»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
[Coke] sorear: some references to $*compiler.verbose which doesn't exist. 00:10
sorear what is the context of that noun phrase? 00:11
[Coke] sorear: gist.github.com/4145972 00:14
I imagine that's several of the new failures. 00:15
yes, definitely is a lot of the new failures. 00:19
dalek ecza: f4918b8 | sorear++ | / (5 files):
Lift the (X11 license) line editor from csharp.exe. it's not readline but it's something. fixes #148
 00:37
sorear [Coke]: oh dear, it would seem "eval" is completely busted 00:39
#1 priority 00:40
[Coke] sorear__ 00:43
sorear__ 00:44
sorear++ #eek!
dalek ecza: dbbc6c7 | sorear++ | lib/Kernel.cs:
Improve dynamic variable handling to hopefully fix eval, Coke++
 01:20
sorear runs a spectest
colomon sorear++ 01:32
sorear [Coke]: 12 failing files now 01:36
[Coke] you have a list you gist? 01:38
sorear gist.github.com/4146151 01:39
[Coke] sorear: how to clear niecza's precompiled cache? 01:46
sorear rm -r ~/.local/share/NieczaModuleCache 01:47
it ought to be a lot better about handling stale files now though 01:48
dalek ecza: 7984f8d | sorear++ | Makefile:
Add an install target
 01:50
[Coke] You got a todo pass on t/spec/integration/weird-errors.t # 3 - still fails here. 01:53
(I'm getting different results on a lot of those. hurm.) 01:56
but tomorrow's run will be muuuch better. sorear++
sorear the is_run stuff is pretty dicey, especially if you don't have a usable glib# installed :/ 01:57
dalek ecza: 0247fe5 | sorear++ | Makefile:
Add boot-from-installed for bonsaikitten
 01:59
sorear yay, I'
yay, I've now finished everything I said I'd do today 02:00
diakopter ++
[Coke] sorear++ 02:00
japhb sorear, by 6 PM no less, good on ya! 02:03
sorear Mac users: What IDE do you use/would you recommend for C? 02:14
Tene If I were on a mac, I'd use vim. I doubt answering the right question, though. 02:47
benabik uses Vim. 02:48
Xcode is actually pretty decent.
dalek ecza: b04ce0c | sorear++ | docs/announce.v24:
Draft v24 announce
 03:33
sorear benabik: cool, I'll try that 03:34
japhb sorear, did v23 just not happen? Or is v24 an accidental double-bump? 04:22
sorear japhb: there was no v23 04:24
benabik We call it "the version that shall not be numbered".
japhb sorear, OK, roger that. 04:51
dalek rl6-bench: 0f05d4c | (Geoffrey Broadwell)++ | bench:
Rename some uses of 'tag' in extract command to 'rev' where that is the real semantic meaning, in preparation for being able to extract arbitrary rev specs
 05:09
rl6-bench: 0cfc0e5 | (Geoffrey Broadwell)++ | bench:
Make sure explode-components returns strings and not path objects for checkout names
rl6-bench: 376370b | (Geoffrey Broadwell)++ | bench:
s:g/'.IO.d'/.path.d/
rl6-bench: 83a5474 | (Geoffrey Broadwell)++ | bench:
Allow arbitrary revs to be extracted, as long as they can be recognized by 'git rev-parse'
rl6-bench: a273c4c | (Geoffrey Broadwell)++ | components.json:
Add build instructions for nqp in components.json
rl6-bench: 0dccbcc | (Geoffrey Broadwell)++ | analyze:
Factor out a little boilerplate for finding the list of enabled Perls; correctly show SKIP in analysis when test does not apply to a given compiler group
rl6-bench: d1e4438 | (Geoffrey Broadwell)++ | analyze:
Show all versions found when merging timing runs from different bench versions
japhb Huh. I half expected to accidentally cause dalek to kick itself.
or rather, to get itself kicked. 05:10
dalek rl6-bench: da28593 | (Geoffrey Broadwell)++ | analyze:
Make compare_times resilient to all compilers failing a test
 06:48
rl6-bench: eb3c239 | (Geoffrey Broadwell)++ | timeall:
Add more while tests; fix off-by-one error in while loop counters
japhb Benchmark comparison of current NQP, Rakudo, and Niecza releases, with startup time removed: gist.github.com/4146931 06:57
sorear japhb: my first project after the release is for/range and ~ optimization 06:58
japhb: did you read the announce.v24 or just comment on its title?
japhb skimmed it 06:59
Well, read it without all of my brain actually working, more like.
