Camelia, the Perl 6 bug

IRC log for #parrot, 2008-06-21

Parrot | source cross referenced

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 toddr joined #parrot
00:07 Whiteknight joined #parrot
00:09 AndyA joined #parrot
00:10 davidfetter joined #parrot
00:18 japhb joined #parrot
00:27 * DietCoke works on binary search, version 0.2
00:27 DietCoke sadly, it's blocking on a bug in Iterator atm. =-)
00:48 Whiteknight joined #parrot
01:09 Auzon joined #parrot
01:38 Whiteknight joined #parrot
01:47 Zaba_ joined #parrot
02:09 kid51 joined #parrot
02:18 cotto_yapc joined #parrot
02:19 DietCoke grumble. Perl::Critic latest doesn't like how we're using it.
02:31 tetragon joined #parrot
02:42 Zaba joined #parrot
02:44 kid51 joined #parrot
02:44 kid51 left #parrot
02:44 kid51 joined #parrot
02:44 kid51 DietCoke:  Yes.  I just upgraded and got a message to the effect that there were no standards in force.
02:48 nopaste "kid51" at 71.247.46.29 pasted "Perl::Critic our $VERSION = '1.086': "There are no enabled policies"" (14 lines) at http://nopaste.snit.ch/13332
02:51 Andy joined #parrot
02:52 kid51 message => 'There are no enabled policies.',
02:52 purl Sorry, I've never seen => before.
02:54 DietCoke yup. sent a list to the P::C mailing list with a diagnosis and asking what they'd recommend.
02:54 DietCoke I suspect "completely redo your test this way" will be the answer.
02:54 DietCoke just hoping they can save me some effort on the rewrite.
02:55 DietCoke gnite, jim
02:55 DietCoke -> zzz
02:55 dalek r28588 | pmichaud++ | trunk:
02:55 dalek : [rakudo]:
02:55 dalek : * Update tools/test_summary.pl to summarize skipped tests.
02:55 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28588
03:11 dalek r28589 | chromatic++ | trunk:
03:11 dalek : [IMCC] Cleaned up SymReg handling in IMCC, partially:
03:11 dalek :     - removed some unused and duplicate functions
03:11 dalek :     - tidied the code
03:11 dalek :     - added some function documentation
03:11 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28589
03:22 dalek r28590 | jkeenan++ | trunk:
03:22 dalek : Number of tests in SKIP block should have been 3, not 2.
03:22 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28590
03:23 bacek rakudo: my ($a, $b); $a=42; say $a
03:24 bacek wrong window.
03:25 bacek Scope  not found for PAST::Var '$a'
03:25 Tene bacek: that works here with the name 'perl6:'
03:25 bacek Tene, thanks
03:25 bacek perl6: my ($a, $b); $a=42; say $a
03:25 polyglotbot OUTPUT[Scope  not found for PAST::Var '$a'␤current instr.: 'parrot;PCT::HLLCompiler;panic' pc 156 (src/PCT/HLLCompiler.pir:103)␤called from Sub 'parrot;PAST::Compiler;post_children' pc 1692 (src/PAST/Compiler.pir:364)␤called from Sub 'parrot;PAST::Compiler;call' pc 3067
03:25 polyglotbot ..(src/PAST/Compiler.pir:823)␤called from Sub 'parrot;PAST::Compiler;post_chil...
03:25 bacek (infix:,)--
03:26 Tene There's also perl6.parse:, perl6.past:, perl6.pir:, and perl6.paste:
03:27 bacek perl6.paste?
03:27 Tene sends output ot pastebin instead of here.
03:28 Tene perl6.paste: for 1..1000 { say $_ }
03:28 nopaste "polyglotbot" at 193.200.132.146 pasted "perl6 paste" (1000 lines) at http://nopaste.snit.ch/13333
03:28 bacek Tene, ok. Thanks again :)
03:28 bacek perl6.past: my ($a, $b);
03:28 nopaste "polyglotbot" at 193.200.132.146 pasted "perl6 past paste" (64 lines) at http://nopaste.snit.ch/13334
03:37 japhb Tene: Can you make aliases for the language names that polyglotbot recognizes?  So rakudo := perl6, plumhead := php, and so on?
03:37 japhb That will tend to DWIM instead of have people wondering which is the right thing to use ...
03:38 japhb Any cygwin users around?
03:38 Tene Right now it just uses the names in languages/ because that reduced my typing.  I doubt I'll actually get around to fixing that in the near future, but it's on feather3 if you want to change it yourself.
03:38 * japhb once had feather access, but cannot for the life of him remember username/password there
03:39 japhb It's been a year and a half since I used it, I think.
03:39 diakopter japhb: I can reset your password on feather(1), but not feather3
03:39 Tene Just harass Juerd to set up an account for you on feather3.
03:40 japhb diakopter: yes, would you mind resetting me?  I may be japhb or geoffb or just geoff, I don't remember.
03:40 dalek r28591 | chromatic++ | trunk:
03:40 dalek : [src] Fixed a memory leak; I forgot to free the const string cache previously.
03:40 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28591
03:41 japhb I need to get around to putting a password safe on a thumb drive somewhere ....
03:41 diakopter how about japhy
03:41 purl japhy is engaged.
03:41 japhb diakopter: nope, that's not me.  The one variant that isn't.
03:42 diakopter nope, geoffb
03:42 diakopter yeah
03:48 japhb Juerd: Can I get my feather account copied to feather3, so that I can help Tene with polyglotbot?
03:49 diakopter Tene has sudo access on feather3; he can create an acct for you there :P
03:49 Tene Oh, I have sudo, I could give you... yeah, that.
03:49 Tene I'll do that now.
