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