Why?
sorear my release procedure requires getting someone on the channel to sanitycheck the announcement 07:00
I'm also going to ask a windows person to check that the new line editor is usable on that platform 07:01
japhb The draft announcement certainly reads OK. 07:03
sorear starts putting together an RC 07:03
japhb I've not tested a build of niecza/master, but you know, I have the tools to try that easily now ...
sorear you didn't before? 07:04
japhb Within the benchmarking suite, I mean: ./bench extract niecza/master; ./bench build niecza/master
sorear feather.perl6.nl/~sorear/niecza-24-rc.zip # I'd like a windows person to check that the line editor works before releasing 07:09
diakopter volunteers
diakopter wonders whether mono is still on this computer
wow, Chrome *really* doesn't want me to download that .zip 07:10
"niecza-24-rc.zip is not commonly downloaded and could be dangerous. Discard. (click button to display other button Keep). <click Keep>. Are you sure? Don't say we didn't warn you." 07:13
dalek rl6-bench: da9a2dd | (Geoffrey Broadwell)++ | README:
Bring README up to current state of bench usage
 07:15
diakopter erm 07:16
japhb starts a benchmarking run for niecza/master and wanders off to relax for a bit
dalek nda: 1361fcb | (Caleb Cushing)++ | bootstrap.pl:
better debugging and recursively creates path

this fixes a problem where panda does not attempt to create the full directory structure if it does not exist (which is likely on a new install). Also provide a better error message that includes the path panda is trying to use in the event it is not writeable
Signed-off-by: Caleb Cushing [email@hidden.address]
nda: f4f19b5 | tadzik++ | bootstrap.pl:
Merge pull request #25 from xenoterracide/master

fixes issue with unwritable directory
diakopter sorear: seems to work 07:17
diakopter up/down/left/arrows do what I'd expect 07:17
sorear \o/
thanks. diakopter++
mtymula hello can anyone tell me how can i enter this room using my android phone?? I've downloaded client and now i need a server address and a port 08:36
moritz server address is irc.freenode.org 08:37
mtymula and a port?
moritz 6667 I think
japhb sorear, FWIW niecza/master is not very different from niecza/v22 in my benchmarks. I assume that's expected. :-) 08:40
japhb sleep, perchance to dream & 08:40
sorear expected, yeah 08:42
gnight japhb
mtymula nope 08:52
port 6667
is an asimov faun channel
;)
sorear I'm connected to irc.freenode.net port 6667 08:55
you connected to the right place
freenode's servers are named after authors
after joining asimov.freenode.net, you join #perl6
on PC clients it's usually done by typing /join #perl6 08:56
dunno about your android clienrt
the servers share channel information, so whichever server you get is good
when you connect you will be routed to a random server, for load balancing
mtymula89 Yeah!! 08:57
mtymula89 What a fool i am:) 08:57
sorear mtymula89: welcome 08:58
mtymula89 Hello now i will be on this irc hannel all day long:) 08:59
mtymula89 When i write a subroutine with arbitrary numberrs of parameters there is a *@name as a argument passed. What exactly does this * do? 09:03
moritz it means "bind any number of arguments to an array @name" 09:05
r: sub f(@a) { say @a.perl }; f(1, 2);
p6eval rakudo e6e158: OUTPUT«===SORRY!===␤CHECK FAILED:␤Calling 'f' will never work with argument types (Int, Int) (line 1)␤ Expected: :(@a)␤»
moritz r: sub f(*@a) { say @a.perl }; f(1, 2);
p6eval rakudo e6e158: OUTPUT«Array.new(1, 2)␤»
moritz see the difference? :-)
mtymula89 Ok so nobody knows howmany elements of an array are there untill the sub is salled upon? 09:06
Yes i see it now. 09:08
felher Can one use 'perlbug' to submit bugs to Rakudo? 09:23
moritz no 09:24
felher k, thnx
jnthn morning o/ 09:58
FROGGS_ morning 09:59
jnthn will has another tuit-limited week, but December should be better