03:49 japhb heh
03:49 japhb thx
03:50 Tene what username?
03:50 purl username is the first entry?
03:50 japhb Tene: how about 'japhb'
03:50 japhb ?
03:51 Tene Juerd: I just set up an account for japhb and gave him sudo access.  Tell me if that was inappropriate.
03:53 Tene japhb: it's in my home directory, currently running inside a screen session.
03:53 japhb OK, fair enough
03:53 * japhb is a big fan of screen
03:55 japhb oh, that's interesting
03:55 Tene hm?
03:55 japhb I can't take your screen, because you don't have permission to affect my tty ...
03:55 Tene yeah
03:55 Tene just chmod your tty
03:55 Tene or setfacl it, if that filesystem has acls
03:56 Tene ah, no setfacl.
03:56 japhb It's an old debian box, I doubt it's running ACLs unless Juerd did it manually
03:56 Tene yeah, looks like
03:56 Tene chmod to the rescue!
03:56 Andy joined #parrot
04:00 * japhb going crazy without 'less'
04:03 japhb Word to the wise -- do not use 'more' on pbot's TODO while inside a screen ... ==> locked window, ctrl-A K required
04:05 bacek perl6: my $a; my $a; $a=42; say $a;
04:05 polyglotbot OUTPUT[42␤]
04:05 bacek yak...
04:05 tetragon Yay, the Unicode newline character
04:07 japhb Dunno why, but I've never been able to see that one.  I'm mildly curious which Debian font has it, if anyone knows ...
04:08 tetragon I'm not certain which fonts in Debian have that range, but one of the fonts on my box with SYMBOL FOR NEWLINE is "Arial Unicode MS"
04:08 japhb Do you have it in any fixed-width (and preferably programmer-friendly) fonts?
04:08 dalek Christoph Otto | Plumhead:
04:08 dalek link: http://www.perlfoundation.or​g/parrot/index.cgi?plumhead
04:09 tetragon I'm not certain about how well Monaco would work
04:11 tetragon And then there's also the ugly "Ming(for ISO10646)" that I was using for xelatex/Template::Latex trials
04:11 bacek japhb, 'Monospace'
04:11 japhb bacek, doesn't seem to work for me
04:12 tetragon Is your client set up for Unicode?
04:12 bacek japhb, strange. Works for me in xchat.
04:12 japhb tetragon: It's xchat-gnome.  Should work.
04:12 japhb bacek: hmmm, I wonder if Monospace is just a symlink ...
04:13 tetragon japhb: With xchat aqua I had to set a preference before it would
04:13 bacek $ fc-match Monospace
04:13 bacek Courier_New.ttf: "Courier New" "Normal"
04:13 bacek heh
04:13 tetragon japhb: I had to set the server character set
04:13 japhb VeraMono.ttf: "Bitstream Vera Sans Mono" "Roman"
04:14 japhb ... and that would be the problem, I guess
04:14 japhb tetragon: hmmm
04:14 tetragon japhb: The Aqua port at the very least doesn't default to UTF-8
04:14 japhb tetragon: nod.  I'm looking around for something obvious
04:15 dalek Christoph Otto | plumhead_renaming:
04:15 dalek link: http://www.perlfoundation.org/pa​rrot/index.cgi?plumhead_renaming
04:15 shorten dalek's url is at http://xrl.us/d2gpk
04:16 japhb tetragon: found it; it was already set to UTF-8
04:17 cotto_yapc msg barney Take a look at http://www.perlfoundation.org/pa​rrot/index.cgi?plumhead_renaming when you have a minute or two.
04:17 purl Message for barney stored.
04:17 shorten cotto_yapc's url is at http://xrl.us/d2gpk
04:17 cotto_yapc msg fperrad Take a look at http://www.perlfoundation.org/pa​rrot/index.cgi?plumhead_renaming when you have a minute or two.
04:17 purl Sorry, I've never seen fperrad before.
04:17 shorten cotto_yapc's url is at http://xrl.us/d2gpk
04:17 cotto_yapc yes you have
04:23 japhb joined #parrot
04:27 japhb
04:31 Zaba_ joined #parrot
04:40 bacek perl6: my ($a,$b); my(1,2);
04:40 polyglotbot OUTPUT[Scope  not found for PAST::Var '$a'␤current instr.: 'parrot;PCT::HLLCompiler;panic' pc 156 (src/PCT/HLLCompiler.pir:103)␤called from Sub 'parrot;PAST::Compiler;post_children' pc 1692 (src/PAST/Compiler.pir:364)␤called from Sub 'parrot;PAST::Compiler;call' pc 3067
04:40 polyglotbot ..(src/PAST/Compiler.pir:823)␤called from Sub 'parrot;PAST::Compiler;post_chil...
04:41 bacek perl6: my ($a,$b); say 'hey'
04:41 polyglotbot OUTPUT[Scope  not found for PAST::Var '$a'␤current instr.: 'parrot;PCT::HLLCompiler;panic' pc 156 (src/PCT/HLLCompiler.pir:103)␤called from Sub 'parrot;PAST::Compiler;post_children' pc 1692 (src/PAST/Compiler.pir:364)␤called from Sub 'parrot;PAST::Compiler;call' pc 3067
04:41 polyglotbot ..(src/PAST/Compiler.pir:823)␤called from Sub 'parrot;PAST::Compiler;post_chil...
04:42 mateo319 joined #parrot
05:02 Eevee er: make: *** No rule to make target `src/pmc/version.pmc', needed by `src/pmc/version.dump'.  Stop.
