Commit | Line | Data |
---|---|---|
7711098a GS |
1 | =head1 NAME |
2 | ||
3 | perltodo - Perl TO-DO List | |
4 | ||
5 | =head1 DESCRIPTION | |
e50bb9a1 | 6 | |
722d2a37 | 7 | This is a list of wishes for Perl. Send updates to |
e50bb9a1 GS |
8 | I<perl5-porters@perl.org>. If you want to work on any of these |
9 | projects, be sure to check the perl5-porters archives for past ideas, | |
10 | flames, and propaganda. This will save you time and also prevent you | |
11 | from implementing something that Larry has already vetoed. One set | |
12 | of archives may be found at: | |
13 | ||
14 | http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/ | |
15 | ||
722d2a37 | 16 | =head1 To do during 5.6.x |
e50bb9a1 | 17 | |
722d2a37 | 18 | =head2 Support for I/O disciplines |
e50bb9a1 | 19 | |
722d2a37 SC |
20 | C<perlio> provides this, but the interface could be a lot more |
21 | straightforward. | |
e50bb9a1 | 22 | |
4b3b956a | 23 | =head2 Autoload bytes.pm |
e50bb9a1 | 24 | |
4b3b956a JH |
25 | When the lexer sees, for instance, C<bytes::length>, it should |
26 | automatically load the C<bytes> pragma. | |
27 | ||
28 | =head2 Make "\u{XXXX}" et al work | |
29 | ||
30 | Danger, Will Robinson! Discussing the semantics of C<"\x{F00}">, | |
31 | C<"\xF00"> and C<"\U{F00}"> on P5P I<will> lead to a long and boring | |
32 | flamewar. | |
e50bb9a1 | 33 | |
c6287c21 | 34 | =head2 Create a char *sv_pvprintify(sv, STRLEN *lenp, UV flags) |
0562c0e3 JH |
35 | |
36 | For displaying PVs with control characters, embedded nulls, and Unicode. | |
37 | This would be useful for printing warnings, or data and regex dumping, | |
38 | not_a_number(), and so on. | |
39 | ||
f35392ae JH |
40 | Requirements: should handle both byte and UTF8 strings. isPRINT() |
41 | characters printed as-is, character less than 256 as \xHH, Unicode | |
0661e9a4 JH |
42 | characters as \x{HHH}. Don't assume ASCII-like, either, get somebody |
43 | on EBCDIC to test the output. | |
f35392ae JH |
44 | |
45 | Possible options, controlled by the flags: | |
0661e9a4 | 46 | - whitespace (other than ' ' of isPRINT()) printed as-is |
f35392ae JH |
47 | - use isPRINT_LC() instead of isPRINT() |
48 | - print control characters like this: "\cA" | |
49 | - print control characters like this: "^A" | |
0661e9a4 JH |
50 | - non-PRINTables printed as '.' instead of \xHH |
51 | - use \OOO instead of \xHH | |
52 | - use the C/Perl-metacharacters like \n, \t | |
f35392ae JH |
53 | - have a maximum length for the produced string (read it from *lenp) |
54 | - append a "..." to the produced string if the maximum length is exceeded | |
0661e9a4 | 55 | - really fancy: print unicode characters as \N{...} |
f35392ae | 56 | |
c5fc23ff JH |
57 | NOTE: pv_display(), pv_uni_display(), sv_uni_display() are doing |
58 | something like the above. | |
59 | ||
722d2a37 | 60 | =head2 Overloadable regex assertions |
e50bb9a1 | 61 | |
722d2a37 SC |
62 | This may or may not be possible with the current regular expression |
63 | engine. The idea is that, for instance, C<\b> needs to be | |
64 | algorithmically computed if you're dealing with Thai text. Hence, the | |
65 | B<\b> assertion wants to be overloaded by a function. | |
e50bb9a1 | 66 | |
776f8809 JH |
67 | =head2 Unicode |
68 | ||
69 | =over 4 | |
70 | ||
71 | =item * | |
e50bb9a1 | 72 | |
f34dec15 JH |
73 | Allow for long form of the General Category Properties, e.g |
74 | C<\p{IsOpenPunctuation}>, not just the abbreviated form, e.g. | |
75 | C<\p{IsPs}>. | |
76 | ||
77 | =item * | |
78 | ||
1ac13f9a JH |
79 | Allow for the metaproperties: C<XID Start>, C<XID Continue>, |
80 | C<NF*_NO>, C<NF*_MAYBE> (require the DerivedCoreProperties and | |
81 | DerviceNormalizationProperties files). | |
f34dec15 | 82 | |
71d929cb JH |
83 | There are also multiple value properties still unimplemented: |
84 | C<Numeric Type>, C<East Asian Width>. | |
f34dec15 JH |
85 | |
86 | =item * | |
87 | ||
722d2a37 | 88 | Case Mappings? http://www.unicode.org/unicode/reports/tr21/ |
e50bb9a1 | 89 | |
6f16a292 JH |
90 | lc(), uc(), lcfirst(), and ucfirst() work only for some of the |
91 | simplest cases, where the mapping goes from a single Unicode character | |
92 | to another single Unicode character. See lib/unicore/SpecCase.txt | |
93 | (and CaseFold.txt). | |
ac1256e8 | 94 | |
776f8809 | 95 | =item * |
e50bb9a1 | 96 | |
c6287c21 JH |
97 | They have some tricks Perl doesn't yet implement like character |
98 | class subtraction. | |
e50bb9a1 | 99 | |
722d2a37 | 100 | http://www.unicode.org/unicode/reports/tr18/ |
e50bb9a1 | 101 | |
776f8809 JH |
102 | =back |
103 | ||
104 | See L<perlunicode/UNICODE REGULAR EXPRESSION SUPPORT LEVEL> for what's | |
f34dec15 JH |
105 | there and what's missing. Almost all of Levels 2 and 3 is missing, |
106 | and as of 5.8.0 not even all of Level 1 is there. | |