sorear morning jnthn 10:00
jnthn o/ sorear
sorear same situation I think
exciting times :D
jnthn Early Decemberish I hope to be pushing some fledgling QAST => JVM stuff. 10:02
FROGGS_ jnthn: is there a spec about how the grammar works? I hit a bug in rakudo/src/Perl6/Grammar.pm and I need to capture more to make a better exception message
felher o/
moritz FROGGS_: extra captures are generally OK
sorear jnthn: \o/ 10:03
jnthn FROGGS_: Well, it's a subset of normal Perl 6 grammars
sorear jnthn: early Decemberish is also when I'm probably going to be able to start the niecza C-backend experiments too :D :D
moritz FROGGS_: STD.pm6 is kinda standard, but if we need more stuff than STD captures, that's OK 10:04
sorear you're always allowed to add new stuff to STD if it improves the error messages and doesn't slow us down too much
kresike hello all you happy perl6 people 10:06
moritz jnthn: do you have some time to discuss sink stuff? 10:10
jnthn moritz: I sorta got time while @student here to exercises, but may need to go give help at any moment 10:11
*do
moritz jnthn: ok
r: role A { }; nqp::say nqp::does A, 'foo'
p6eval rakudo e6e158: OUTPUT«===SORRY!===␤Error while compiling block : Error while compiling op call: Error while compiling block : Error while compiling op say: Error while compiling op does: No registered operation handler for 'does'␤»
moritz r: role A { }; nqp::say nqp::can A, 'foo'
p6eval rakudo e6e158: OUTPUT«1␤»
moritz that's the first problem 10:12
r: module A { }; nqp::say nqp::can A, 'foo'
p6eval rakudo e6e158: OUTPUT«No such method 'foo' for invocant of type 'Any'␤ in any find_method at src/gen/Metamodel.pm:149␤ in block at /tmp/6cIyHASplD:1␤␤»
moritz that's the second problem
jnthn uh, second one is wtf
moritz currently I emit the equivalent of .sink() if nqp::can($_, 'sink') for sink context 10:12
and those two make the whole thing blow up
jnthn yeah, trouble is, a role's find_method always returns a punner 10:13
Probably need to be a bit smarter on that.
The other one is just an outright bug
moritz where? in the code emitting, or in can?
FROGGS_ thanks you both, gtg to work, but will read logs 10:14
jnthn I think needs a fix somewhere in RolePunning.pm or whatever it's called
In src/Perl6/Metamodel
jnthn We can can on the pun. 10:14
moritz what a pun :-) 10:15
jnthn :D
sorear sleep &
felher o/ sorear
jnthn o/ sorear
huf if you continue this much further, someone's gonna take some elephants accross the alps
jnthn The module one is probably something throwing that should not throw but instead return a null/type object 10:16
ok, back to teaching... 10:19
FROGGS back for good 10:49
moritz *sing* I just want you back for gooooood *sing* 11:00
FROGGS ya, lets unbutton ourselfs 11:02
FROGGS .oO( imagines a bunch of geeks with bare nipples... ) 11:03
jnthn FROGGS: I...don't think I'm going to adopt that style while presenting at Perl confs :P 11:18
FROGGS well, the ppl would keep you in mind though 11:36
jnthn FROGGS: I try to do that by giving good talks :P 12:53
...but it's good to have a backup plan :P 12:54
mtymuls . 13:02
[Coke] yawns. 13:41
mtymuls :P 13:42
PerlJam good cyber-monday everybody! 13:57
dalek p/x_backslash: e9ab312 | (Tobias Leich)++ | src/ (2 files):
add indirection for 'Unrecognized backslash sequence'
 13:58
FROGGS r: "\u" 14:05
p6eval rakudo e6e158: OUTPUT«===SORRY!===␤Method 'sorry' not found for invocant of class 'Perl6::QGrammar+{qq}+{stop}'␤»
dalek p/gh67-threads: 756d06b | rurban++ | / (2 files):
add t/nqp/67-threads.t, fix .gitignore for /nqp
 14:56
[Coke] FROGG's exception there looks notgood. 15:07
FROGGS [Coke]: already fixed (pull request) 15:09
rindolf FROGGS: hi. 15:13
[Coke] FROGGS++
FROGGS hi rindolf 15:17
thanks [Coke]
rindolf FROGGS: what's up? 15:23
FROGGS: you might be interested to know that I worked on PySolFC the other day.