05:09 Psyche^ joined #parrot
05:16 Eevee okay, make didn't care on attempt 2
05:16 Eevee but this is interesting
05:16 Eevee perl6: my $a = 'foo'; $a.say;
05:16 polyglotbot OUTPUT[foo␤]
05:17 Eevee perl6: $_ = 'foo'; $_.say;
05:17 polyglotbot OUTPUT[too few arguments passed (0) - 1 params expected␤current instr.: '_block11' pc 0 (EVAL_7:3)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:469)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1088 (src/PCT/HLLCompiler.pir:598)␤called from Sub
05:17 polyglotbot ..'parrot;PCT::HLLCompiler;command_line' pc 1267 (src/PCT/HL...
05:18 Eevee $_ only work within loops so far?
05:19 bacek perl6: say ~(1..10).grep({$_ % 2})
05:19 polyglotbot OUTPUT[Method 'grep' not found for invocant of class 'Range'␤current instr.: '_block11' pc 63 (EVAL_14:25)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:469)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1088 (src/PCT/HLLCompiler.pir:598)␤called from Sub
05:19 polyglotbot ..'parrot;PCT::HLLCompiler;command_line' pc 1267 (sr...
05:19 bacek perl6: say ~<1..10>.grep({$_ % 2})
05:19 polyglotbot OUTPUT[Method 'grep' not found for invocant of class 'Perl6Str'␤current instr.: '_block11' pc 45 (EVAL_12:22)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:469)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1088 (src/PCT/HLLCompiler.pir:598)␤called from Sub
05:19 polyglotbot ..'parrot;PCT::HLLCompiler;command_line' pc 1267 ...
05:20 bacek perl6: say ~(<1..10>).grep({$_ % 2})
05:20 polyglotbot OUTPUT[Method 'grep' not found for invocant of class 'Perl6Str'␤current instr.: '_block11' pc 45 (EVAL_12:22)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:469)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1088 (src/PCT/HLLCompiler.pir:598)␤called from Sub
05:20 polyglotbot ..'parrot;PCT::HLLCompiler;command_line' pc 1267 ...
05:20 bacek perl6: say ~grep {$_ % 2}, 1..10;
05:20 polyglotbot OUTPUT[1 3 5 7 9␤]
05:20 bacek but it is not '$_'
05:21 Eevee is it faked as $^_?
05:21 Eevee wow that was some line noise
05:25 bacek perl6: say ~grep {$^_ % 2}, 1..10;
05:25 polyglotbot OUTPUT[1 3 5 7 9␤]
05:25 bacek Eevee, looks like...
05:26 Eevee haha, that's so gross
05:26 Eevee okay
06:04 Zaba joined #parrot
06:05 zarchne joined #parrot
06:13 japhb Juerd: ping
06:32 japhb msg Juerd Do you have any objections to apt installs on feather3 -- like, for instance, 'less'?
06:32 purl Message for juerd stored.
07:17 ank joined #parrot
07:22 iblechbot joined #parrot
07:25 Ademan joined #parrot
08:01 barney joined #parrot
08:03 slightlyoff joined #parrot
08:04 slightlyoff left #parrot
08:05 Zaba_ joined #parrot
08:11 dalek bernhard.schmalhofer@gmx.de | Plumhead:
08:11 dalek link: http://www.perlfoundation.or​g/parrot/index.cgi?plumhead
08:12 dalek bernhard.schmalhofer@gmx.de | Plumhead:
08:12 dalek link: http://www.perlfoundation.or​g/parrot/index.cgi?plumhead
08:25 dalek r28592 | bernhard++ | trunk:
08:25 dalek : [Plumhead]
08:25 dalek : Clean up generated *.pbc files.
08:26 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28592
08:31 bacek joined #parrot
08:36 masak joined #parrot
08:46 dalek r28593 | bernhard++ | trunk:
08:46 dalek : Satisfy the c_parens codingstd test.
08:46 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28593
09:20 pjcj joined #parrot
09:31 ruoso joined #parrot
09:33 Zaba joined #parrot
10:21 Whiteknight joined #parrot
11:01 iblechbot joined #parrot
12:04 jq joined #parrot
12:14 tetragon joined #parrot
12:34 jjuran joined #parrot
12:45 Zaba joined #parrot
12:49 Juerd japhb: No, please go ahead. feather3 is free play as long as you fix what you break :)
12:53 kid51 joined #parrot
12:56 radhios joined #parrot
13:13 ambs joined #parrot
13:14 ambs Given that TimToady is around, let me just suggest to add a new constrction to perl 6:   forever { ... }   :D
13:15 pmichaud we already have such a construction:  loop
13:15 ambs oops
13:15 ambs loop { .. }
13:15 ambs ?
13:16 pmichaud yes,   loop { ... }
13:16 pmichaud see S04.
13:16 ambs damn, everytime I get an idea, somebody had it already.
13:16 pmichaud from S04:  As in C, the parentheses are required if you supply the 3-part spec; however, the 3-part loop spec may be entirely omitted to write an infinite loop. That is,
13:16 pmichaud loop {...}
13:16 pmichaud is equivalent to the Cish idiom:
13:16 pmichaud loop (;;) {...}
13:16 ambs then, for is dead?
13:16 pmichaud no, for is an iterator.
13:16 pmichaud for @list { ... }
13:17 ambs ok, C for is now loop
13:17 ambs Perl for, is for
13:17 ambs got it
13:18 * ambs needs to start reading S*
13:45 masak ambs: not only that, but p5 `while` is sometimes `for` in p6
13:45 masak specifically, when you're reading lines from file
13:54 ambs so, basically, a completely new language
13:54 ambs :)
13:54 ruoso joined #parrot
13:55 diakopter ambs: not backward-compatible, if that's what you mean.  there are a lot of similarities with Perl 5
13:58 ambs diakopter, no. I mean it will take some time to be self-confident writing p6 code.
