17
617eabfa
NC
18What can we offer you in return? Fame, fortune, and everlasting glory? Maybe
19not, but if your patch is incorporated, then we'll add your name to the
20F<AUTHORS> file, which ships in the official distribution. How many other
21programming languages offer you 1 line of immortality?
938c8732 22
0bdfc961 23=head1 Tasks that only need Perl knowledge
e50bb9a1 24
412f19a0
NC
25=head2 merge common code in installperl and installman
26
27There are some common subroutines and a common C<BEGIN> block in F<installperl>
28and F<installman>. These should probably be merged. It would also be good to
29check for duplication in all the utility scripts supplied in the source
30tarball. It might be good to move them all to a subdirectory, but this would
31require careful checking to find all places that call them, and change those
32correctly.
33
0bdfc961 34=head2 common test code for timed bail out
e50bb9a1 35
0bdfc961
NC
36Write portable self destruct code for tests to stop them burning CPU in
37infinite loops. This needs to avoid using alarm, as some of the tests are
38testing alarm/sleep or timers.
e50bb9a1 39
87a942b1 40=head2 POD -E<gt> HTML conversion in the core still sucks
e50bb9a1 41
938c8732 42Which is crazy given just how simple POD purports to be, and how simple HTML
adebf063
NC
43can be. It's not actually I<as> simple as it sounds, particularly with the
44flexibility POD allows for C<=item>, but it would be good to improve the
45visual appeal of the HTML generated, and to avoid it having any validation
46errors. See also L</make HTML install work>, as the layout of installation tree
47is needed to improve the cross-linking.
938c8732 48
dc0fb092
SP
49The addition of C<Pod::Simple> and its related modules may make this task
50easier to complete.
51
8537f021
RGS
52=head2 merge checkpods and podchecker
53
54F<pod/checkpods.PL> (and C<make check> in the F<pod/> subdirectory)
55implements a very basic check for pod files, but the errors it discovers
56aren't found by podchecker. Add this check to podchecker, get rid of
57checkpods and have C<make check> use podchecker.
58
aa237293
NC
59=head2 Parallel testing
60
b2e2905c 61(This probably impacts much more than the core: also the Test::Harness
02f21748
RGS
62and TAP::* modules on CPAN.)
63
aa237293
NC
64The core regression test suite is getting ever more comprehensive, which has
65the side effect that it takes longer to run. This isn't so good. Investigate
66whether it would be feasible to give the harness script the B<option> of
67running sets of tests in parallel. This would be useful for tests in
68F<t/op/*.t> and F<t/uni/*.t> and maybe some sets of tests in F<lib/>.
69
70Questions to answer
71
72=over 4
73
74=item 1
75
76How does screen layout work when you're running more than one test?
77
78=item 2
79
80How does the caller of test specify how many tests to run in parallel?
81
82=item 3
83
84How do setup/teardown tests identify themselves?
85
86=back
87
88Pugs already does parallel testing - can their approach be re-used?
89
0bdfc961 90=head2 Make Schwern poorer
e50bb9a1 91
613bd4f7 92We should have tests for everything. When all the core's modules are tested,
0bdfc961
NC
93Schwern has promised to donate to $500 to TPF. We may need volunteers to
94hold him upside down and shake vigorously in order to actually extract the
95cash.
3958b146 96
0bdfc961 97=head2 Improve the coverage of the core tests
e50bb9a1 98
02f21748
RGS
99Use Devel::Cover to ascertain the core modules's test coverage, then add
100tests that are currently missing.
30222c0f 101
0bdfc961 102=head2 test B
e50bb9a1 103
0bdfc961 104A full test suite for the B module would be nice.
e50bb9a1 105
636e63cb
NC
106=head2 Deparse inlined constants
107
108Code such as this
109
110 use constant PI => 4;
111 warn PI
112
113will currently deparse as
114
115 use constant ('PI', 4);
116 warn 4;
117
118because the tokenizer inlines the value of the constant subroutine C<PI>.
119This allows various compile time optimisations, such as constant folding
120and dead code elimination. Where these haven't happened (such as the example
121above) it ought be possible to make B::Deparse work out the name of the
122original constant, because just enough information survives in the symbol
123table to do this. Specifically, the same scalar is used for the constant in
124the optree as is used for the constant subroutine, so by iterating over all
125symbol tables and generating a mapping of SV address to constant name, it
126would be possible to provide B::Deparse with this functionality.
