pugscode.org/ | nopaste: sial.org/pbot/perl6 | pugs: [~] <m oo se> (or rakudo:, kp6:, elf: etc.) (or perl6: for all) | irclog: irc.pugscode.org/
Set by Tene on 29 July 2008.
00:01 hercynium left 00:03 Ontolog left 00:15 justatheory left
ispy_ @seen rindolf 00:15
lambdabot I saw rindolf leaving #perl6 4h 30m 57s ago, and .
00:17 justatheory joined 00:18 kst```` left 00:20 kst joined 00:24 justatheory left 00:37 kanru left
pugs_svn r22272 | ruoso++ | [smop] the leaks are still there, but now array iterator is working... 00:45
00:48 kanru joined 00:50 hercynium joined, BinGOs left 00:51 wknight8111 joined 00:54 baest left
ruoso . o O ( now we're at a very close step from having map implemented... ) 01:00
01:41 abra joined 01:46 mncharity joined
mncharity perl6: say(<a b c>.Str) 01:46
p6eval rakudo 31216: OUTPUT[Method 'Str' not found for invocant of class 'List'␤current instr.: '_block11' pc 46 (EVAL_12:18)␤]
..elf 22272: OUTPUT[abc␤]
..pugs: OUTPUT[*** No such method in class Array: "&Str"␤ at /tmp/ic67Au3jeE line 1, column 5-16␤]
meppl good night
mncharity g'night meppl :)
perl6: say(<a b c>)
p6eval elf 22272, pugs, rakudo 31216: OUTPUT[abc␤]
meppl ;)
01:47 meppl left
mncharity perl6: my $arrayref = <ok 4 - say stringifies its args>; say $arrayref; 01:47
p6eval pugs, rakudo 31216: OUTPUT[ok 4 - say stringifies its args␤]
..elf 22272: OUTPUT[ok4-saystringifiesitsargs␤]
mncharity perl6: my $arrayref = <ok 4 - say stringifies its args>; say $arrayref.WHAT; 01:48
p6eval elf 22272, pugs, rakudo 31216: OUTPUT[Array␤]
01:59 araujo joined 02:03 Limbic_Region left
s1n moritz_: where do i find the actual definition of is() that's used in Test.pm? 02:24
err well, n/m...
02:27 wknight8111 left 02:32 mberends joined
pugs_svn r22273 | putter++ | [elf] ++ and -- now work. Makefile tweaks. Test results synced. 02:33
r22273 | putter++ | New Makefile target test_ok_files tests only files which run_tests said passed.
r22273 | putter++ | It uses prove, which is more picky than run_tests, so it doesn't agree.
r22273 | putter++ | There's still no simple yes/no answer to "did I break something?". moritz++
mncharity closest answer is make test . ie, elf needs to continue to compile itself, and any changed results to t/ should ideally be understood. though in practice, that second part is often neglected. optional at this point. 02:35
good night all.
02:35 mncharity left 02:48 sri_work left 02:51 z80ASM joined, slavik joined
slavik is there a perl6 code editor that has autocomplete and syntax highlighting for perl6? 02:52
also, am I correct in assuming that pugs has rules?
s1n @tell pmichaud i have it tested and it works, sort of, there's a bug i don't know how to solve so i need your expertise 03:13
lambdabot Consider it noted.