Well, yesterday actually. 15:24
FROGGS hi, $work and a bit bug hunting / error message improving... 15:25
rindolf: this PySolFC looks quiet nice, what did you do? 15:26
skids Wait, there's $work for solitaire? I must have missed that booth at the job fair! 15:29
japhb skids: Wouldn't that be Zynga? Although I guess that would be "social solitaire" ... 15:34
rindolf FROGGS: I made sure every module can be loaded (or "import"ed in Python-speak) separately. 15:47
FROGGS ahh, cool 15:48
rindolf skids: I wasn't using "work" in the exact literal meaning. 15:56
rindolf clubs skids senseless with the server hosting en.wiktionary.org .
Or the entire cluster of them. 15:57
sergot hi o/ ! 16:00
FROGGS hi sergot 16:01
rjbs- eeLag is my enemy. 16:11
Evidenced, for example, in how this message happened in the wrong channel. :-/
moritz rjbs++ # autopun 16:12
rurban FROGGS: #ifdef enum_class_Proxy is wrong. It is a enum not a preprocessor define 16:13
I'm fixing it in the branch gh67-threads 16:14
FROGGS rurban: k, thank you 16:16
rurban For now no problem, because it will always be false. But I try to add proper threads support to nqp 16:17
[Coke] finds rdstar.wordpress.com/2012/11/22/the...partially/ , which appears to have no author. 16:29
(and isn't on planet sixperl) 16:32
FROGGS well, when seeing the snd link I'd say lue posted it 16:33
[Coke] danke. 16:38
phenny: ask lue if he wants rdstar to be added to the planet sixperl feed.
phenny [Coke]: I'll pass that on when lue is around.
[Coke] lue++ 16:41
moritz with less than two years of age, my daughter is now a youtube consumer 16:51
she has her favorite video which she demands to see occasionally 16:52
and then she selects second-best favorite videos by pointing to the preview images
though we don't have a touchpad, and she doesn't know how to use the mouse yet, we must click for her 16:53
FROGGS moritz: ya, the mouse must be something oldschool and non-intuitive... 16:56
my sons, almost 3 and the other 19months can both skip through images on my android, and they always want to click on thing on my laptop ;o) 16:57
but the older one can now use a touchpad too, even if he is very slow with it
moritz erm, s/touchpad/touch screen/ above 16:59
moritz wonders if he should cut a star release 17:00
FROGGS wonders if he should test all these modules, since he didnt do it for ages 17:01
[Coke] (kids) at one point we found a trackball helpful 17:02
(my thumb hurts just thinking about it. :)
masak greetings, #perl6. 17:18
japhb Greetings, Earthling!
FROGGS masak o/ 17:23
see y'all at home
masak I think rt.perl.org:443/rt3/Ticket/Display...?id=115916 is a dupe 17:25
also, does our RT instance suddenly have a port number?
moritz \o masak 17:26
masak: in some links it has for months at least
masak huh.
dalek ar: aa7e45c | moritz++ | docs/announce/2012.11:
announcement for 2012.11 release
 17:29
ar: 35d12ef | moritz++ | / (2 files):
bump rakudo and nqp versions
ar: dda13ae | moritz++ | modules/ (11 files):
bump module versions
moritz everybody, please review the release announcement, and fix (preferrably) or point out any mistakes/omissions 17:30
dalek ar: b9b42ec | moritz++ | tools/build/Makefile.in:
forgot a version bump (Makefile.in)
masak moritz: I reviewed the release announcement -- didn't see anything out of the ordinary. looks good. 17:32
moritz masak: thanks
masak "advanced macros" is a little vague, but I don't have a better suggestion.