776f8809 | 107 | |
722d2a37 | 108 | =head2 use Thread for iThreads |
e50bb9a1 | 109 | |
722d2a37 SC |
110 | Artur Bergman's C<iThreads> module is a start on this, but needs to |
111 | be more mature. | |
e50bb9a1 | 112 | |
dd0afe54 AB |
113 | =head2 make perl_clone optionally clone ops |
114 | ||
115 | So that pseudoforking, mod_perl, iThreads and nvi will work properly | |
116 | (but not as efficiently) until the regex engine is fixed to be threadsafe. | |
117 | ||
722d2a37 | 118 | =head2 Work out exit/die semantics for threads |
e50bb9a1 | 119 | |
722d2a37 | 120 | =head2 Typed lexicals for compiler |
e50bb9a1 | 121 | |
722d2a37 | 122 | =head2 Compiler workarounds for Win32 |
e50bb9a1 | 123 | |
722d2a37 | 124 | =head2 AUTOLOADing in the compiler |
e50bb9a1 | 125 | |
722d2a37 | 126 | =head2 Fixing comppadlist when compiling |
e50bb9a1 | 127 | |
722d2a37 | 128 | =head2 Cleaning up exported namespace |
e50bb9a1 | 129 | |
722d2a37 | 130 | =head2 Complete signal handling |
e50bb9a1 | 131 | |
722d2a37 SC |
132 | Add C<PERL_ASYNC_CHECK> to opcodes which loop; replace C<sigsetjmp> with |
133 | C<sigjmp>; check C<wait> for signal safety. | |
e50bb9a1 | 134 | |
722d2a37 | 135 | =head2 Out-of-source builds |
e50bb9a1 | 136 | |
722d2a37 | 137 | This was done for 5.6.0, but needs reworking for 5.7.x |
e50bb9a1 | 138 | |
722d2a37 | 139 | =head2 POSIX realtime support |
e50bb9a1 | 140 | |
722d2a37 SC |
141 | POSIX 1003.1 1996 Edition support--realtime stuff: POSIX semaphores, |
142 | message queues, shared memory, realtime clocks, timers, signals (the | |
143 | metaconfig units mostly already exist for these) | |
e50bb9a1 | 144 | |
722d2a37 | 145 | =head2 UNIX98 support |
e50bb9a1 | 146 | |
722d2a37 | 147 | Reader-writer locks, realtime/asynchronous IO |
e50bb9a1 | 148 | |
722d2a37 | 149 | =head2 IPv6 Support |
e50bb9a1 | 150 | |
722d2a37 SC |
151 | There are non-core modules, such as C<Net::IPv6>, but these will need |
152 | integrating when IPv6 actually starts to really happen. See RFC 2292 | |
153 | and RFC 2553. | |
e50bb9a1 | 154 | |
722d2a37 | 155 | =head2 Long double conversion |
e50bb9a1 | 156 | |
722d2a37 | 157 | Floating point formatting is still causing some weird test failures. |
e50bb9a1 | 158 | |
722d2a37 | 159 | =head2 Locales |
e50bb9a1 | 160 | |
722d2a37 SC |
161 | Locales and Unicode interact with each other in unpleasant ways. |
162 | One possible solution would be to adopt/support ICU: | |
e50bb9a1 | 163 | |
722d2a37 | 164 | http://oss.software.ibm.com/developerworks/opensource/icu/project/ |
e50bb9a1 | 165 | |
722d2a37 | 166 | =head2 Thread-safe regexes |
e50bb9a1 | 167 | |
722d2a37 | 168 | The regular expression engine is currently non-threadsafe. |
e50bb9a1 | 169 | |
722d2a37 | 170 | =head2 Arithmetic on non-Arabic numerals |
e50bb9a1 | 171 | |
722d2a37 | 172 | C<[1234567890]> aren't the only numerals any more. |
e50bb9a1 | 173 | |
722d2a37 | 174 | =head2 POSIX Unicode character classes |
e50bb9a1 | 175 | |
210b36aa | 176 | (C<[=a=]> for equivalance classes, C<[.ch.]> for collation.) |
722d2a37 | 177 | These are dependent on Unicode normalization and collation. |
e50bb9a1 | 178 | |
722d2a37 | 179 | =head2 Factoring out common suffices/prefices in regexps (trie optimization) |
c47ff5f1 | 180 | |
722d2a37 SC |
181 | Currently, the user has to optimize C<foo|far> and C<foo|goo> into |
182 | C<f(?:oo|ar)> and C<[fg]oo> by hand; this could be done automatically. | |
e50bb9a1 | 183 | |
722d2a37 | 184 | =head2 Security audit shipped utilities |
e50bb9a1 | 185 | |
722d2a37 SC |
186 | All the code we ship with Perl needs to be sensible about temporary file |
187 | handling, locking, input validation, and so on. | |
e50bb9a1 | 188 | |
722d2a37 | 189 | =head2 Custom opcodes |
e50bb9a1 | 190 | |
722d2a37 SC |
191 | Have a way to introduce user-defined opcodes without the subroutine call |
192 | overhead of an XSUB; the user should be able to create PP code. Simon | |
193 | Cozens has some ideas on this. | |
e50bb9a1 | 194 | |
722d2a37 | 195 | =head2 DLL Versioning |
e50bb9a1 | 196 | |
d1be9408 | 197 | Windows needs a way to know what version of an XS or C<libperl> DLL it's |
722d2a37 | 198 | loading. |
e50bb9a1 | 199 | |
722d2a37 | 200 | =head2 Introduce @( and @) |
e50bb9a1 | 201 | |
722d2a37 SC |
202 | C<$(> may return "foo bar baz". Unfortunately, since groups can |
203 | theoretically have spaces in their names, this could be one, two or | |
204 | three groups. | |
e50bb9a1 | 205 | |
722d2a37 | 206 | =head2 Floating point handling |
e50bb9a1 | 207 | |
722d2a37 SC |
208 | C<NaN> and C<inf> support is particularly troublesome. |
209 | (fp_classify(), fp_class(), fp_class_d(), class(), isinf(), | |
210 | isfinite(), finite(), isnormal(), unordered(), <ieeefp.h>, | |
211 | <fp_class.h> (there are metaconfig units for all these) (I think), | |