rakudo_svn r31218 | tene++ | Merge exceptionmagic branch into trunk. 03:35
03:35 drbean_ joined
Tene slavik: There's decent vim support. Look in the editors/ directory. 03:36
slavik hmm, ty
I'll have to look up vim auto tab stuffs 03:37
03:50 kanru left 03:52 drbean left 03:56 kanru joined 03:59 Bzek left 04:08 Bzek joined 04:11 Bzek left 04:12 z80ASM left 04:15 Bzek joined 04:16 Bzek left, Bzek joined 04:25 abra left 04:27 scrottie joined
scrottie beppu, awwaiid. yo. 04:28
04:57 ispy_ left 04:58 ispy_ joined, xinming_ joined 05:11 xinming left 05:18 sail0r joined 05:22 sail0r left 05:37 hercynium left 05:54 slavik left 05:56 BinGOs joined 05:58 Psyche^ joined 06:07 ashizawa joined 06:13 Patterner left, Psyche^ is now known as Patterner, Ontolog joined 06:22 baest joined 06:39 mberends left 06:44 alester left 07:04 jferrero joined 07:22 Lorn left 07:29 jferrero left 07:30 zamolxes left 07:38 penk joined 07:45 iblechbot joined
Ontolog what is the command to run all the tests? 07:54
moritz_ make spectest_regression
or make fulltest # also runs condingstd tests
it was r31218 that introduced the segfault, so not your problem ;) 07:55
Ontolog cool but it seems there are problems with my implementation 08:03
i'll look into it
08:24 barney joined 08:29 pmurias joined 08:44 smg left 08:47 Ontolog left
rakudo_svn r31223 | tene++ | Revert my previous exceptions commit. It was wrong. 08:48
pmurias ruoso: have you tried making SMOP_REFERENCE and SMOP_RELEASE print line info or should i try it? 08:50
08:55 Auzon left 09:02 zamolxes joined 09:16 jan_ left, yves left, [particle] left 09:19 fullermd joined, jfredett joined, pjcj joined 09:21 Ontolog joined, jan_ joined, yves joined, [particle] joined, gbacon joined, simcop2387 joined, pasteling joined, rakudo_svn joined 09:22 Ontolog left, Ontolog joined
Ontolog moritz_: did you see the mail about matches? 09:23
moritz_ Ontolog: no, currently I don't have mail access :( 09:24
Ontolog ohh
The question is, should say "ab1cd12ef123gh".match(/\d+/).perl; 09:25
return ["1", "12", "1", "2", "123", "12", "1", "23", "2", "3"] 09:26
moritz_ rakudo: say "ab1cd12ef123gh".match(/\d+/).perl
p6eval rakudo 31223: OUTPUT[{}␤]
Ontolog match is not implemented
i implemented it
on my machine
but all my match does is iterate over matches by calling match.'next'() and getting match.'text'() 09:27
and the result is the list shown
moritz_ thought he had applied a patch from cjfields that implements match
that's wrong 09:28
Ontolog well in Str.pir there is no sub 'match'
moritz_ it could be in src/builtins/any-str.pir
Ontolog if you can get at your mail, check out the diff i sent
i see
well running .match on a string doesn't work normally
moritz_ rakudo: say "abc".match(/b./) 09:29
p6eval rakudo 31223: OUTPUT[bc␤]
Ontolog my point is not so much to implement match(), but if I did that's cool, my point is to illustrate what is happening
09:29 bennymack left
Ontolog hm? 09:29
i guess it is my bad..
let me check it out again....
moritz_ I think that match.next doesn't do what we think it does, or is broken, or ... dunno
what we need for Perl 6 is non-overlapping matches 09:30
so 1, 12, 123 in your example above 09:31
Ontolog yes
that's what I expected
moritz_ it looks as though match.next gives all possible matches
Ontolog exactly
moritz_ maybe there's a way to supply a start position to a regex match? 09:32
Ontolog btw, the match I implemented in Str.pir behaves differently than the existing match
moritz_ something like regex(string, start)
(I don't have a copy of parrt here, so I can't check :( 09:33
Ontolog how will that fix the problem in match?
moritz_ in perl psudo code: $previous = 0; while $regex($str, $previous) { push @ret, ~match; $previous = $match.to) 09:35
errm, think $/ instead of match
Ontolog I see what you are saying but is that just a workaround for a bug? 09:42
09:42 felipe joined
moritz_ no 09:42
I think PGE needs backtracking match.next semantics 09:43
Ontolog ok
moritz_ lunch &
Ontolog as to the impl of match in any-str.pir and the one i did for myself in Str.pir
ok ttyl
when you return: shouldn't match.(/rx/) return a List? 09:44
09:46 bennymack joined
ruoso pmurias, please, try it... I still didn't try... 09:52
pmurias what c flag should be used to enable such tracing? 09:54
ruoso there's already one... 09:55
SMOP_LOWLEVEL_MEM_DEBUG
pugs_svn r22274 | pmurias++ | [smop] SMOP_RELEASE and SMOP_REFERENCE print line,function and file with SMOP_LOWLEVEL_MEM_DEBUG 10:02
pmurias ruoso: i'm not even sure what the leaking Code is 10:06
10:08 Jedai joined 10:09 masak joined 10:10 iblechbot left 10:15 smg joined
ruoso pmurias, test/32 is an example... 10:18
ruoso later &
10:18 ruoso left 10:20 pmurias left
rakudo_svn r31225 | jonathan++ | [rakudo] Handle the case where a fractional part of a number contains underscores; PIR doesn't allow this, so just whip 'em out. Resolves RT#58998. 10:28
10:36 elmex joined
rakudo_svn r31226 | jonathan++ | [rakudo] Make use happen at BEGIN time. This lets us use modules that contain classes. However, we still have various issues to follow up on with this (for example, making it work when you use a module that does a use itself, which gives very odd errors at the moment, and 10:47
..making this play well when a pre-compiled module does a use).