14:04 masak ambs: well, yes or no, depending on what 'completely new' means to you
14:04 masak but the file-line idiom will have to be relearnt
14:05 masak it's true that p5 and p6 are very different in some respects
14:05 * ambs just found http://feather.perl6.nl/syn/
14:05 masak the sense of Perl is definitely still in p6, tho'
14:06 ambs :)
14:06 tetragon joined #parrot
14:09 radhios joined #parrot
14:12 dalek r28594 | pmichaud++ | trunk:
14:12 dalek : [rakudo]:
14:12 dalek : * Update spectest-progress
14:12 dalek : * 65 files, 794 passing tests
14:12 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28594
14:14 jonathan Ooh, not too many off 1000. :-)
14:15 pmichaud yes, when I get the <term> rule refactored we'll get a bunch of tests passing from enabling 'pi'
14:16 pmichaud src/ops needs to have svn:ignore set.
14:16 gmansi joined #parrot
14:17 jonathan Nice...what else is on your coming soon list?
14:17 pmichaud $_ handling, <term> rule refactor, and hopefully pre-compiled Test.pm
14:17 dalek r28595 | pmichaud++ | trunk:
14:17 dalek : [rakudo]:
14:17 dalek : * Update README.
14:17 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28595
14:17 pmichaud (getting Test.pm to work requires a bit of re-think of :load :init subs)
14:19 jonathan Sounds good.
14:19 jonathan Pre-compiled stuff would be great.
14:19 pmichaud yes, it will speed up the testing quite a bit.
14:19 pmichaud then we'd also be able to write runtime libraries in Perl 6 :-)
14:20 jonathan Will you try and do it for general stuff too, so we can have a perl6 --bytecode myscript.p6 and get myscript.pbc?
14:20 pmichaud I don't know of a way to generate bytecode from within PIR.
14:20 pmichaud but one could do  --target=pir and then compile that result to .pbc
14:20 mateo319 joined #parrot
14:20 jonathan You take an eval PMC
14:20 jonathan And I think
14:20 jonathan $S0 = $P0 where $P0 is the eval PMC
14:21 jonathan Then write $S0 to file.
14:21 pmichaud hrm.
14:21 pmichaud hadn't heard that one.  In previous times when I've asked about this the answer came back as "oh, we don't have a way to do that yet."  :-)
14:21 jonathan I'm sure I've done that in the past.
14:21 pmichaud if that works I should add it to HLL compiler.
14:23 jonathan oh, it's not quite like that
14:23 jonathan But see languages/dotnet/src/net2pbc.pir
14:24 nopaste "pmichaud" at 76.183.97.54 pasted "problem with lexical scoping" (15 lines) at http://nopaste.snit.ch/13336
14:26 jonathan That's bizzare...
14:26 pmichaud I think it has to do with closure handling.  I'm coming up with a small .pir demo case.
14:30 nopaste "pmichaud" at 76.183.97.54 pasted "problem with lexical scoping (pir)" (31 lines) at http://nopaste.snit.ch/13337
14:44 iblechbot joined #parrot
14:53 Limbic_Region joined #parrot
14:55 dalek r28596 | pmichaud++ | trunk:
14:55 dalek : [rakudo]:
14:55 dalek : * Set svn:ignore properties on src/ops .
14:55 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28596
14:56 dalek r28597 | pmichaud++ | trunk:
14:56 dalek : [rakudo]:
14:56 dalek : * Handle empty statements (RT#55594)
14:56 dalek : * Patch courtesy Ronald Schmidt <ronaldxs@software-path.com>
14:56 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28597
15:04 mateo319 joined #parrot
15:05 dalek r28598 | pmichaud++ | trunk:
15:05 dalek : [rakudo]:
15:06 dalek : * Allow 'make spectest_regression' to work even when "svn up" fails on t/spec.
15:06 dalek : * RT#56050 patch courtesy Deven T. Corzine <deven@twist.ties.org>
15:06 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28598
15:17 dalek r28599 | pmichaud++ | trunk:
15:17 dalek : [rakudo]:
15:17 dalek : * Add chop.t and remove comb.t from spectest_regression (RT#56068, moritz++)
15:17 dalek : * Patch courtesy Moritz Lenz <moritz@casella.verplant.org>
15:17 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28599
15:22 DietCoke removing things from regression doesn't soundgood!
15:24 pmichaud we were skipping all of the tests.
15:25 jonathan r28598++
15:25 jonathan (I'm only kinda half here at the moment...)
15:25 pmichaud I'm about to be not-here -- have a LUG to attend in 30 mins.
15:26 jonathan pmichaud: Not sure - I need to look at what newclosure does more closely to really understand what's going on here...
15:27 pmichaud jonathan: no problem -- just wanted to get the ticket on record.  I'll probably bring it up at #parrotsketch if nobody's addressed it by then.
15:28 pmichaud theoretically we don't need the newclosure in this case (for the immediate block), but when I don't generate those as closures I get other bugs -- which probably also deserve tickets.
15:34 jonathan pmichaud: Would be good to give the whole thing a review.
15:34 jonathan Maybe I get some free time to look over it later today or tomorrow.
15:36 pmichaud there's also the issue with :outer referencing :load :init subs
15:36 pmichaud which allison has fixed in the pdd25cx branch
15:36 pmichaud I don't know if that will fix this or not.
15:36 jonathan Hmm, OK.
15:36 pmichaud might try testing the .pir in pdd25cx, though, just to see.
15:36 jonathan Do you have any idea how far we are off merging that into trunk?