127
0bdfc961 128=head2 A decent benchmark
e50bb9a1 129
617eabfa 130C<perlbench> seems impervious to any recent changes made to the perl core. It
0bdfc961
NC
131would be useful to have a reasonable general benchmarking suite that roughly
132represented what current perl programs do, and measurably reported whether
133tweaks to the core improve, degrade or don't really affect performance, to
134guide people attempting to optimise the guts of perl. Gisle would welcome
135new tests for perlbench.
6168cf99 136
0bdfc961 137=head2 fix tainting bugs
6168cf99 138
0bdfc961
NC
139Fix the bugs revealed by running the test suite with the C<-t> switch (via
140C<make test.taintwarn>).
e50bb9a1 141
0bdfc961 142=head2 Dual life everything
e50bb9a1 143
0bdfc961
NC
144As part of the "dists" plan, anything that doesn't belong in the smallest perl
145distribution needs to be dual lifed. Anything else can be too. Figure out what
146changes would be needed to package that module and its tests up for CPAN, and
147do so. Test it with older perl releases, and fix the problems you find.
e50bb9a1 148
a393eb28
RGS
149To make a minimal perl distribution, it's useful to look at
150F<t/lib/commonsense.t>.
151
0bdfc961 152=head2 Improving C<threads::shared>
722d2a37 153
0bdfc961
NC
154Investigate whether C<threads::shared> could share aggregates properly with
155only Perl level changes to shared.pm
722d2a37 156
0bdfc961 157=head2 POSIX memory footprint
e50bb9a1 158
0bdfc961
NC
159Ilya observed that use POSIX; eats memory like there's no tomorrow, and at
160various times worked to cut it down. There is probably still fat to cut out -
161for example POSIX passes Exporter some very memory hungry data structures.
e50bb9a1 162
eed36644
NC
163=head2 embed.pl/makedef.pl
164
165There is a script F<embed.pl> that generates several header files to prefix
166all of Perl's symbols in a consistent way, to provide some semblance of
167namespace support in C<C>. Functions are declared in F<embed.fnc>, variables
907b3e23 168in F<interpvar.h>. Quite a few of the functions and variables
eed36644
NC
169are conditionally declared there, using C<#ifdef>. However, F<embed.pl>
170doesn't understand the C macros, so the rules about which symbols are present
171when is duplicated in F<makedef.pl>. Writing things twice is bad, m'kay.
172It would be good to teach C<embed.pl> to understand the conditional
173compilation, and hence remove the duplication, and the mistakes it has caused.
e50bb9a1 174
801de10e
NC
175=head2 use strict; and AutoLoad
176
177Currently if you write
178
179 package Whack;
180 use AutoLoader 'AUTOLOAD';
181 use strict;
182 1;
183 __END__
184 sub bloop {
185 print join (' ', No, strict, here), "!\n";
186 }
187
188then C<use strict;> isn't in force within the autoloaded subroutines. It would
189be more consistent (and less surprising) to arrange for all lexical pragmas
190in force at the __END__ block to be in force within each autoloaded subroutine.
191
773b3597
RGS
192There's a similar problem with SelfLoader.
193
0bdfc961 194=head1 Tasks that need a little sysadmin-type knowledge
e50bb9a1 195
0bdfc961
NC
196Or if you prefer, tasks that you would learn from, and broaden your skills
197base...
e50bb9a1 198
cd793d32 199=head2 make HTML install work
e50bb9a1 200
adebf063
NC
201There is an C<installhtml> target in the Makefile. It's marked as
202"experimental". It would be good to get this tested, make it work reliably, and
203remove the "experimental" tag. This would include
204
205=over 4
206
207=item 1
208
209Checking that cross linking between various parts of the documentation works.
210In particular that links work between the modules (files with POD in F<lib/>)
211and the core documentation (files in F<pod/>)
212
213=item 2
214
617eabfa
NC
215Work out how to split C<perlfunc> into chunks, preferably one per function
216group, preferably with general case code that could be used elsewhere.