Ontolog yay, i will soon have a full week to work on perl6! muahahaha 10:54
if only someone will lay down the law with regards to match() behavior 10:55
i could write a decent patch ;-)
10:55 wknight8111 joined 10:59 Ontolog left
rakudo_svn r31227 | jonathan++ | [rakudo] Sort out keys, values and kv on mappings (and thus hashes) by exporting them, as we do for List. Gets us passing another 8 TODO'd spectests. 11:11
11:38 iblechbot joined 11:41 masak left 11:43 Ontolog joined
rakudo_svn r31228 | jonathan++ | [rakudo] We currently always store subs in the namespace without sigil, so if we get a sub parameter with the & sigil then we need to strip it off. Resolves RT#58828. 11:44
11:51 wknight8111 left 11:56 penk left 12:04 BinGOs left, BinGOs joined 12:10 ispy_ left 12:21 abra joined 12:37 ispy_ joined 12:38 sirhc joined 12:50 ashizawa left 12:56 REPLeffect left 13:07 macae joined
Juerd Let's call undef "unvalue" :) 13:28
Antivalue :)
13:30 macae left 13:31 pmurias joined
pmurias Juerd: a NULL pointer is the antivalue ;) 13:31
13:33 Front_slash joined
Juerd Heheh 13:34
13:35 Front_slash left 13:36 ispy_1 joined 13:39 Lorn joined 13:40 Front_slash joined 13:44 c1sung joined 13:47 c1sung left, c1sung joined 13:48 TJCRI joined 13:52 ispy_ left 14:01 mberends joined 14:02 ispy_1 left 14:03 alester joined, NeatFires joined
NeatFires when? 14:03
pmurias NeatFires: ? 14:04
pugs_svn r22275 | pmurias++ | [smop] SMOP_LOWLEVEL_MEM_DEBUG shouldn't be set by default 14:07
14:22 ruoso_ joined
ruoso_ Hello! 14:22
14:27 omichael joined
ruoso_ pmurias: I'm adding a little more debug information... 14:44
ruoso_ back to hunting leaks
pugs_svn r22276 | moritz++ | [t/spec] unfudged passing TODO tests for rakudo
r22277 | moritz++ | [t/spec] more unfudging 14:45
r22278 | moritz++ | [t/spec] more unfudging of passing TODO tests 14:47
PerlJam 5/away
14:54 sri_work joined
pugs_svn r22279 | moritz++ | [t/spec] tests for code objects in signatures 15:01
r22280 | moritz++ | [t/spec] svn props on newly added test file 15:02
15:05 Exodist joined 15:11 ruoso_ left 15:12 ruoso__ joined 15:16 ruoso__ left
rakudo_svn r31229 | moritz++ | [rakudo] added tests for the fix in r31228 to spectest_regression 15:26
15:31 pmurias left 15:33 kanru left 15:36 exodist_ joined, zamolxes left 15:41 Exodist left 15:48 barney left 16:10 wknight8111 joined 16:21 aindilis joined 16:24 justatheory joined 16:32 baest left 16:35 ashleyb joined 16:39 wknight8111 left 16:42 smg left 16:49 wknight8111 joined
Juerd I haven't been active on the mailinglists for a while now 16:52
But now I've decided to stop reading them too. Sorry 16:53
Perl 6 activity is too much for me to follow :)
[particle] we're not slowing down just so you can catch up :P 16:54
Juerd No, I don't want you to. I lack tuits to catch up.