moritz masak: just advance them :-) 17:37
PerlJam Are we doing monthly releases of R* now? 17:40
masak making macros suck less is a prerequisite to advancing them, I think... :)
moritz PerlJam: currently, yes 17:41
PerlJam moritz: why is that? The recent feature list has been "impressive enough"? 17:43
moritz PerlJam: mostly yes (more) 17:48
PerlJam: first of all we didn't announce that we'd switch to a longer release cycle, so people expect a new release
PerlJam: secondly we have some many nice features that it would be a shame not to get them to the users quickly 17:49
hey, heredocs! Operators with custom prec and assoc 17:50
and, quite important IMHO, precompilation fixes
jnthn++
masak indeed. jnthn++
dalek ast: c196552 | moritz++ | S (5 files):
rakudo autounfudges
 17:51
PerlJam I didn't realize that we hadn't announced the longer release cycle. 17:52
moritz PerlJam: speaking just from memory, I think that only before the switch to nom did we have quarterly releases 17:53
PerlJam: and ever since monthly releases
PerlJam weird. I guess I just had "quarterly" stuck on my brain for Star releases. 17:55
jnthn evenin o/
PerlJam greetings jnthn++
jnthn oh, moritz++ is on with the Star release 17:57
That's good; I'm pretty tight for time this week.
moritz moritz.faui2k3.org/tmp/rakudo-star-...RC0.tar.gz # release candidate 18:02
jnthn moritz: Maybe worth mentioning quote adverbs too? 18:07
And perhaps explicitly call out the pre-compilation fixes... 18:08
need to go buy some noms...will commit those when back if you don't beat me to it... 18:09
dalek ar: 3a2846f | moritz++ | tools/star/release-guide.pod:
[release guide] add list of previous releases
 18:14
moritz jnthn: beat you to it :-) 18:16
dalek ar: d4e9c5d | moritz++ | docs/announce/2012.11:
[announce] mention quote adverbs and precompilation fixes
moritz I'll wait with the R* release until tomorrow; maybe soembody has more feedback until then 18:20
if not, I'll make a release code-identical to the RC0, but with updated release annoucement 18:21
phenny: tell FROGGS if you want to do more typed exception in nqp, you can do them directly in master; they are low risk, and don't break rakudo 18:23
phenny moritz: I'll pass that on when FROGGS is around.
moritz now if only I could build nqp again on my laptop... 18:24
dalek rl6-roast-data: 1da7959 | coke++ | / (3 files):
today (automated commit)
 18:32
[Coke] sorear++ - niecza back down to 15 failures.
masak moritz++ # cutting R* 18:42
moritz is a bit surprised by panda commit 1361fcbc0e9d298b38f80c3fb1068f034205e532 -- does it work to use a module in bootstrap.pl? if yes, why? 18:43
it does not 18:45
===SORRY!===
Could not find Shell::Command in any of: ...
dalek nda: a5f2ff8 | moritz++ | bootstrap.pl:
unbreak bootstrap.pl
 18:46
moritz ponders how to best get panda to use %*CUSTOM_LIB 18:47
is there a cross-make compatible way to execute a command and put the result in a variable? 18:49
given what I know about the portability of Makefiles, I guess the answer is "no" 18:50
[Coke] I would probably run the command and put the output in a file as a stopgap. 18:51
moritz this feels more like (not so) expressive dance than programming 18:53
japhb moritz, writing portable makefiles will often cause one to express oneself, though "dance" is not how I would tend to describe that expression. ;-) 18:57
moritz
.oO( expressive non-dancing )
 18:57
r: say @*INC.join(' ')
p6eval rakudo e6e158: OUTPUT«/home/p6eval/nom-inst/lib/parrot/4.8.0-devel/languages/perl6/lib /home/p6eval/nom-inst/lib/parrot/4.8.0-devel/languages/perl6/vendor/lib /home/p6eval/nom-inst/lib/parrot/4.8.0-devel/languages/perl6/site/lib /home/p6eval/.perl6/2012.11-4-ge6e1585/lib /home/p6eva…
japhb Is it just me, or does p6eval seem to cut output to a shorter length than the IRC server normally would? 18:59
moritz yes, it has its own cutoff
japhb It seems too conservative.
masak I can't think of a case where I felt it was too short. 19:00
japhb Huh. I feel that rather often. :-/
FROGGS talking about sex?
phenny FROGGS: 18:23Z <moritz> tell FROGGS if you want to do more typed exception in nqp, you can do them directly in master; they are low risk, and don't break rakudo
FROGGS will read logs now
moritz: k, will do 19:01
masak the things that hit the cutoffs tend to be errors/exceptions, and they tend to be rather void of meaning after line 1.
masak I say that as someone who strips a lot of useless stacktraces in bug reports. :) 19:01
japhb masak, I see it often when someone is trying to get a method list from some class, or in this case see all of @*INC.