217Challenges here are correctly identifying the groups of functions that go
218together, and making the right named external cross-links point to the right
219page. Things to be aware of are C<-X>, groups such as C<getpwnam> to
220C<endservent>, two or more C<=items> giving the different parameter lists, such
221as
adebf063
NC
222
223 =item substr EXPR,OFFSET,LENGTH,REPLACEMENT
adebf063 224 =item substr EXPR,OFFSET,LENGTH
adebf063
NC
225 =item substr EXPR,OFFSET
226
227and different parameter lists having different meanings. (eg C<select>)
228
229=back
3a89a73c 230
0bdfc961
NC
231=head2 compressed man pages
232
233Be able to install them. This would probably need a configure test to see how
234the system does compressed man pages (same directory/different directory?
235same filename/different filename), as well as tweaking the F<installman> script
236to compress as necessary.
237
30222c0f
NC
238=head2 Add a code coverage target to the Makefile
239
240Make it easy for anyone to run Devel::Cover on the core's tests. The steps
241to do this manually are roughly
242
243=over 4
244
245=item *
246
247do a normal C<Configure>, but include Devel::Cover as a module to install
248(see F<INSTALL> for how to do this)
249
250=item *
251
252 make perl
253
254=item *
255
256 cd t; HARNESS_PERL_SWITCHES=-MDevel::Cover ./perl -I../lib harness
257
258=item *
259
260Process the resulting Devel::Cover database
261
262=back
263
264This just give you the coverage of the F<.pm>s. To also get the C level
265coverage you need to
266
267=over 4
268
269=item *
270
271Additionally tell C<Configure> to use the appropriate C compiler flags for
272C<gcov>
273
274=item *
275
276 make perl.gcov
277
278(instead of C<make perl>)
279
280=item *
281
282After running the tests run C<gcov> to generate all the F<.gcov> files.
283(Including down in the subdirectories of F<ext/>
284
285=item *
286
287(From the top level perl directory) run C<gcov2perl> on all the C<.gcov> files
288to get their stats into the cover_db directory.
289
290=item *
291
292Then process the Devel::Cover database
293
294=back
295
296It would be good to add a single switch to C<Configure> to specify that you
297wanted to perform perl level coverage, and another to specify C level
298coverage, and have C<Configure> and the F<Makefile> do all the right things
299automatically.
300
02f21748 301=head2 Make Config.pm cope with differences between built and installed perl
0bdfc961
NC
302
303Quite often vendors ship a perl binary compiled with their (pay-for)
304compilers. People install a free compiler, such as gcc. To work out how to
305build extensions, Perl interrogates C<%Config>, so in this situation
306C<%Config> describes compilers that aren't there, and extension building
307fails. This forces people into choosing between re-compiling perl themselves
308using the compiler they have, or only using modules that the vendor ships.
309
310It would be good to find a way teach C<Config.pm> about the installation setup,
311possibly involving probing at install time or later, so that the C<%Config> in
312a binary distribution better describes the installed machine, when the
313installed machine differs from the build machine in some significant way.
314
728f4ecd
NC
315=head2 linker specification files
316
317Some platforms mandate that you provide a list of a shared library's external
318symbols to the linker, so the core already has the infrastructure in place to
319do this for generating shared perl libraries. My understanding is that the
320GNU toolchain can accept an optional linker specification file, and restrict
321visibility just to symbols declared in that file. It would be good to extend
322F<makedef.pl> to support this format, and to provide a means within
323C<Configure> to enable it. This would allow Unix users to test that the
324export list is correct, and to build a perl that does not pollute the global
325namespace with private symbols.
326
a229ae3b
RGS
327=head2 Cross-compile support
328
329Currently C<Configure> understands C<-Dusecrosscompile> option. This option
330arranges for building C<miniperl> for TARGET machine, so this C<miniperl> is
331assumed then to be copied to TARGET machine and used as a replacement of full
332C<perl> executable.