[particle] you'll find them some day. we'll be here.
Juerd It's just that I have been reading only 10% of the mail anyway, and still it's eating too much time
I'm cutting down on mailinglists and irc 16:55
Just removed 6 mailing lists from mutt's check-these-mailboxes-for-new-mail list
[particle] i went from 10 irc channels down to three. i'm up to six now :(
Juerd 20 here :( 16:56
I find it very hard to choose which I'm leaving.
But I'll have to do it soon. 16:57
[particle] 20 is way too many
Juerd For quite a while I've had a highlight on /utf-8/ in any channel =~ /perl/
But I can't be everyone's unicode help.
Also, finally my message got through to most of the community so I no longer have to :) 16:58
moritz_ yes, Juerd++ 16:59
rakudo_svn r31234 | moritz++ | [rakudo] implement Any.split(/regex/), Ontolog++
r31234 | moritz++ | Patch courtesy by Chris Davaz, <cdavaz at gmail dot com>
r31234 | moritz++ | Note that currently Str.split(Str) mask this method for strings, which should be
r31234 | moritz++ | fixed by moving that to Any as well.
Juerd still wants tuits to rewrite p5 unicode docs, but that'll probably never happen
16:59 meppl joined
Juerd By the time I have enough tuits for that, Perl 6 is probably already there, and I can just write P6 docs :) 17:00
I sincerely hope that Perl 6 will have way less Unicode documentation. It's part of life now. 17:01
rakudo_svn r31236 | moritz++ | [rakudo] fixed trailing whitespaces 17:02
moritz_ hpes that too
afk&
17:02 Chillance joined
[particle] juerd: have you looked at parrot's strings pdd? 17:07
17:07 justatheory left
Juerd [particle]: A while ago 17:07
[particle]: I'm curious if it'll be as efficient as Perl 5's strings. I fear it might not be.
17:08 exodist_ left
[particle] i'm not familiar enough with either design to comment 17:08
but i hope we're able to get review from p5pers who are
Juerd The main important thing is that there ought to be TWO string types. 17:09
One binary, one text.
Text is a sequence of encoded numbers 17:10
TimToady the important new idea is the NFG encoding, which gives you one integer per grapheme, which should be very efficient (after transcoding) 17:12
17:14 rindolf joined 17:16 Exodist joined 17:18 exodist_ joined, exodist_ left, exodist_ joined, exodist_ left 17:22 Exodist left 17:28 Bzek left 17:30 Exodist joined 17:39 ruoso joined 17:40 justatheory joined, zamolxes joined
ruoso Hello! 17:42
pugs_svn r22281 | ruoso++ | [smop] fixed two memory leaks. changed a bit the SMOP_LOWLEVEL_MEM_DEBUG code and added SMOP_MOLD_DEBUG and SMOP_SLIME_DEBUG that prints when it is evaluating some call both the name of the method as well as the type 17:46
ruoso @tell pmurias, I left a commented line with all the debug options in CMakeLists.txt... I'm using "./build/test/foo 2>&1 | less" to trace the leaks... 17:47
lambdabot Consider it noted.
rindolf Hi ruoso 17:55
ruoso hi rindolf
rindolf ruoso: what's up? 17:56
ruoso leaks everywhere ;)
heh... I'm doing some plumbing
18:01 Front_slash left 18:04 Exodist left, Exodist joined
ruoso Are packages called "::Foo" or "Foo::" ? 18:06
or "::Foo::?
PerlJam Just "Foo" unless you need to give perl more clue, then "::Foo" (AFAIK)
or even ::(Foo) 18:07
ruoso Just "Foo" refers to the class, doesn't it?