15:36 pmichaud anyway, I have to head off to a meeting -- bbl.
15:36 jonathan OK, cu
15:37 pmichaud I don't know how far off a merge is.  We got PGE to compile in the branch, but that pointed up some other failures.
15:37 pmichaud afk # meeting
15:49 davidfetter joined #parrot
16:35 ron joined #parrot
16:38 ron joined #parrot
16:51 Zaba joined #parrot
17:03 ank joined #parrot
17:19 gmansi joined #parrot
17:29 zarchne joined #parrot
17:56 Ademan joined #parrot
18:01 tetragon joined #parrot
18:15 gmansi joined #parrot
18:18 Khisanth joined #parrot
18:47 gmansi joined #parrot
19:10 dalek r28600 | chromatic++ | trunk:
19:10 dalek : [IMCC] Plugged the keychain memory leak in IMCC.  All key-type structures
19:10 dalek : leaked memory proportional to the number of elements in the key.  Now they
19:10 dalek : don't.
19:10 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28600
19:26 dalek r28601 | chromatic++ | trunk:
19:26 dalek : [IMCC] Tidied the rest of the SymReg code.  There are no functional changes.
19:26 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28601
19:43 dalek r28602 | chromatic++ | trunk:
19:43 dalek : [src] Fixed Hash declarations so that both C and C++ compile again (NotFound,
19:43 dalek : RT #55960).
19:43 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28602
19:46 gryphon joined #parrot
19:48 gmansi joined #parrot
19:50 cognominal joined #parrot
19:56 kid51 joined #parrot
20:06 dalek r28603 | jkeenan++ | searchdocs:
20:06 dalek : Branch was merged into trunk on June 1; no longer needed.
20:06 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28603
20:08 dalek r28604 | jkeenan++ | searchdocs-27802:
20:08 dalek : Branch to which this tag corresponded has been removed from trunk; tag; no longer needed.
20:08 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28604
20:09 dalek r28605 | jkeenan++ | nom4:
20:09 dalek : Creating nom4 in https://svn.perl.org/parrot/branches
20:09 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28605
20:09 dalek r28606 | jkeenan++ | nom4-28604:
20:09 dalek : Tagging trunk at r28604 so that the nom4 can later be synched to it.
20:09 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28606
20:11 ruoso joined #parrot
20:20 dalek r28607 | jkeenan++ | nom4:
20:20 dalek : In branch, remove config step class auto::m4 and associated test file.  Update MANIFEST.
20:20 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28607
20:23 dalek r28608 | jkeenan++ | trunk:
20:23 dalek : File needed updating.
20:23 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28608
20:33 jhorwitz joined #parrot
20:36 dalek r28609 | jkeenan++ | opsrenum:
20:36 dalek : Creating opsrenum in https://svn.perl.org/parrot//branches
20:36 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28609
20:36 dalek r28610 | jkeenan++ | opsrenum-28608:
20:36 dalek : Tagging trunk at r28608 so that the opsrenum can later be synched to it.
20:36 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28610
20:51 vhold joined #parrot
20:59 slightlyoff joined #parrot
21:00 slightlyoff left #parrot
21:00 paco joined #parrot
21:01 dalek r28611 | pmichaud++ | trunk:
21:01 dalek : [rakudo]:
21:01 dalek : * Fix trailing space problem in src/parser/actions.pm .
21:01 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28611
21:03 dalek r28612 | pmichaud++ | trunk:
21:03 dalek : [rakudo]:
21:03 dalek : * Add item() function.
21:03 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28612
21:22 Zaba joined #parrot
21:37 bacek morning everyone
21:37 bacek perl6: sub my ($a,$b) { say $a+$b }; my $a=40; my $b=2;
21:37 polyglotbot RESULT[2]
21:37 bacek perl6: sub my ($a,$b) { say $a+$b }; my $a=40; my $b=2; my ($a,$b);
21:37 polyglotbot OUTPUT[too few arguments passed (1) - 2 params expected␤current instr.: 'my' pc 100 (EVAL_12:40)␤called from Sub '_block11' pc 94 (EVAL_12:34)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:469)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1088
21:37 polyglotbot ..(src/PCT/HLLCompiler.pir:598)␤called from Sub 'parrot;PCT::H...
21:38 bacek perl6: sub my ($a) { say $a }; my $a=40; my $b=2; my ($a,$b);
21:38 polyglotbot OUTPUT[402␤]
21:38 dalek r28613 | fperrad++ | libs4php:
21:38 dalek : [phc] add boolean & NULL
21:38 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28613
21:38 diakopter perl6: sub say { undef.say }; say();
21:38 polyglotbot OUTPUT[[oops; continuation 0xb7292cc4 of type 21 is trying to jump from runloop 1421 to runloop 98]␤[oops; continuation 0xb72932ac of type 21 is trying to jump from runloop 1421 to runloop 98]␤[oops; continuation 0xb7293894 of type 21 is trying to jump from runloop 1421 to runloop 98]␤[oops;
21:38 polyglotbot ..continuation 0xb7293e7c of type 21 is trying to jump f...
21:39 dalek r28614 | fperrad++ | libs4php:
21:39 dalek : [php] php_sape_name uses the framework
21:39 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28614
21:40 bacek pmichaud, it's a bug in handling 'scoped' rule in grammar.
21:42 jonathan [my|has|our] (...) is, I think, known as unimplemented.
21:42 jonathan Or as far as I last knew, anyway.
21:44 bacek jonathan, it's not implemented.