212 | fp_setmask(), fp_getmask(), fp_setround(), fp_getround() | |
213 | (no metaconfig units yet for these). Don't forget finitel(), fp_classl(), | |
214 | fp_class_l(), (yes, both do, unfortunately, exist), and unorderedl().) | |
e50bb9a1 | 215 | |
210b36aa | 216 | As of Perl 5.6.1, there is a Perl macro, Perl_isnan(). |
e50bb9a1 | 217 | |
722d2a37 | 218 | =head2 IV/UV preservation |
e50bb9a1 | 219 | |
722d2a37 SC |
220 | Nicholas Clark has done a lot of work on this, but work is continuing. |
221 | C<+>, C<-> and C<*> work, but guards need to be in place for C<%>, C</>, | |
222 | C<&>, C<oct>, C<hex> and C<pack>. | |
e50bb9a1 | 223 | |
722d2a37 | 224 | =head2 Replace pod2html with something using Pod::Parser |
83df6a1d | 225 | |
722d2a37 SC |
226 | The CPAN module C<Malik::Pod::Html> may be a more suitable basis for a |
227 | C<pod2html> convertor; the current one duplicates the functionality | |
228 | abstracted in C<Pod::Parser>, which makes updating the POD language | |
229 | difficult. | |
e50bb9a1 | 230 | |
722d2a37 | 231 | =head2 Automate module testing on CPAN |
e50bb9a1 | 232 | |
722d2a37 SC |
233 | When a new Perl is being beta tested, porters have to manually grab |
234 | their favourite CPAN modules and test them - this should be done | |
235 | automatically. | |
e50bb9a1 | 236 | |
722d2a37 | 237 | =head2 sendmsg and recvmsg |
83df6a1d | 238 | |
722d2a37 SC |
239 | We have all the other BSD socket functions but these. There are |
240 | metaconfig units for these functions which can be added. To avoid these | |
241 | being new opcodes, a solution similar to the way C<sockatmark> was added | |
242 | would be preferable. (Autoload the C<IO::whatever> module.) | |
e50bb9a1 | 243 | |
722d2a37 | 244 | =head2 Rewrite perlre documentation |
e50bb9a1 | 245 | |
722d2a37 SC |
246 | The new-style patterns need full documentation, and the whole document |
247 | needs to be a lot clearer. | |
e50bb9a1 | 248 | |
722d2a37 | 249 | =head2 Convert example code to IO::Handle filehandles |
e50bb9a1 | 250 | |
722d2a37 | 251 | =head2 Document Win32 choices |
e50bb9a1 | 252 | |
722d2a37 | 253 | =head2 Check new modules |
e50bb9a1 | 254 | |
722d2a37 | 255 | =head2 Make roffitall find pods and libs itself |
e50bb9a1 | 256 | |
722d2a37 | 257 | Simon Cozens has done some work on this but it needs a rethink. |
e50bb9a1 | 258 | |
722d2a37 | 259 | =head1 To do at some point |
e50bb9a1 | 260 | |
722d2a37 SC |
261 | These are ideas that have been regularly tossed around, that most |
262 | people believe should be done maybe during 5.8.x | |
e50bb9a1 | 263 | |
722d2a37 | 264 | =head2 Remove regular expression recursion |
e50bb9a1 | 265 | |
722d2a37 SC |
266 | Because the regular expression engine is recursive, badly designed |
267 | expressions can lead to lots of recursion filling up the stack. Ilya | |
268 | claims that it is easy to convert the engine to being iterative, but | |
269 | this has still not yet been done. There may be a regular expression | |
270 | engine hit squad meeting at TPC5. | |
e50bb9a1 | 271 | |
722d2a37 | 272 | =head2 Memory leaks after failed eval |
e50bb9a1 | 273 | |
722d2a37 SC |
274 | Perl will leak memory if you C<eval "hlagh hlagh hlagh hlagh">. This is |
275 | partially because it attempts to build up an op tree for that code and | |
276 | doesn't properly free it. The same goes for non-syntactically-correct | |
277 | regular expressions. Hugo looked into this, but decided it needed a | |
278 | mark-and-sweep GC implementation. | |
e50bb9a1 | 279 | |
722d2a37 SC |
280 | Alan notes that: The basic idea was to extend the parser token stack |
281 | (C<YYSTYPE>) to include a type field so we knew what sort of thing each | |
210b36aa | 282 | element of the stack was. The F<perly.c> code would then have to be |
722d2a37 SC |
283 | postprocessed to record the type of each entry on the stack as it was |
284 | created, and the parser patched so that it could unroll the stack | |
285 | properly on error. | |
e50bb9a1 | 286 | |
722d2a37 SC |
287 | This is possible to do, but would be pretty messy to implement, as it |
288 | would rely on even more sed hackery in F<perly.fixer>. | |
e50bb9a1 | 289 | |
722d2a37 | 290 | =head2 pack "(stuff)*" |
e50bb9a1 | 291 | |
722d2a37 | 292 | That's to say, C<pack "(sI)40"> would be the same as C<pack "sI"x40> |
e50bb9a1 | 293 | |
722d2a37 | 294 | =head2 bitfields in pack |
e50bb9a1 | 295 | |
722d2a37 | 296 | =head2 Cross compilation |
e50bb9a1 | 297 | |
722d2a37 | 298 | Make Perl buildable with a cross-compiler. This will play havoc with |
da75cd15 | 299 | Configure, which needs to know how the target system will respond to |
722d2a37 SC |
300 | its tests; maybe C<microperl> will be a good starting point here. |
301 | (Indeed, Bart Schuller reports that he compiled up C<microperl> for | |