18:08 jhorwitz joined
PerlJam This is where I would say to go read the Synopsis as my memory isn't that good :) 18:10
ruoso :P 18:11
18:13 smg joined 18:14 pmurias joined 18:20 wknight8111 left
meppl good night 18:20
18:23 meppl left 18:34 jan_ left 18:37 pmurias left 18:42 ashleyb left 18:45 jan_ joined 18:59 Auzon joined 19:01 abra left 19:07 REPLeffect joined 19:10 larsen_ joined
ruoso wonders why he can't connect to irc.perl.org 19:10
rindolf ruoso: let me try. 19:11
ruoso: I can connect there - not sure about login.
ruoso: I'm on #moose and #perl-qa now.
ruoso hmm... there's a bogus proxy configuration somewhere... 19:13
ruoso better now... 19:16
19:22 explorer joined 19:47 baest joined 19:52 explorer left 19:57 braceta joined 19:59 justatheory left 20:13 justatheory joined 20:17 pmurias joined
ruoso hi pmurias 20:20
20:24 araujo left 20:27 araujo joined
ingy seen TimToady 20:31
seen TimToady?
meh
pmurias ruoso: hi 20:33
ruoso @seen TimToady
lambdabot TimToady is in #perl6. I last heard TimToady speak 3h 21m 24s ago.
ruoso ingy, ^
ingy :) 20:34
rindolf Maybe I should try seeing if my Perl6 Graham Function program works in Rakudo. 20:35
ingy TimToady: you said at some point that you had certain unicode issues with YAML::XS. I am in a good spot to pay attention, so please email [email@hidden.address] with any YAML requests. 20:36
ruoso pmurias, have you seen the changes I made? 20:37
pmurias yes 20:38
it shouldn't have been causing the leak 20:39
we need a proper way of creating prototypes for RI's instead of using the RI is it's own meta RI
trick
20:41 mtrimpe joined
pmurias shower& 20:43
ruoso: anyway clearing up the Code leaks makes the remaining on in 21_code much clearer 20:55
ruoso yep
20:56 larsen_ left 21:00 Southen joined
pmurias found it 21:02
pugs_svn r22282 | pmurias++ | [smop] fixed one leak in 21_code 21:04
pmurias ruoso: the convention is that normal c functions don't RELEASE their argument?
21:05 wknight8111 joined, eternaleye left
ruoso pmurias++ 21:05
pmurias, there's no convention for normal c functions...
pmurias c functions in smop
pugs_svn r22283 | pmurias++ | [smop] fixed remaining leak 21:10
21:13 braceta left 21:16 eternaleye joined
pmurias is there a TAP way of marking that a test is excpected to fail? 21:18
[particle] like # TODO ? 21:21
21:21 eternaleye_ joined
[particle] svn.perl.org/parrot/trunk/languages...y/01-tap.t 21:22
lambdabot tinyurl.com/445tfo
21:23 TJCRI left
pmurias [particle]: it doesn't work if the test dies 21:24
ruoso: we have no leaks now ;) 21:25
[particle] ah, dies.
ruoso pmurias++ for 1..*
[particle] you can't mark that in the tap, unless you don't use a plan 21:26
if you have a function like eval_dies_ok available, you can use that
pmurias ruoso: map should be a function not a map method 21:30
ruoso so the call should be postcircumfix:<()>, right? 21:31
moritz_ wtf is the readproctitle process on feather? 21:33
pmurias ruoso: yes 21:34
moritz_ and why does it need so much CPU?
pmurias ruoso: maybe map should be written in Perl 6?
ruoso pmurias, cool... do you want to take that to do now? I'm a bit busy now... 21:35
pmurias, at least in m0ld
that's for sure
21:36 eternaleye left
pmurias ruoso: i should be sleeping now, the pugs backend will need some work first 21:37
ruoso :P 21:38
pmurias g'night
21:38 pmurias left
ruoso later & 21:54
21:54 ruoso left 21:58 eternaleye_ left, eternaleye_ joined 22:02 gispita joined 22:04 NeatFires left 22:09 rindolf left 22:10 meppl joined
Juerd moritz_: readproctitle is an in-$0-logging mechanism 22:10
moritz_: It's using so much CPU because whatever it's logging, is broken and keeps logging stuff. At this point, we should be happy that it's not a file based log :) 22:11
It was commitbit. Killed. 22:13
22:26 gispita left 22:46 sri_work left 22:50 Auzon left 22:54 Lorn_ joined, omichael left 23:00 hercynium joined 23:06 Lorn left 23:17 Lorn joined 23:24 iblechbot left 23:30 Chillance left 23:31 justatheory left 23:32 Lorn_ left

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!