Oh, I can definitely understand the desire to avoid a mess. 19:02
moritz well, the full method list from a class is often too long for IRC anyway :-)
japhb Maybe treat the cutoff for STDERR and STDOUT distinctly?
moritz ETOOMUCHEFFORT 19:02
FROGGS masak: 443 is the default port for https/ssl connections
japhb heh
masak FROGGS: ah.
masak I'd rather p6eval spam the channel less. 19:02
moritz r: say Int.^methods(:all).elems 19:03
p6eval rakudo e6e158: OUTPUT«217␤»
masak if someone wants a long listing of stuff, they can use their local terminal window, not #perl6 :)
japhb I figured masak new that and was just commenting on the redundant URL formatting ...
japhb er knew 19:03
moritz thought so too
masak no, I just suddenly noticed the port in the URL.
japhb moritz, granted, that ^methods(:all) is going to be long. But I often see ^methods(:local) cut off when it doesn't need to be. 19:04
masak I know very little of ports. perhaps I knew at some point that 443 means https/ssl, but I definitely didn't remember.
r" say Int.^methods(:local)
r: say Int.^methods(:local)>>.name 19:05
p6eval rakudo e6e158: OUTPUT«Int Num Rat FatRat abs Bridge chr sqrt base expmod is-prime floor round ceiling sign conj rand sin asin cos acos tan atan atan2 sec asec cosec acosec cotan acotan sinh asinh cosh acosh tanh atanh sech asech cosech acosech cotanh acotanh unpolar cis Complex log exp …
japhb But in any case, no worries. Does the cutoff also apply to PM with p6eval, or just channel responses?
moritz also pm
japhb Okie-doke.
moritz there's not much engineering in p6eval, except the bare minimum needed to provide the functionality we need here 19:06
tadzik bah, so I _did_ break panda bootstrap. Neh 19:08
moritz tadzik: never mind, I fixed it 19:09
tadzik yeah, seen it. moritz++
I need vacation :)
masak .oO( "vacation" has to do with looking at cows, right? ) 19:10
just like a "vacant look" has to do with looking like a cow. 19:11
:P
FROGGS hehe
masak (shamelessly plagiarized from, I think, MJD's twitter feed) 19:15
FROGGS thats fine, there is not much left to invent in the year 2012 19:16
moritz: did you had a chance to look at my pull request? github.com/rakudo/rakudo/pull/91 19:17
FROGGS don't wanna annoy you though :/ 19:19
masak FROGGS: I used to think that was the case; however, with each year that passes, I realize how the whole CS and PL thing is still in its infancy, or at least young adolescence.
skids
.oO(me decides not to ponder other meanings of "I need vacation.", especially with cows in the picture.)
 19:20
masak .oO( cow tipping ) 19:21
<popl> Leven*shteen*! 19:26
popl: L*o*venshteen! :P
FROGGS masak: for a non-native-english-reader, whats the CS and PL thing? 19:30
moritz: wouldnt it make sense to drop 'make test' or mention the other two option in the first step? paste.scsys.co.uk/215400 19:31
masak FROGGS: "Computer Science", and "Programming Language"
FROGGS ahh
ya, tru 19:32
masak ...which is part of why I like being involved with Perl 6. 19:33
FROGGS and I am a community supporting rules loving german guy 19:34
and Perl6 is pretty awesome
masak .oO( "[community supporting rules] loving" or "[community supporting], [rules loving]"? ) :P 19:35
FROGGS hehe 19:41
tadzik parsing people is hard
I once signed up for "Advanced Programming in C++"
I expected (Advanced Programming) in C++
got Advanced (Programming in C++)
moritz FROGGS: the current factoring of the messages is not bad because it doesn't flood the user with information 19:42
FROGGS moritz: alright 19:43
dalek nda: 5ae49a8 | tadzik++ | / (2 files):
Fix installing random files from lib/
 19:49
nda: 6b9161c | tadzik++ | testmodules/compiled__module/blib/lib/baz.js:
Add forgotten file for tests
nda: 4c73acb | tadzik++ | / (3 files):
Make Panda::Builder copy over all the files it doesn't recognize
tadzik phenny: tell azawawi I told panda to install random files from lib as well :) Hope this suits you
phenny tadzik: I'll pass that on when azawawi is around.