302 | the Agenda PDA and it works fine.) A really big spanner in the works | |
303 | is the bootstrapping build process of Perl: if the filesystem the | |
304 | target systems sees is not the same what the build host sees, various | |
305 | input, output, and (Perl) library files need to be copied back and forth. | |
e50bb9a1 | 306 | |
f86a8bc5 JH |
307 | As of 5.8.0 Configure mostly works for cross-compilation |
308 | (used successfully for iPAQ Linux), miniperl gets built, | |
309 | but then building DynaLoader (and other extensions) fails | |
310 | since MakeMaker knows nothing of cross-compilation. | |
311 | (See INSTALL/Cross-compilation for the state of things.) | |
312 | ||
722d2a37 | 313 | =head2 Perl preprocessor / macros |
e50bb9a1 | 314 | |
722d2a37 SC |
315 | Source filters help with this, but do not get us all the way. For |
316 | instance, it should be possible to implement the C<??> operator somehow; | |
317 | source filters don't (quite) cut it. | |
e50bb9a1 | 318 | |
722d2a37 | 319 | =head2 Perl lexer in Perl |
a45bd81d | 320 | |
722d2a37 | 321 | Damian Conway is planning to work on this, but it hasn't happened yet. |
e50bb9a1 | 322 | |
722d2a37 | 323 | =head2 Using POSIX calls internally |
e50bb9a1 | 324 | |
210b36aa | 325 | When faced with a BSD vs. SysV -style interface to some library or |
722d2a37 SC |
326 | system function, perl's roots show in that it typically prefers the BSD |
327 | interface (but falls back to the SysV one). One example is getpgrp(). | |
328 | Other examples include C<memcpy> vs. C<bcopy>. There are others, mostly in | |
210b36aa | 329 | F<pp_sys.c>. |
e50bb9a1 | 330 | |
722d2a37 SC |
331 | Mostly, this item is a suggestion for which way to start a journey into |
332 | an C<#ifdef> forest. It is not primarily a suggestion to eliminate any of | |
333 | the C<#ifdef> forests. | |
e50bb9a1 | 334 | |
722d2a37 SC |
335 | POSIX calls are perhaps more likely to be portable to unexpected |
336 | architectures. They are also perhaps more likely to be actively | |
337 | maintained by a current vendor. They are also perhaps more likely to be | |
338 | available in thread-safe versions, if appropriate. | |
e50bb9a1 | 339 | |
722d2a37 | 340 | =head2 -i rename file when changed |
e50bb9a1 | 341 | |
722d2a37 SC |
342 | It's only necessary to rename a file when inplace editing when the file |
343 | has changed. Detecting a change is perhaps the difficult bit. | |
e50bb9a1 | 344 | |
722d2a37 | 345 | =head2 All ARGV input should act like E<lt>E<gt> |
e50bb9a1 | 346 | |
2d84a16a DM |
347 | eg C<read(ARGV, ...)> doesn't currently read across multiple files. |
348 | ||
722d2a37 | 349 | =head2 Support for rerunning debugger |
e50bb9a1 | 350 | |
722d2a37 | 351 | There should be a way of restarting the debugger on demand. |
e50bb9a1 | 352 | |
c6287c21 JH |
353 | =head2 Test Suite for the Debugger |
354 | ||
355 | The debugger is a complex piece of software and fixing something | |
356 | here may inadvertently break something else over there. To tame | |
357 | this chaotic behaviour, a test suite is necessary. | |
358 | ||
722d2a37 | 359 | =head2 my sub foo { } |
c47ff5f1 | 360 | |
722d2a37 SC |
361 | The basic principle is sound, but there are problems with the semantics |
362 | of self-referential and mutually referential lexical subs: how to | |
363 | declare the subs? | |
c47ff5f1 | 364 | |
722d2a37 | 365 | =head2 One-pass global destruction |
c47ff5f1 | 366 | |
722d2a37 SC |
367 | Sweeping away all the allocated memory in one go is a laudable goal, but |
368 | it's difficult and in most cases, it's easier to let the memory get | |
369 | freed by exiting. | |
e50bb9a1 | 370 | |
722d2a37 | 371 | =head2 Rewrite regexp parser |
e50bb9a1 | 372 | |
722d2a37 SC |
373 | There has been talk recently of rewriting the regular expression parser |
374 | to produce an optree instead of a chain of opcodes; it's unclear whether | |
375 | or not this would be a win. | |
e50bb9a1 | 376 | |
722d2a37 | 377 | =head2 Cache recently used regexps |
e50bb9a1 | 378 | |
722d2a37 | 379 | This is to speed up |
e50bb9a1 | 380 | |
722d2a37 SC |
381 | for my $re (@regexps) { |
382 | $matched++ if /$re/ | |
383 | } | |
e50bb9a1 | 384 | |
722d2a37 SC |
385 | C<qr//> already gives us a way of saving compiled regexps, but it should |
386 | be done automatically. | |
e50bb9a1 | 387 | |
722d2a37 | 388 | =head2 Re-entrant functions |
e50bb9a1 | 389 | |
722d2a37 SC |
390 | Add configure probes for C<_r> forms of system calls and fit them to the |
391 | core. Unfortunately, calling conventions for these functions and not | |
392 | standardised. | |
04c70446 | 393 | |
722d2a37 | 394 | =head2 Cross-compilation support |
04c70446 | 395 | |
722d2a37 SC |
396 | Bart Schuller reports that using C<microperl> and a cross-compiler, he |