21:44 dalek r28615 | fperrad++ | libs4php:
21:44 dalek : [php] use PHP types
21:44 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28615
21:45 bacek perl6: sub foo {}; say foo;
21:45 polyglotbot OUTPUT[Null PMC access in get_string()␤current instr.: 'print' pc 9632 (src/gen_builtins.pir:6354)␤called from Sub 'say' pc 9654 (src/gen_builtins.pir:6364)␤called from Sub '_block11' pc 38 (EVAL_10:16)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:469)␤called from
21:45 polyglotbot ..Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1088 ...
21:46 bacek yet another bug...
21:47 bacek perl6: say sqrt(1i),' ', (1+1i)/sqrt(2)
21:47 polyglotbot OUTPUT[1 0.707107+0.707107i␤]
21:50 diakopter perl6: say sqrt(1     i),' ', (1+1   i)/sqrt(2)
21:50 polyglotbot OUTPUT[1 0.707107+0.707107i␤]
21:51 diakopter perl6: sub i { 5 }; if 1 {3i i};
21:51 polyglotbot RESULT[0+3i]
21:53 diakopter perl6: sub i { 5 }; if i() {3i i i};
21:53 polyglotbot RESULT[0+3i]
21:53 * diakopter stares
21:53 diakopter perl6: sub i { 5 }; if i {3i i i};
21:53 polyglotbot OUTPUT[Could not find non-existent sub if␤current instr.: '_block11' pc 46 (EVAL_11:21)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:469)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1088 (src/PCT/HLLCompiler.pir:598)␤called from Sub
21:53 polyglotbot ..'parrot;PCT::HLLCompiler;command_line' pc 1267 (src/PCT/HLLCompiler.p...
21:59 diakopter perl6: sub i { say 4; [5] }; if =i(i) {say, 3;i(i)};
21:59 polyglotbot OUTPUT[4␤4␤␤4␤4␤]
22:00 diakopter this helps a bit
22:00 diakopter perl6: sub i { say 4; [5] }; if =i(i) {say 2, 3;i(i)};
22:00 polyglotbot OUTPUT[4␤4␤23␤4␤4␤]
22:07 pmichaud my ($a, $b)  isn't implemented yet.  See RT#53902.
22:09 bacek pmichaud, I checked src/parser/gammar :)
22:10 pmichaud I'm just saying it's a known bug.  :-)
22:10 bacek it's not implemented and I can't implement it... 30+ seconds for recompile killing me...
22:10 bacek perl6: sub foo {}; say foo;
22:10 polyglotbot OUTPUT[Null PMC access in get_string()␤current instr.: 'print' pc 9632 (src/gen_builtins.pir:6354)␤called from Sub 'say' pc 9654 (src/gen_builtins.pir:6364)␤called from Sub '_block11' pc 38 (EVAL_10:16)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:469)␤called from
22:10 polyglotbot ..Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1088 ...
22:11 bacek is parrot or rakudo bug?
22:11 pmichaud probably not a parrot bug.
22:11 pmichaud might be a pct bug.
22:11 bacek perl6.past: sub foo {}; say foo;
22:11 nopaste "polyglotbot" at 193.200.132.146 pasted "perl6 past paste" (91 lines) at http://nopaste.snit.ch/13338
22:12 pmichaud it may be that we need an empty block to return undef.  Or perhaps an empty hash.
22:12 bacek perl6: sub foo {undef}; say foo;
22:12 polyglotbot OUTPUT[␤]
22:12 pmichaud since {} is really a hash composer.  :-)
22:12 bacek yak :)
22:13 bacek pmichaud, btw, in S29-list/map.t block fudged with 'empty statement in closure' works for me.
22:14 bacek perl6: say ~(1,2,3,4).map:{ $^a + $^b }
22:14 polyglotbot OUTPUT[Statement not terminated properly at line 1, near ":{ $^a + $"␤current instr.: 'parrot;PGE::Util;die' pc 120 (runtime/parrot/library/PGE/Util.pir:82)␤called from Sub 'parrot;Perl6::Grammar;eat_terminator' pc 20609 (src/gen_grammar.pir:2813)␤called from Sub
22:14 polyglotbot ..'parrot;Perl6::Grammar;statementlist' pc 19622 (src/gen_grammar.pir:2450)␤called fr...
22:14 pmichaud adverbial closures not implemented.
22:15 bacek i know...
22:16 bacek what do :: in grammar?
22:17 pmichaud it's a 'cut' operation
22:17 pmichaud prevents backtracking past the ::
22:19 bacek pmichaud, thanks.
22:21 bacek pmichaud, another question: http://rt.perl.org/rt3/Ticket/Att​achment/431776/197156/test2.diff
22:21 shorten bacek's url is at http://xrl.us/ef2ce
22:21 bacek this is modified Test.pm and test for Test.pm
22:22 bacek (and small patch for eval() to prevent dieing on parsing)
22:26 pmichaud interesting.  I'll consider the patch for a while (but testing Test.pm isn't high on my priority list atm)
22:27 pmichaud seems like we could get almost as good by just introducing $expect_fail into Test.pm directly, rather than trying to override the sub.
22:30 anr joined #parrot
22:32 apeiron I'm trying to submit a smoke for parrot 0.6.3 on FreeBSD and I'm getting this error: * Sending data to smokeserver "http://smoke.parrotcode.org/smoke/"... error: The submitted smoke has an invalid format! Am I missing something, or did I do something incorrectly?
22:32 bacek pmichaud, It's 'pugs' approach.
22:33 bacek They use named 'todo' arg.
22:33 diakopter apeiron: usually that's b/c you did svn export, not svn checkout
22:33 apeiron diakopter, This is directly from a 0.6.3 tarball.
22:33 pmichaud we're not using named 'todo' for the test suite.
22:33 diakopter apeiron: that too :)
22:33 apeiron That machine doesn't even *have* svn.