333
d1307786 334This could be done little differently. Namely C<miniperl> should be built for
a229ae3b 335HOST and then full C<perl> with extensions should be compiled for TARGET.
d1307786 336This, however, might require extra trickery for %Config: we have one config
87a942b1
JH
337first for HOST and then another for TARGET. Tools like MakeMaker will be
338mightily confused. Having around two different types of executables and
339libraries (HOST and TARGET) makes life interesting for Makefiles and
340shell (and Perl) scripts. There is $Config{run}, normally empty, which
341can be used as an execution wrapper. Also note that in some
342cross-compilation/execution environments the HOST and the TARGET do
343not see the same filesystem(s), the $Config{run} may need to do some
344file/directory copying back and forth.
0bdfc961 345
8537f021
RGS
346=head2 roffitall
347
348Make F<pod/roffitall> be updated by F<pod/buildtoc>.
349
0bdfc961
NC
350=head1 Tasks that need a little C knowledge
351
352These tasks would need a little C knowledge, but don't need any specific
353background or experience with XS, or how the Perl interpreter works
354
fbf638cb
RGS
355=head2 Modernize the order of directories in @INC
356
357The way @INC is laid out by default, one cannot upgrade core (dual-life)
358modules without overwriting files. This causes problems for binary
3d14fd97
AD
359package builders. One possible proposal is laid out in this
360message:
361L<http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2002-04/msg02380.html>.
fbf638cb 362
bcbaa2d5
RGS
363=head2 -Duse32bit*
364
365Natively 64-bit systems need neither -Duse64bitint nor -Duse64bitall.
366On these systems, it might be the default compilation mode, and there
367is currently no guarantee that passing no use64bitall option to the
368Configure process will build a 32bit perl. Implementing -Duse32bit*
369options would be nice for perl 5.12.
370
0bdfc961 371=head2 Make it clear from -v if this is the exact official release
89007cb3 372
617eabfa
NC
373Currently perl from C<p4>/C<rsync> ships with a F<patchlevel.h> file that
374usually defines one local patch, of the form "MAINT12345" or "RC1". The output
375of perl -v doesn't report that a perl isn't an official release, and this
89007cb3 376information can get lost in bugs reports. Because of this, the minor version
fa11829f 377isn't bumped up until RC time, to minimise the possibility of versions of perl
89007cb3
NC
378escaping that believe themselves to be newer than they actually are.
379
380It would be useful to find an elegant way to have the "this is an interim
381maintenance release" or "this is a release candidate" in the terse -v output,
382and have it so that it's easy for the pumpking to remove this just as the
383release tarball is rolled up. This way the version pulled out of rsync would
384always say "I'm a development release" and it would be safe to bump the
385reported minor version as soon as a release ships, which would aid perl
386developers.
387
0bdfc961
NC
388This task is really about thinking of an elegant way to arrange the C source
389such that it's trivial for the Pumpking to flag "this is an official release"
390when making a tarball, yet leave the default source saying "I'm not the
391official release".
392
fee0a0f7 393=head2 Profile Perl - am I hot or not?
62403a3c 394
fee0a0f7
NC
395The Perl source code is stable enough that it makes sense to profile it,
396identify and optimise the hotspots. It would be good to measure the
397performance of the Perl interpreter using free tools such as cachegrind,
398gprof, and dtrace, and work to reduce the bottlenecks they reveal.
399
400As part of this, the idea of F<pp_hot.c> is that it contains the I<hot> ops,
401the ops that are most commonly used. The idea is that by grouping them, their
402object code will be adjacent in the executable, so they have a greater chance
403of already being in the CPU cache (or swapped in) due to being near another op
404already in use.
62403a3c
NC
405
406Except that it's not clear if these really are the most commonly used ops. So
fee0a0f7
NC
407as part of exercising your skills with coverage and profiling tools you might
408want to determine what ops I<really> are the most commonly used. And in turn
409suggest evictions and promotions to achieve a better F<pp_hot.c>.
62403a3c 410
98fed0ad
NC
411=head2 Allocate OPs from arenas
412
413Currently all new OP structures are individually malloc()ed and free()d.
414All C<malloc> implementations have space overheads, and are now as fast as
415custom allocates so it would both use less memory and less CPU to allocate
416the various OP structures from arenas. The SV arena code can probably be
417re-used for this.
418
539f2c54
JC
419Note that Configuring perl with C<-Accflags=-DPL_OP_SLAB_ALLOC> will use
420Perl_Slab_alloc() to pack optrees into a contiguous block, which is
421probably superior to the use of OP arenas, esp. from a cache locality
422standpoint. See L<Profile Perl - am I hot or not?>.