397 | got Perl working on the Agenda PDA. However, one cannot build a full | |
398 | Perl because Configure needs to get the results for the target platform, | |
399 | for the host. | |
e50bb9a1 | 400 | |
722d2a37 | 401 | =head2 Bit-shifting bitvectors |
e50bb9a1 | 402 | |
722d2a37 | 403 | Given: |
e50bb9a1 | 404 | |
722d2a37 | 405 | vec($v, 1000, 1) = 1; |
e50bb9a1 | 406 | |
722d2a37 | 407 | One should be able to do |
e50bb9a1 | 408 | |
722d2a37 | 409 | $v <<= 1; |
e50bb9a1 | 410 | |
722d2a37 | 411 | and have the 999'th bit set. |
e50bb9a1 | 412 | |
722d2a37 SC |
413 | Currently if you try with shift bitvectors you shift the NV/UV, instead |
414 | of the bits in the PV. Not very logical. | |
e50bb9a1 | 415 | |
722d2a37 | 416 | =head2 debugger pragma |
e50bb9a1 | 417 | |
722d2a37 SC |
418 | The debugger is implemented in Perl in F<perl5db.pl>; turning it into a |
419 | pragma should be easy, but making it work lexically might be more | |
420 | difficult. Fiddling with C<$^P> would be necessary. | |
e50bb9a1 | 421 | |
722d2a37 | 422 | =head2 use less pragma |
e50bb9a1 | 423 | |
722d2a37 SC |
424 | Identify areas where speed/memory tradeoffs can be made and have a hint |
425 | to switch between them. | |
e50bb9a1 | 426 | |
722d2a37 | 427 | =head2 switch structures |
e50bb9a1 | 428 | |
722d2a37 SC |
429 | Although we have C<Switch.pm> in core, Larry points to the dormant |
430 | C<nswitch> and C<cswitch> ops in F<pp.c>; using these opcodes would be | |
431 | much faster. | |
e50bb9a1 | 432 | |
722d2a37 | 433 | =head2 Cache eval tree |
e50bb9a1 | 434 | |
722d2a37 | 435 | =head2 rcatmaybe |
e50bb9a1 | 436 | |
722d2a37 | 437 | =head2 Shrink opcode tables |
e50bb9a1 | 438 | |
722d2a37 | 439 | =head2 Optimize away @_ |
e50bb9a1 | 440 | |
722d2a37 | 441 | Look at the "reification" code in C<av.c> |
e50bb9a1 | 442 | |
722d2a37 | 443 | =head2 Prototypes versus indirect objects |
e50bb9a1 | 444 | |
722d2a37 | 445 | Currently, indirect object syntax bypasses prototype checks. |
e50bb9a1 | 446 | |
210b36aa | 447 | =head2 Install HTML |
e50bb9a1 | 448 | |
722d2a37 SC |
449 | HTML versions of the documentation need to be installed by default; a |
450 | call to C<installhtml> from C<installperl> may be all that's necessary. | |
e50bb9a1 | 451 | |
722d2a37 | 452 | =head2 Prototype method calls |
e50bb9a1 | 453 | |
722d2a37 | 454 | =head2 Return context prototype declarations |
e50bb9a1 | 455 | |
722d2a37 | 456 | =head2 magic_setisa |
e50bb9a1 | 457 | |
722d2a37 | 458 | =head2 Garbage collection |
e50bb9a1 | 459 | |
722d2a37 SC |
460 | There have been persistent mumblings about putting a mark-and-sweep |
461 | garbage detector into Perl; Alan Burlison has some ideas about this. | |
e50bb9a1 | 462 | |
722d2a37 | 463 | =head2 IO tutorial |
e50bb9a1 | 464 | |
722d2a37 | 465 | Mark-Jason Dominus has the beginnings of one of these. |
e50bb9a1 | 466 | |
722d2a37 | 467 | =head2 pack/unpack tutorial |
e50bb9a1 | 468 | |
722d2a37 | 469 | Simon Cozens has the beginnings of one of these. |
e50bb9a1 | 470 | |
722d2a37 | 471 | =head2 Rewrite perldoc |
e50bb9a1 | 472 | |
722d2a37 SC |
473 | There are a few suggestions for what to do with C<perldoc>: maybe a |
474 | full-text search, an index function, locating pages on a particular | |
475 | high-level subject, and so on. | |
e50bb9a1 | 476 | |
3958b146 | 477 | =head2 Install .3p manpages |
e50bb9a1 | 478 | |
3958b146 | 479 | This is a bone of contention; we can create C<.3p> manpages for each |
722d2a37 SC |
480 | built-in function, but should we install them by default? Tcl does this, |
481 | and it clutters up C<apropos>. | |
e50bb9a1 | 482 | |
722d2a37 | 483 | =head2 Unicode tutorial |
e50bb9a1 | 484 | |
722d2a37 | 485 | Simon Cozens promises to do this before he gets old. |
e50bb9a1 | 486 | |
722d2a37 | 487 | =head2 Update POSIX.pm for 1003.1-2 |
3958b146 | 488 | |
722d2a37 | 489 | =head2 Retargetable installation |
e50bb9a1 | 490 | |
722d2a37 | 491 | Allow C<@INC> to be changed after Perl is built. |
e50bb9a1 | 492 | |
722d2a37 | 493 | =head2 POSIX emulation on non-POSIX systems |
e50bb9a1 | 494 | |
722d2a37 SC |
495 | Make C<POSIX.pm> behave as POSIXly as possible everywhere, meaning we |
496 | have to implement POSIX equivalents for some functions if necessary. | |
e50bb9a1 | 497 | |
722d2a37 | 498 | =head2 Rename Win32 headers |
e50bb9a1 | 499 | |
722d2a37 SC |
500 | =head2 Finish off lvalue functions |
501 | ||
502 | They don't work in the debugger, and they don't work for list or hash | |
503 | slices. | |
e50bb9a1 | 504 | |
722d2a37 | 505 | =head2 Update sprintf documentation |
e50bb9a1 | 506 | |
722d2a37 | 507 | Hugo van der Sanden plans to look at this. |
e50bb9a1 | 508 | |
722d2a37 | 509 | =head2 Use fchown/fchmod internally |
e50bb9a1 | 510 | |
722d2a37 SC |
511 | This has been done in places, but needs a thorough code review. |