tadzik phenny: thanks
FROGGS moritz: rakudo-test and rakudo-spectest passes, but should modules-test work? 19:54
sorear good * #perl6
moritz FROGGS: it should, but I don't think it did in the last few releases 19:55
FROGGS open3: exec of ... failed at /usr/share/perl/5.14/TAP/Parser/Iterator/Process.pm line 168.
moritz t/custom-headers-and-content.t .. Cannot call 'match'; none of these signatures match: 19:56
:(Cool:D : $target, *%adverbs)
while testing perl6-lwp-simple
moritz Bailador fails t/05-post-content.t 19:57
tadzik oh noes 19:58
tadzik I merged 2 pull requests today trusting them to be good 19:58
erm, merged 3
tadzik and looks like 2 of them broke stuff 19:59
FROGGS tadzik: paste.scsys.co.uk/215403 20:00
masak netsplits are useful for remembering who's lurking on the channel. 20:03
tadzik hm 20:04
tadzik I'm not sure what that is :) 20:04
moritz FROGGS: might be a brokenness of the test harness 20:05
I'm pretty sure it's supposed to execute perl6 $testfile, not just $testfile
FROGGS ohh, hold 20:06
on
moritz jsonrpc fails t/server.t
FROGGS I didnt make install -.-
moritz and I failed at reading the error message 20:07
FROGGS ;o)
tadzik hrm. I wonder, what to do to LWP::Simple to marry it with MuEvent 20:14
I need to somehow say it "you know, when you call recv(), you should basically set a MuEvent handler. That'd be cool"
without rewriting half of it
dalek o: 0755ad0 | moritz++ | bin/ufo:
install to any(%*CUSTOM_LIB<site home>)

This takes the easy and portable route by writing the prefix into the Makefile, at the expense of having to re-run ufo whenever that prefix changes -- which is quite often in the case of the home lib.
moritz masak: I hope you're OK with that change; it's the least insane solution I could find :-) 20:15
tadzik being able to convince it to use my kind of IO::Socket, I could maybe preempt it somehow inside those read()s
I could maybe emulate this sort of behaviour using gather-take
so many "I could maybe" 20:16
masak moritz: on phone; will look later. 20:20
moritz masak: ok, no hurry
FROGGS k, most of the module-tests are passing, the failing ones you already posted have </yoda> 20:21
moritz FROGGS: it's your comedy day, isn't it? :-)
FROGGS I'm just in a good mood ;o) 20:22
sitting in the dark with laptop and my almost sleeping kid...
moritz that's quite a feeling, I know :-) 20:23
moritz $daugher has been sleeping for an hour 20:24
FROGGS ya, I can be lucky most days, both are sleeping within 10 minutes usually
think he is drifting away now... 20:25
so, shhhhhh
:o)
moritz hopes FROGGS' IRC client doesn't make a sound for each incoming line :-) 20:26
FROGGS just dont use the BELL ascii char
masak r: say "\b" 20:26
p6eval rakudo e6e158: OUTPUT«␤»
masak :D
moritz would like to visit the BELL labs
FROGGS \b? really? 20:27
masak ya rly.
FROGGS I'd like to visit Xerox and tell them that they are stupid
masak no wait
r: say "\b"
p6eval rakudo e6e158: OUTPUT«␤»
masak r: say "\a"
p6eval rakudo e6e158: OUTPUT«␤»
masak there we go. 20:28
FROGGS r: say "\u"
p6eval rakudo e6e158: OUTPUT«===SORRY!===␤Method 'sorry' not found for invocant of class 'Perl6::QGrammar+{qq}+{stop}'␤»
masak r: say ord "\a"
p6eval rakudo e6e158: OUTPUT«7␤»
FROGGS ;o)
masak submits rakudobug
doy heh, irssi doesn't strip ^G
that's... kind of annoying
FROGGS masak: nooooooo!
masak r: say ord "\a\a\a\a\a\a\a" 20:29
p6eval rakudo e6e158: OUTPUT«7␤»
masak r: say "\a\a\a\a\a\a\a"
p6eval rakudo e6e158: OUTPUT«␤»
skids hush.