22:33 pmichaud using :todo was from when the test suite was only testing pugs.
22:34 diakopter yes, non-svn-checkouts can't submit smokes
22:34 apeiron ah.
22:34 apeiron Now you tell me.
22:34 diakopter apeiron: actually
22:34 pmichaud ...why does 'make smoke' require svn, ooc?
22:34 diakopter if you modify the submission file to include some svn revision, the smoke server will accep tit
22:34 diakopter accept it
22:35 diakopter b/c it includes the svn revision in the submission
22:35 apeiron I'm not sure if fooling it like that would be helpful...
22:35 pmichaud seems like we ought to be able to get 'make smoke' to recognize "0.6.3" as a valid revision :-)
22:35 pmichaud for the non-svn checkouts.
22:36 apeiron Not a big deal, I'm installing svn presently.
22:36 apeiron (well, not a big deal for me)
22:36 pmichaud well, we'd like to get smokes from lots of platforms :-)
22:36 diakopter apeiron: I don't know what revision the 0.6.3 release was
22:37 diakopter pmichaud: could the 'make release' target put the r in some text file somewhere?
22:37 dalek r28616 | pmichaud++ | trunk:
22:37 dalek : [rakudo]:
22:37 dalek : * Update rakudo's grammar with various changes to STD.pm .
22:37 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28616
22:37 apeiron pmichaud, I have some thoughts on putting together a federation of VMs using VMWare and using Perl-driven SSH and such to autosmoke.
22:37 pmichaud apeiron: that would be very cool.
22:38 apeiron I'm working with Fusion for the Mac right now as I'm on my MBP.
22:38 apeiron Maybe someone with the appropriate license keys or whatnot can create appliances for different platforms?
22:38 dalek r28617 | fperrad++ | libs4php:
22:38 dalek : [php] is_bool, is_float is_long is_null is_string strval
22:38 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=28617
22:39 diakopter apeiron: VMWare server 2.0 beta is free... and pretty neat. linux or windows tho
22:39 apeiron diakopter, I tried the Fusion 2.0 beta and it b0rked networking. :(
22:39 apeiron pmichaud, The basic premise is: start VM, ssh in, svn up, make smoke, shutdown VM, move onto the next.
22:40 diakopter oh; I don't know about Fusion 2... I just know about Fusion 1 and Server 2 beta
22:40 * Tene really likes KVM for virtualization.
22:40 diakopter Server 2 beta has been around for 1.5 years at least
22:40 apeiron Tene, Heh. Being that $WORK paid for this MBP and its copy of Fusion... :)
22:41 diakopter rakudo: DWIM: { say system('uname -a') }
22:41 diakopter perl6: DWIM: { say system('uname -a') }
22:41 polyglotbot OUTPUT[Statement not terminated properly at line 1, near ": { say sy"␤current instr.: 'parrot;PGE::Util;die' pc 120 (runtime/parrot/library/PGE/Util.pir:82)␤called from Sub 'parrot;Perl6::Grammar;eat_terminator' pc 20609 (src/gen_grammar.pir:2813)␤called from Sub
22:41 polyglotbot ..'parrot;Perl6::Grammar;statementlist' pc 19622 (src/gen_grammar.pir:2450)␤called fr...
22:41 diakopter no block labels?
22:43 pmichaud not yet.
22:43 diakopter apeiron: 0.6.3 was r28470
22:44 diakopter try 'r28470' or '28470' in that spot...
22:44 diakopter it should submit
22:44 pmichaud also, system() isn't valid Perl6.  Try 'run'.
22:44 diakopter pmichaud: ah
22:44 apeiron diakopter, All right. Will give smoking that a shot once svn is installed.
22:44 pmichaud perl6:   say run('uname -a');
22:44 polyglotbot OUTPUT[Linux feather3 2.6.18-6-xen-686 #1 SMP Sun Feb 10 22:43:13 UTC 2008 i686 GNU/Linux␤0␤]
22:51 teknomunk joined #parrot
22:57 nopaste "bacek" at 202.7.166.165 pasted "Slightly reworked Test.pm and test for pmichaud" (142 lines) at http://nopaste.snit.ch/13339
23:01 bacek yak... I forgot to change comments...
23:04 nopaste "bacek" at 202.7.166.165 pasted "Slightly reworked Test.pm and test for pmichaud (once again)" (142 lines) at http://nopaste.snit.ch/13340
23:05 bacek pmichaud, it this version ok?
23:05 pmichaud I like it better, yes.
23:07 bacek send it to RT?
23:08 bacek pmichaud, You promise to take a look :)  http://bacek.homeip.net/cgit/parrot/diff/​languages/perl6/src/classes/List.pir?id=5​6a09b46629e68af6f496ad1dddc966bdb50c37f
23:08 shorten bacek's url is at http://xrl.us/ego7m
23:09 bacek it's lazy list (first cut). Keep working on it or throw away as full crap?
23:09 Zaba joined #parrot
23:16 cjfields joined #parrot
23:16 Limbic_Region pmichaud ping
23:19 bacek Limbic_Region, you shouldn't "ping" pmichaud. You should "summon" him :)
23:19 bacek summon pmichaud!
23:20 bacek spell failed...
23:20 purl failed... looks funny
23:20 diakopter spell purl
23:20 purl Possible spellings for purl: URL burl curl furl hurl Perl pull pure purr
23:21 * Limbic_Region heard that patience is a virtue
23:21 Limbic_Region not one that I have, but I can wait :-)
23:22 apeiron Also, just 'cause people say something doesn't mean it's true. :) (cf "perl is dying" tripe)
23:25 diakopter apeiron: but something that'd dead can't be dying..