512 | Also fchdir is available in some platforms. | |
e50bb9a1 | 513 | |
d45541b3 | 514 | =head2 Make v-strings overloaded objects |
c5fc23ff | 515 | |
d45541b3 JH |
516 | Instead of having to guess whether a string is a v-string and thus |
517 | needs to be displayed with %vd, make v-strings (readonly) objects | |
518 | (class "vstring"?) with a stringify overload. | |
c5fc23ff | 519 | |
722d2a37 | 520 | =head1 Vague ideas |
e50bb9a1 | 521 | |
722d2a37 | 522 | Ideas which have been discussed, and which may or may not happen. |
e50bb9a1 | 523 | |
722d2a37 | 524 | =head2 ref() in list context |
e50bb9a1 | 525 | |
722d2a37 SC |
526 | It's unclear what this should do or how to do it without breaking old |
527 | code. | |
e50bb9a1 | 528 | |
f86a8bc5 | 529 | =head2 Make tr/// return histogram of characters in list context |
e50bb9a1 | 530 | |
722d2a37 | 531 | There is a patch for this, but it may require Unicodification. |
e50bb9a1 | 532 | |
722d2a37 | 533 | =head2 Compile to real threaded code |
3958b146 | 534 | |
722d2a37 | 535 | =head2 Structured types |
3958b146 | 536 | |
722d2a37 | 537 | =head2 Modifiable $1 et al. |
e50bb9a1 | 538 | |
722d2a37 SC |
539 | ($x = "elephant") =~ /e(ph)/; |
540 | $1 = "g"; # $x = "elegant" | |
e50bb9a1 | 541 | |
722d2a37 SC |
542 | What happens if there are multiple (nested?) brackets? What if the |
543 | string changes between the match and the assignment? | |
e50bb9a1 | 544 | |
722d2a37 | 545 | =head2 Procedural interfaces for IO::*, etc. |
e50bb9a1 | 546 | |
722d2a37 SC |
547 | Some core modules have been accused of being overly-OO. Adding |
548 | procedural interfaces could demystify them. | |
e50bb9a1 | 549 | |
722d2a37 | 550 | =head2 RPC modules |
e50bb9a1 | 551 | |
722d2a37 | 552 | =head2 Attach/detach debugger from running program |
e50bb9a1 | 553 | |
722d2a37 SC |
554 | With C<gdb>, you can attach the debugger to a running program if you |
555 | pass the process ID. It would be good to do this with the Perl debugger | |
556 | on a running Perl program, although I'm not sure how it would be done. | |
e50bb9a1 | 557 | |
722d2a37 | 558 | =head2 Alternative RE syntax module |
e50bb9a1 | 559 | |
722d2a37 SC |
560 | use Regex::Newbie; |
561 | $re = Regex::Newbie->new | |
562 | ->start | |
563 | ->match("foo") | |
564 | ->repeat(Regex::Newbie->class("char"),3) | |
565 | ->end; | |
566 | /$re/; | |
e50bb9a1 | 567 | |
722d2a37 | 568 | =head2 GUI::Native |
e50bb9a1 | 569 | |
722d2a37 SC |
570 | A non-core module that would use "native" GUI to create graphical |
571 | applications. | |
e50bb9a1 | 572 | |
722d2a37 | 573 | =head2 foreach(reverse ...) |
e50bb9a1 | 574 | |
722d2a37 | 575 | Currently |
e50bb9a1 | 576 | |
722d2a37 | 577 | foreach (reverse @_) { ... } |
e50bb9a1 | 578 | |
722d2a37 SC |
579 | puts C<@_> on the stack, reverses it putting the reversed version on the |
580 | stack, then iterates forwards. Instead, it could be special-cased to put | |
581 | C<@_> on the stack then iterate backwards. | |
e50bb9a1 | 582 | |
722d2a37 | 583 | =head2 Constant function cache |
e50bb9a1 | 584 | |
722d2a37 | 585 | =head2 Approximate regular expression matching |
e50bb9a1 | 586 | |
722d2a37 | 587 | =head1 Ongoing |
e50bb9a1 | 588 | |
722d2a37 | 589 | These items B<always> need doing: |
e50bb9a1 | 590 | |
722d2a37 | 591 | =head2 Update guts documentation |
e50bb9a1 | 592 | |
722d2a37 SC |
593 | Simon Cozens tries to do this when possible, and contributions to the |
594 | C<perlapi> documentation is welcome. | |
e50bb9a1 | 595 | |
722d2a37 | 596 | =head2 Add more tests |
e50bb9a1 | 597 | |
722d2a37 SC |
598 | Michael Schwern will donate $500 to Yet Another Society when all core |
599 | modules have tests. | |
e50bb9a1 | 600 | |
722d2a37 | 601 | =head2 Update auxiliary tools |
e50bb9a1 | 602 | |
722d2a37 | 603 | The code we ship with Perl should look like good Perl 5. |
e50bb9a1 | 604 | |
722d2a37 | 605 | =head1 Recently done things |
e50bb9a1 | 606 | |
722d2a37 SC |
607 | These are things which have been on the todo lists in previous releases |
608 | but have recently been completed. | |
e50bb9a1 | 609 | |
722d2a37 | 610 | =head2 Safe signal handling |
e50bb9a1 | 611 | |
722d2a37 SC |
612 | A new signal model went into 5.7.1 without much fanfare. Operations and |
613 | C<malloc>s are no longer interrupted by signals, which are handled | |
614 | between opcodes. This means that C<PERL_ASYNC_CHECK> now actually does | |
615 | something. However, there are still a few things that need to be done. | |
e50bb9a1 | 616 | |
722d2a37 | 617 | =head2 Tie Modules |
e50bb9a1 | 618 | |
722d2a37 SC |
619 | Modules which implement arrays in terms of strings, substrings or files |
620 | can be found on the CPAN. | |
e50bb9a1 | 621 | |
722d2a37 | 622 | =head2 gettimeofday |
e50bb9a1 | 623 | |