423
a229ae3b 424=head2 Improve win32/wince.c
0bdfc961 425
a229ae3b 426Currently, numerous functions look virtually, if not completely,
02f21748 427identical in both C<win32/wince.c> and C<win32/win32.c> files, which can't
6d71adcd
NC
428be good.
429
c5b31784
SH
430=head2 Use secure CRT functions when building with VC8 on Win32
431
432Visual C++ 2005 (VC++ 8.x) deprecated a number of CRT functions on the basis
433that they were "unsafe" and introduced differently named secure versions of
434them as replacements, e.g. instead of writing
435
436 FILE* f = fopen(__FILE__, "r");
437
438one should now write
439
440 FILE* f;
441 errno_t err = fopen_s(&f, __FILE__, "r");
442
443Currently, the warnings about these deprecations have been disabled by adding
444-D_CRT_SECURE_NO_DEPRECATE to the CFLAGS. It would be nice to remove that
445warning suppressant and actually make use of the new secure CRT functions.
446
447There is also a similar issue with POSIX CRT function names like fileno having
448been deprecated in favour of ISO C++ conformant names like _fileno. These
26a6faa8 449warnings are also currently suppressed by adding -D_CRT_NONSTDC_NO_DEPRECATE. It
c5b31784
SH
450might be nice to do as Microsoft suggest here too, although, unlike the secure
451functions issue, there is presumably little or no benefit in this case.
452
6b632b43
NC
453=head2 __FUNCTION__ for MSVC-pre-7.0
454
455Jarkko notes that one can things morally equivalent to C<__FUNCTION__>
456(or C<__func__>) even in MSVC-pre-7.0, contrary to popular belief.
457See L<http://www.codeproject.com/debug/extendedtrace.asp> if you feel like
458making C<PERL_MEM_LOG> more useful on Win32.
459
6d71adcd
NC
460=head1 Tasks that need a knowledge of XS
461
462These tasks would need C knowledge, and roughly the level of knowledge of
463the perl API that comes from writing modules that use XS to interface to
464C.
465
6d71adcd
NC
466=head2 autovivification
467
468Make all autovivification consistent w.r.t LVALUE/RVALUE and strict/no strict;
469
470This task is incremental - even a little bit of work on it will help.
471
472=head2 Unicode in Filenames
473
474chdir, chmod, chown, chroot, exec, glob, link, lstat, mkdir, open,
475opendir, qx, readdir, readlink, rename, rmdir, stat, symlink, sysopen,
476system, truncate, unlink, utime, -X. All these could potentially accept
477Unicode filenames either as input or output (and in the case of system
478and qx Unicode in general, as input or output to/from the shell).
479Whether a filesystem - an operating system pair understands Unicode in
480filenames varies.
481
482Known combinations that have some level of understanding include
483Microsoft NTFS, Apple HFS+ (In Mac OS 9 and X) and Apple UFS (in Mac
484OS X), NFS v4 is rumored to be Unicode, and of course Plan 9. How to
485create Unicode filenames, what forms of Unicode are accepted and used
486(UCS-2, UTF-16, UTF-8), what (if any) is the normalization form used,
487and so on, varies. Finding the right level of interfacing to Perl
488requires some thought. Remember that an OS does not implicate a
489filesystem.
490
491(The Windows -C command flag "wide API support" has been at least
492temporarily retired in 5.8.1, and the -C has been repurposed, see
493L<perlrun>.)
494
87a942b1
JH
495Most probably the right way to do this would be this:
496L</"Virtualize operating system access">.
497
6d71adcd
NC
498=head2 Unicode in %ENV
499
500Currently the %ENV entries are always byte strings.
87a942b1 501See L</"Virtualize operating system access">.
6d71adcd 502
1f2e7916
JD
503=head2 Unicode and glob()
504
505Currently glob patterns and filenames returned from File::Glob::glob()
87a942b1 506are always byte strings. See L</"Virtualize operating system access">.
1f2e7916 507
dbb0c492
RGS
508=head2 Unicode and lc/uc operators
509
510Some built-in operators (C<lc>, C<uc>, etc.) behave differently, based on
511what the internal encoding of their argument is. That should not be the
512case. Maybe add a pragma to switch behaviour.
