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