23:25 pmichaud Limbic_Region: pong
23:26 apeiron diakopter, You're not saying perl is dead, are you?
23:26 pmichaud bacek: I still don't see why we need a ListIterator class.
23:27 diakopter apeiron: sorry; was kidding; a bad/botched joke/reply
23:27 diakopter mega-botched
23:27 apeiron diakopter, FWIW I don't believe you'd say such a thing so I'm not about to call you heathen. :)
23:28 diakopter I shoot my mouth off too often.. you shouldn't give me the benefit of the doubt like that ;)
23:28 bacek pmichaud, removing it in TODO.
23:28 * apeiron is reminded of old school Daffy Duck cartoons...
23:29 Limbic_Region pmichaud - sorry, have to take care of the 2 year old - bbias
23:29 cjfields do we have a PIR coding std?
23:30 cjfields (I have seen the C one in docs)
23:30 bacek perl6: my @a=(1..2); say ~@a; say ~@a;
23:30 polyglotbot OUTPUT[1 2␤1 2␤]
23:30 Limbic_Region pmichaud - in your PCT talk, I am pretty sure all of the functions you demo'd had existing parrot ops - if you are creating a function that does not - what are your options?  PIR/PASM only?
23:31 pmichaud one can always write a new op or method in C
23:31 pmichaud (or, of course, anything that compiles down to PIR/PASM :-)
23:32 cjfields I have attached a patch to RT#55492 for transliteration which is passing tests (t/spec/S05-transliteration/trans.t)
23:32 cjfields though tr/// is skipped (not implemented yet)
23:36 Eevee joined #parrot
23:36 Limbic_Region pmichaud - but if one were going to write a new op/method in C, one would need to get it integrated into parrot right?
23:36 bacek t/spec/S29-str/comb.rakudo......................  10   10    0    0    0   10
23:36 bacek t/spec/S29-str/sprintf.rakudo...................  37   37    0    0    0   37
23:37 bacek pmichaud, we should remove them from spectest_regression.data...
23:37 pmichaud bacek: comb has already been removed.
23:37 bacek (Or implement them :)
23:38 pmichaud I'm not sure what to do with sprintf.  I suppose we could just try parrot's sprintf and see what happens.  :-)
23:38 pmichaud l_r:  NCI and dynamic ops allow things to be written in C w/o integrating them into parrot's core.
23:39 bacek pmichaud, No. comb.t still in.
23:39 bacek But commit comment says that it removed.
23:39 Limbic_Region pmichaud - thanks.  That answered my question(s).
23:39 bacek yak. forget it
23:40 Limbic_Region pmichaud - if I wanted to do something stupid like implementing state - where would I look?
23:40 pmichaud I haven't learned 'state' yet.  But probably globals and lexicals.
23:41 pmichaud okay, I _have_ to fix $_ tonight.
23:41 pmichaud the current implementation is really broken -- it always assumes $_ is a parameter.
23:42 diakopter perl6: my $a = sub { say $_; }; $a();
23:42 polyglotbot OUTPUT[too few arguments passed (0) - 1 params expected␤current instr.: '_block17' pc 64 (EVAL_10:33)␤called from Sub '_block11' pc 58 (EVAL_10:27)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:469)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1088
23:42 polyglotbot ..(src/PCT/HLLCompiler.pir:598)␤called from Sub 'parrot;P...
23:44 cjfields pmichaud: Hate to bug you.  When you have time, can you take a look at RT#55492 (patch)?  It works around the .HLL mapping issue and passes all transliteration method tests.
23:45 Limbic_Region pmichaud - WRT to state variables, see http://perlmonks.org/?node=473584 # Larry's explanation there is better than the one in S04
23:46 Limbic_Region pmichaud - also see http://use.perl.org/~nicholas/journal/36270 for a good test case
23:47 * cjfields thinking pm's gotta full plate right now...
23:47 diakopter cjfields: not to mention a full stomach
23:47 cjfields ;)
23:48 cjfields speaking of, I need to get something...
23:49 pmichaud l_r:  I'm not sure what storage mechanism state variables will use, but we'll probably use the :viviself parameter of PAST::Var nodes to initialize them on declaration.
23:49 pmichaud we may end up having to have a :scope('state') type added to PAST::Var, too.
23:50 pmichaud I'll have to think about it all a bit.  Anything dealing with "first" initialization is often a bit tricky.
23:51 pmichaud cjfields: don't all of the arguments first need to be mapped to a single hash before applying them?  it looks to me as though this algorithm applies the first argument, then the second, then the next, etc.
23:51 Limbic_Region pmichaud - ok, all that meant to me is that it is beyond my current knowledge/understanding so I will make any attempt on my own
23:52 pmichaud cjfields: or maybe I'm misreading... just a sec.
23:53 cjfields pmichaud: the problem with trans (as described in S05) is the Pair key/value could be an Array, a Range, or a List.
23:53 cjfields or combinations thereof.
23:54 cjfields s/List/String
23:54 pmichaud right.  So I'm thinking you need to loop through all of them to build a "master" key=>value mapping, and then use it to perform the substitution.
23:54 pmichaud ...which it appears is close to what you're doing.
23:55 pmichaud in
23:55 pmichaud +     orig = substr tmps, pos, 1
23:55 pmichaud ...what if the thing being substituted isn't a single character?
23:55 * cjfields looking back at S05...
23:57 cjfields Do you means (in S05): 'The array version can map one-or-more characters to one-or-more characters:'
23:57 pmichaud yes.
23:59 cjfields I'll add some tests for that.  I think I'll just need to rework the actual mapping bit at the end.

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

Parrot | source cross referenced