513
6d71adcd
NC
514=head2 use less 'memory'
515
516Investigate trade offs to switch out perl's choices on memory usage.
517Particularly perl should be able to give memory back.
518
519This task is incremental - even a little bit of work on it will help.
520
521=head2 Re-implement C<:unique> in a way that is actually thread-safe
522
523The old implementation made bad assumptions on several levels. A good 90%
524solution might be just to make C<:unique> work to share the string buffer
525of SvPVs. That way large constant strings can be shared between ithreads,
526such as the configuration information in F<Config>.
527
528=head2 Make tainting consistent
529
530Tainting would be easier to use if it didn't take documented shortcuts and
531allow taint to "leak" everywhere within an expression.
532
533=head2 readpipe(LIST)
534
535system() accepts a LIST syntax (and a PROGRAM LIST syntax) to avoid
536running a shell. readpipe() (the function behind qx//) could be similarly
537extended.
538
539=head2 strcat(), strcpy(), strncat(), strncpy(), sprintf(), vsprintf()
540
541Maybe create a utility that checks after each libperl.a creation that
542none of the above (nor sprintf(), vsprintf(), or *SHUDDER* gets())
543ever creep back to libperl.a.
544
545 nm libperl.a | ./miniperl -alne '$o = $F[0] if /:$/; print "$o $F[1]" if $F[0] eq "U" && $F[1] =~ /^(?:strn?c(?:at|py)|v?sprintf|gets)$/'
546
547Note, of course, that this will only tell whether B<your> platform
548is using those naughty interfaces.
549
550=head2 Audit the code for destruction ordering assumptions
551
552Change 25773 notes
553
554 /* Need to check SvMAGICAL, as during global destruction it may be that
555 AvARYLEN(av) has been freed before av, and hence the SvANY() pointer
556 is now part of the linked list of SV heads, rather than pointing to
557 the original body. */
558 /* FIXME - audit the code for other bugs like this one. */
559
560adding the C<SvMAGICAL> check to
561
562 if (AvARYLEN(av) && SvMAGICAL(AvARYLEN(av))) {
563 MAGIC *mg = mg_find (AvARYLEN(av), PERL_MAGIC_arylen);
564
565Go through the core and look for similar assumptions that SVs have particular
566types, as all bets are off during global destruction.
567
749904bf
JH
568=head2 Extend PerlIO and PerlIO::Scalar
569
570PerlIO::Scalar doesn't know how to truncate(). Implementing this
571would require extending the PerlIO vtable.
572
573Similarly the PerlIO vtable doesn't know about formats (write()), or
574about stat(), or chmod()/chown(), utime(), or flock().
575
576(For PerlIO::Scalar it's hard to see what e.g. mode bits or ownership
577would mean.)
578
579PerlIO doesn't do directories or symlinks, either: mkdir(), rmdir(),
580opendir(), closedir(), seekdir(), rewinddir(), glob(); symlink(),
581readlink().
582
94da6c29
JH
583See also L</"Virtualize operating system access">.
584
3236f110
NC
585=head2 -C on the #! line
586
587It should be possible to make -C work correctly if found on the #! line,
588given that all perl command line options are strict ASCII, and -C changes
589only the interpretation of non-ASCII characters, and not for the script file
590handle. To make it work needs some investigation of the ordering of function
591calls during startup, and (by implication) a bit of tweaking of that order.
592
81622873
NC
593=head2 Propagate const outwards from Perl_moreswitches()
594
595Change 32057 changed the parameter and return value of C<Perl_moreswitches()>
596from <char *> to <const char *>. It should now be possible to propagate
597const-correctness outwards to C<S_parse_body()>, C<Perl_moreswitches()>
598and C<Perl_yylex()>.
599
3236f110 600
0bdfc961 601=head1 Tasks that need a knowledge of the interpreter
3298bd4d 602
0bdfc961
NC
603These tasks would need C knowledge, and knowledge of how the interpreter works,
604or a willingness to learn.