210b36aa | 624 | C<Time::HiRes> has been integrated into the core. |
e50bb9a1 | 625 | |
722d2a37 | 626 | =head2 setitimer and getimiter |
e50bb9a1 | 627 | |
210b36aa | 628 | Adding C<Time::HiRes> got us this too. |
e50bb9a1 | 629 | |
722d2a37 SC |
630 | =head2 Testing __DIE__ hook |
631 | ||
632 | Tests have been added. | |
633 | ||
634 | =head2 CPP equivalent in Perl | |
e50bb9a1 | 635 | |
722d2a37 SC |
636 | A C Yardley will probably have done this by the time you can read this. |
637 | This allows for a generalization of the C constant detection used in | |
638 | building C<Errno.pm>. | |
e50bb9a1 | 639 | |
722d2a37 | 640 | =head2 Explicit switch statements |
e50bb9a1 | 641 | |
722d2a37 SC |
642 | C<Switch.pm> has been integrated into the core to give you all manner of |
643 | C<switch...case> semantics. | |
e50bb9a1 | 644 | |
722d2a37 | 645 | =head2 autocroak |
e50bb9a1 | 646 | |
722d2a37 | 647 | This is C<Fatal.pm>. |
e50bb9a1 | 648 | |
722d2a37 | 649 | =head2 UTF/EBCDIC |
e50bb9a1 | 650 | |
722d2a37 | 651 | Nick Ing-Simmons has made UTF-EBCDIC (UTR13) work with Perl. |
e50bb9a1 | 652 | |
722d2a37 | 653 | EBCDIC? http://www.unicode.org/unicode/reports/tr16/ |
e50bb9a1 | 654 | |
722d2a37 | 655 | =head2 UTF Regexes |
e50bb9a1 | 656 | |
722d2a37 SC |
657 | Although there are probably some small bugs to be rooted out, Jarkko |
658 | Hietaniemi has made regular expressions polymorphic between bytes and | |
659 | characters. | |
e50bb9a1 | 660 | |
722d2a37 | 661 | =head2 perlcc to produce executable |
e50bb9a1 | 662 | |
722d2a37 SC |
663 | C<perlcc> was recently rewritten, and can now produce standalone |
664 | executables. | |
e50bb9a1 | 665 | |
722d2a37 | 666 | =head2 END blocks saved in compiled output |
e50bb9a1 | 667 | |
722d2a37 | 668 | =head2 Secure temporary file module |
e50bb9a1 | 669 | |
722d2a37 | 670 | Tim Jenness' C<File::Temp> is now in core. |
e50bb9a1 | 671 | |
722d2a37 | 672 | =head2 Integrate Time::HiRes |
e50bb9a1 | 673 | |
722d2a37 | 674 | This module is now part of core. |
e50bb9a1 | 675 | |
722d2a37 | 676 | =head2 Turn Cwd into XS |
e50bb9a1 | 677 | |
722d2a37 | 678 | Benjamin Sugars has done this. |
e50bb9a1 | 679 | |
722d2a37 | 680 | =head2 Mmap for input |
e50bb9a1 | 681 | |
722d2a37 | 682 | Nick Ing-Simmons' C<perlio> supports an C<mmap> IO method. |
e50bb9a1 | 683 | |
722d2a37 | 684 | =head2 Byte to/from UTF8 and UTF8 to/from local conversion |
e50bb9a1 | 685 | |
722d2a37 | 686 | C<Encode> provides this. |
e50bb9a1 | 687 | |
722d2a37 | 688 | =head2 Add sockatmark support |
e50bb9a1 | 689 | |
722d2a37 | 690 | Added in 5.7.1 |
e50bb9a1 | 691 | |
722d2a37 SC |
692 | =head2 Mailing list archives |
693 | ||
694 | http://lists.perl.org/, http://archive.develooper.com/ | |
695 | ||
696 | =head2 Bug tracking | |
697 | ||
698 | Richard Foley has written the bug tracking system at http://bugs.perl.org/ | |
e50bb9a1 | 699 | |
722d2a37 | 700 | =head2 Integrate MacPerl |
e50bb9a1 | 701 | |
722d2a37 SC |
702 | Chris Nandor and Matthias Neeracher have integrated the MacPerl changes |
703 | into 5.6.0. | |
e50bb9a1 | 704 | |
722d2a37 | 705 | =head2 Web "nerve center" for Perl |
e50bb9a1 | 706 | |
722d2a37 | 707 | http://use.perl.org/ is what you're looking for. |
e50bb9a1 | 708 | |
722d2a37 | 709 | =head2 Regular expression tutorial |
e50bb9a1 | 710 | |
722d2a37 | 711 | C<perlretut>, provided by Mark Kvale. |
e50bb9a1 | 712 | |
722d2a37 | 713 | =head2 Debugging Tutorial |
e50bb9a1 | 714 | |
722d2a37 | 715 | C<perldebtut>, written by Richard Foley. |
e50bb9a1 | 716 | |
722d2a37 | 717 | =head2 Integrate new modules |
e50bb9a1 | 718 | |
722d2a37 | 719 | Jarkko has been integrating madly into 5.7.x |
e50bb9a1 | 720 | |
722d2a37 | 721 | =head2 Integrate profiler |
e50bb9a1 | 722 | |
722d2a37 | 723 | C<Devel::DProf> is now a core module. |
e50bb9a1 | 724 | |
722d2a37 | 725 | =head2 Y2K error detection |
e50bb9a1 | 726 | |
722d2a37 SC |
727 | There's a configure option to detect unsafe concatenation with "19", and |
728 | a CPAN module. (C<D'oh::Year>) | |
e50bb9a1 | 729 | |
722d2a37 | 730 | =head2 Regular expression debugger |
e50bb9a1 | 731 | |
722d2a37 SC |
732 | While not part of core, Mark-Jason Dominus has written C<Rx> and has |
733 | also come up with a generalised strategy for regular expression | |
734 | debugging. | |
e50bb9a1 | 735 | |
722d2a37 | 736 | =head2 POD checker |
e50bb9a1 | 737 | |
722d2a37 | 738 | That's, uh, F<podchecker> |
e50bb9a1 | 739 | |
722d2a37 | 740 | =head2 "Dynamic" lexicals |
e50bb9a1 | 741 | |
722d2a37 | 742 | =head2 Cache precompiled modules |
e50bb9a1 | 743 | |
722d2a37 | 744 | =head1 Deprecated Wishes |
e50bb9a1 | 745 | |
722d2a37 SC |
746 | These are items which used to be in the todo file, but have been |
747 | deprecated for some reason. | |
e50bb9a1 | 748 | |
722d2a37 | 749 | =head2 Loop control on do{} |