3298bd4d 605
636e63cb
NC
606=head2 state variable initialization in list context
607
608Currently this is illegal:
609
610 state ($a, $b) = foo();
611
612The current Perl 6 design is that C<state ($a) = foo();> and
613C<(state $a) = foo();> have different semantics, which is tricky to implement
614in Perl 5 as currently the produce the same opcode trees. It would be useful
615to clarify that the Perl 6 design is firm, and then implement the necessary
616code in Perl 5. There are comments in C<Perl_newASSIGNOP()> that show the
617code paths taken by various assignment constructions involving state variables.
618
4fedb12c
RGS
619=head2 Implement $value ~~ 0 .. $range
620
621It would be nice to extend the syntax of the C<~~> operator to also
622understand numeric (and maybe alphanumeric) ranges.
a393eb28
RGS
623
624=head2 A does() built-in
625
626Like ref(), only useful. It would call the C<DOES> method on objects; it
627would also tell whether something can be dereferenced as an
628array/hash/etc., or used as a regexp, etc.
629L<http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2007-03/msg00481.html>
630
631=head2 Tied filehandles and write() don't mix
632
633There is no method on tied filehandles to allow them to be called back by
634formats.
4fedb12c 635
d10fc472 636=head2 Attach/detach debugger from running program
1626a787 637
cd793d32
NC
638The old perltodo notes "With C<gdb>, you can attach the debugger to a running
639program if you pass the process ID. It would be good to do this with the Perl
0bdfc961
NC
640debugger on a running Perl program, although I'm not sure how it would be
641done." ssh and screen do this with named pipes in /tmp. Maybe we can too.
1626a787 642
a8cb5b9e
RGS
643=head2 Optimize away empty destructors
644
645Defining an empty DESTROY method might be useful (notably in
646AUTOLOAD-enabled classes), but it's still a bit expensive to call. That
647could probably be optimized.
648
0bdfc961
NC
649=head2 LVALUE functions for lists
650
651The old perltodo notes that lvalue functions don't work for list or hash
652slices. This would be good to fix.
653
654=head2 LVALUE functions in the debugger
655
656The old perltodo notes that lvalue functions don't work in the debugger. This
657would be good to fix.
658
0bdfc961
NC
659=head2 regexp optimiser optional
660
661The regexp optimiser is not optional. It should configurable to be, to allow
662its performance to be measured, and its bugs to be easily demonstrated.
663
02f21748
RGS
664=head2 delete &function
665
666Allow to delete functions. One can already undef them, but they're still
667in the stash.
668
ef36c6a7
RGS
669=head2 C</w> regex modifier
670
671That flag would enable to match whole words, and also to interpolate
672arrays as alternations. With it, C</P/w> would be roughly equivalent to:
673
674 do { local $"='|'; /\b(?:P)\b/ }
675
676See L<http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2007-01/msg00400.html>
677for the discussion.
678
0bdfc961
NC
679=head2 optional optimizer
680
681Make the peephole optimizer optional. Currently it performs two tasks as
682it walks the optree - genuine peephole optimisations, and necessary fixups of
683ops. It would be good to find an efficient way to switch out the
684optimisations whilst keeping the fixups.
685
686=head2 You WANT *how* many
687
688Currently contexts are void, scalar and list. split has a special mechanism in
689place to pass in the number of return values wanted. It would be useful to
690have a general mechanism for this, backwards compatible and little speed hit.
691This would allow proposals such as short circuiting sort to be implemented
692as a module on CPAN.
693
694=head2 lexical aliases
695
696Allow lexical aliases (maybe via the syntax C<my \$alias = \$foo>.
697
698=head2 entersub XS vs Perl
699
700At the moment pp_entersub is huge, and has code to deal with entering both
701perl and XS subroutines. Subroutine implementations rarely change between
702perl and XS at run time, so investigate using 2 ops to enter subs (one for
703XS, one for perl) and swap between if a sub is redefined.
2810d901
NC
704
705=head2 Self ties
706
707self ties are currently illegal because they caused too many segfaults. Maybe
708the causes of these could be tracked down and self-ties on all types re-
709instated.
0bdfc961
NC
710
711=head2 Optimize away @_
712
713The old perltodo notes "Look at the "reification" code in C<av.c>".
714
16fc99ce
NC
715=head2 Properly Unicode safe tokeniser and pads.
716
717The tokeniser isn't actually very UTF-8 clean. C<use utf8;> is a hack -
718variable names are stored in stashes as raw bytes, without the utf-8 flag
719set. The pad API only takes a C<char *> pointer, so that's all bytes too. The
720tokeniser ignores the UTF-8-ness of C<PL_rsfp>, or any SVs returned from
721source filters. All this could be fixed.
722
f092b1f4
RGS
723=head2 The yada yada yada operators
724
725Perl 6's Synopsis 3 says:
726
727I<The ... operator is the "yada, yada, yada" list operator, which is used as
728the body in function prototypes. It complains bitterly (by calling fail)
729if it is ever executed. Variant ??? calls warn, and !!! calls die.>
730
731Those would be nice to add to Perl 5. That could be done without new ops.
732
87a942b1
JH
733=head2 Virtualize operating system access
734
735Implement a set of "vtables" that virtualizes operating system access
736(open(), mkdir(), unlink(), readdir(), getenv(), etc.) At the very
737least these interfaces should take SVs as "name" arguments instead of
738bare char pointers; probably the most flexible and extensible way
e1a3d5d1
JH
739would be for the Perl-facing interfaces to accept HVs. The system
740needs to be per-operating-system and per-file-system
741hookable/filterable, preferably both from XS and Perl level
87a942b1
JH
742(L<perlport/"Files and Filesystems"> is good reading at this point,
743in fact, all of L<perlport> is.)
744
e1a3d5d1
JH
745This has actually already been implemented (but only for Win32),
746take a look at F<iperlsys.h> and F<win32/perlhost.h>. While all Win32
747variants go through a set of "vtables" for operating system access,
748non-Win32 systems currently go straight for the POSIX/UNIX-style
749system/library call. Similar system as for Win32 should be
750implemented for all platforms. The existing Win32 implementation
751probably does not need to survive alongside this proposed new
752implementation, the approaches could be merged.
87a942b1
JH
753
754What would this give us? One often-asked-for feature this would
94da6c29
JH
755enable is using Unicode for filenames, and other "names" like %ENV,
756usernames, hostnames, and so forth.
757(See L<perlunicode/"When Unicode Does Not Happen">.)
758
759But this kind of virtualization would also allow for things like
760virtual filesystems, virtual networks, and "sandboxes" (though as long
761as dynamic loading of random object code is allowed, not very safe
762sandboxes since external code of course know not of Perl's vtables).
763An example of a smaller "sandbox" is that this feature can be used to
764implement per-thread working directories: Win32 already does this.
765
766See also L</"Extend PerlIO and PerlIO::Scalar">.
87a942b1 767
ac6197af
NC
768=head2 Investigate PADTMP hash pessimisation
769
770The peephole optimier converts constants used for hash key lookups to shared
771hash key scalars. Under ithreads, something is undoing this work. See
772See http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2007-09/msg00793.html
773
0bdfc961
NC
774=head1 Big projects
775
776Tasks that will get your name mentioned in the description of the "Highlights
87a942b1 777of 5.12"
0bdfc961
NC
778
779=head2 make ithreads more robust
780
4e577f8b 781Generally make ithreads more robust. See also L</iCOW>
0bdfc961
NC
782
783This task is incremental - even a little bit of work on it will help, and
784will be greatly appreciated.
785
6c047da7
YST
786One bit would be to write the missing code in sv.c:Perl_dirp_dup.
787
59c7f7d5
RGS
788Fix Perl_sv_dup, et al so that threads can return objects.
789
0bdfc961
NC
790=head2 iCOW
791
792Sarathy and Arthur have a proposal for an improved Copy On Write which
793specifically will be able to COW new ithreads. If this can be implemented
794it would be a good thing.
795
796=head2 (?{...}) closures in regexps
797
798Fix (or rewrite) the implementation of the C</(?{...})/> closures.
799
800=head2 A re-entrant regexp engine
801
802This will allow the use of a regex from inside (?{ }), (??{ }) and
803(?(?{ })|) constructs.
6bda09f9 804
6bda09f9
YO
805=head2 Add class set operations to regexp engine
806
807Apparently these are quite useful. Anyway, Jeffery Friedl wants them.
808
809demerphq has this on his todo list, but right at the bottom.