e50bb9a1 | 750 | |
722d2a37 | 751 | This would break old code; use C<do{{ }}> instead. |
e50bb9a1 | 752 | |
722d2a37 | 753 | =head2 Lexically scoped typeglobs |
e50bb9a1 | 754 | |
722d2a37 | 755 | Not needed now we have lexical IO handles. |
e50bb9a1 | 756 | |
722d2a37 | 757 | =head2 format BOTTOM |
3958b146 | 758 | |
722d2a37 | 759 | =head2 report HANDLE |
e50bb9a1 | 760 | |
722d2a37 | 761 | Damian Conway's text formatting modules seem to be the Way To Go. |
e50bb9a1 | 762 | |
722d2a37 | 763 | =head2 Generalised want()/caller()) |
3958b146 | 764 | |
722d2a37 | 765 | =head2 Named prototypes |
e50bb9a1 | 766 | |
722d2a37 | 767 | These both seem to be delayed until Perl 6. |
e50bb9a1 | 768 | |
722d2a37 | 769 | =head2 Built-in globbing |
e50bb9a1 | 770 | |
722d2a37 | 771 | The C<File::Glob> module has been used to replace the C<glob> function. |
e50bb9a1 | 772 | |
722d2a37 | 773 | =head2 Regression tests for suidperl |
e50bb9a1 | 774 | |
722d2a37 | 775 | C<suidperl> is deprecated in favour of common sense. |
e50bb9a1 | 776 | |
722d2a37 | 777 | =head2 Cached hash values |
e50bb9a1 | 778 | |
722d2a37 | 779 | We have shared hash keys, which perform the same job. |
e50bb9a1 | 780 | |
722d2a37 | 781 | =head2 Add compression modules |
e50bb9a1 | 782 | |
722d2a37 SC |
783 | The compression modules are a little heavy; meanwhile, Nick Clark is |
784 | working on experimental pragmata to do transparent decompression on | |
785 | input. | |
e50bb9a1 | 786 | |
722d2a37 | 787 | =head2 Reorganise documentation into tutorials/references |
e50bb9a1 | 788 | |
722d2a37 | 789 | Could not get consensus on P5P about this. |
e50bb9a1 | 790 | |
722d2a37 SC |
791 | =head2 Remove distinction between functions and operators |
792 | ||
793 | Caution: highly flammable. | |
794 | ||
795 | =head2 Make XS easier to use | |
e50bb9a1 | 796 | |
722d2a37 | 797 | Use C<Inline> instead, or SWIG. |
e50bb9a1 | 798 | |
722d2a37 | 799 | =head2 Make embedding easier to use |
e50bb9a1 | 800 | |
722d2a37 | 801 | Use C<Inline::CPR>. |
e50bb9a1 | 802 | |
722d2a37 | 803 | =head2 man for perl |
04c70446 | 804 | |
722d2a37 | 805 | See the Perl Power Tools. (http://language.perl.com/ppt/) |
04c70446 | 806 | |
722d2a37 | 807 | =head2 my $Package::variable |
04c70446 | 808 | |
722d2a37 | 809 | Use C<our> instead. |
04c70446 | 810 | |
722d2a37 | 811 | =head2 "or" tests defined, not truth |
04c70446 | 812 | |
722d2a37 | 813 | Suggesting this on P5P B<will> cause a boring and interminable flamewar. |
04c70446 | 814 | |
722d2a37 | 815 | =head2 "class"-based lexicals |
04c70446 | 816 | |
cbb3fa72 | 817 | Use flyweight objects, secure hashes or, dare I say it, pseudo-hashes instead. |
f86a8bc5 | 818 | (Or whatever will replace pseudohashes in 5.10.) |
04c70446 | 819 | |
722d2a37 | 820 | =head2 byteperl |
04c70446 | 821 | |
722d2a37 | 822 | C<ByteLoader> covers this. |
04c70446 | 823 | |
722d2a37 | 824 | =head2 Lazy evaluation / tail recursion removal |
04c70446 | 825 | |
f86a8bc5 JH |
826 | C<List::Util> gives first() (a short-circuiting grep); tail recursion |
827 | removal is done manually, with C<goto &whoami;>. (However, MJD has | |
828 | found that C<goto &whoami> introduces a performance penalty, so maybe | |
829 | there should be a way to do this after all: C<sub foo {START: ... goto | |
830 | START;> is better.) | |
0562c0e3 JH |
831 | |
832 | =head2 Make "use utf8" the default | |
833 | ||
f86a8bc5 JH |
834 | Because of backward compatibility this is difficult: scripts could not |
835 | contain B<any legacy eight-bit data> (like Latin-1) anymore, even in | |
836 | string literals or pod. Also would introduce a measurable slowdown of | |
837 | at least few percentages since all regular expression operations would | |
838 | be done in full UTF-8. But if you want to try this, add | |
839 | -DUSE_UTF8_SCRIPTS to your compilation flags. | |
840 | ||
3298bd4d JH |
841 | =head2 Unicode collation and normalization |
842 | ||
843 | The Unicode::Collate and Unicode::Normalize modules | |
844 | by SADAHIRO Tomoyuki have been included since 5.8.0. | |
845 | ||
846 | Collation? http://www.unicode.org/unicode/reports/tr10/ | |
847 | Normalization? http://www.unicode.org/unicode/reports/tr15/ | |
0562c0e3 | 848 | |
825b3abc RS |
849 | =head2 Create debugging macros |
850 | ||
851 | Debugging macros (like printsv, dump) can make debugging perl inside a | |
852 | C debugger much easier. A good set for gdb comes with mod_perl. | |
853 | Something similar should be distributed with perl. | |
854 | ||
855 | The proper way to do this is to use and extend Devel::DebugInit. | |
856 | Devel::DebugInit also needs to be extended to support threads. | |
857 | ||
858 | See p5p archives for late May/early June 2001 for a recent discussion | |
859 | on this topic. | |
860 | ||
3298bd4d | 861 | =cut |