masak r: say "\a\a\a\a\a\a\a" x 50
p6eval rakudo e6e158: OUTPUT«…
tadzik eek
masak hides 20:30
doy actually, it does something weird, it's not just "not stripping"
so it has to be intentional behavior?
weird
masak what does it do? 20:31
doy whenever it receives \a, it generates a bell, even if the window isn't active, and switching back and forth to the window doesn't generate bells 20:32
masak s0rear! \0/ 20:41
masak moritz: github.com/masak/ufo/commit/0755ad0ddb looks good 20:45
moritz \o/
masak I have a general feeling of unease related to the commit, but it's not your fault. 20:46
it's just... it feels like the problem being solved is one I can't get a good overview of.
masak all I see is gradual patch-size solutions like this one going in over the years. 20:47
moritz you mean that ufo is slowly growing, with neither a perspective to becoming smaller again, nor having a real design? 20:48
masak oh, I wasn't really thinking of ufo here. 20:49
the lack of design there is probably because I'm kind of an absentee original author.
I was thinking of Rakudo/Perl 6 and library handling.
and it's tricky because I don't really know of a better way to solve this than gradual evolution. 20:50
[Coke] sees MuEvent go by in the scrollback and realizes that he never resurrected cjbot.
[Coke] wonders if it's worth doing until we have a better event loop. 20:51
tadzik not sure
if the issue of it hanging with no reason is till there, then No 20:52
masak .oO( NoEvent )
tadzik if it's not, then we still have the issue of busy waiting, so Probably No
masak what's the issue of busy waiting, and is there a solution in sight? 20:53
FROGGS I'd say if a event loop is busy when waiting than it is missing a delay... 20:54
moritz it's missing a callback :-) 20:55
tadzik moritz: we'll need a poll() that works on more than one socket :)
masak ah.
moritz .oO ( 25.media.tumblr.com/tumblr_m8f57168...1_1280.png )
masak yeah, not having that sounds pretty limiting. 20:56
FROGGS moritz: and I know what episode that is ;o)
.oO( should buy a horgan just to be sure )
 20:57
moritz speaking of which, today in the subway I saw a maybe 13 year old girl reading 50 Shades
I'm pretty sure I didn't read stuff like that when I was 13 :-)
FROGGS I believe there were no such things when we were 13 20:58
and I didnt like books much at that time
jnthn 50 shades of grey? Isn't that the Visual Studio 2012 default color scheme?
moritz :-) 20:59
FROGGS ya, lets call it "color" scheme ;o)
moritz you mean THAT SHOUTY PROGRAM?
jnthn :) 21:00
jnthn I had to use it in a talk a couple of months back. The talk was somewhat memorable for by reference to the DEBUG (I actually shouted it) menu :) 21:01
masak THE_MENUS_ARE_ALL_CONSTANT 21:03
FROGGS maybe MS should hire some designers? to round up the edges of their life-tiles or what it is called and to introduce ucfirst on menus
the thing is, even if you want to use their software (windows phone development tools with VS2012 express), then after installing windows 7 and its sdk for a day in a VM you get stuck when they wanna have 37$ just to let you copy the examples to your phone 21:05
-.-
wasted a day because of that
tadzik maybe they should hire some logical thinkers :) They still didn't respond to my email "how do I bootstrap a windows dvd image" 21:06
I get free copies of windows as a student, but they don't come as .iso files, but as an .exe which downloads an .iso
I asked them if they expect me to come over to piratebay and get a windows there so I can download my legal copy 21:07
moritz try wine :-)
tadzik no, it didn't work :)
it used to, but they fixed something with their downloader and now it doesn't ;)
masak becoming an alcoholic is seldom the answer.
popl No, it didn't work the first time I tried it either tadzik.
I was in a similar situation. 21:08
tadzik oh, they respond to emails eventually?
well, I run a piratebay copy anyway; torrent uploaders are capable of making an install image that installs on an USB disk, microsoft apparently isn't
maybe it's better with windows 8 21:09
popl tadzik: I was referring to using Wine. 21:11
tadzik ah, I see
spider-mario didn’t work for me either 21:12
masak sorear++ # v24 21:28
looking forward to v42 ;)
also looking forward to what the object sharing will lead to. 21:29
japhb_ After v42, he clearly will need to do a v42.bis release. 21:33
[Coke] sounds like one of the refactors jnthn did recently for rakudo. 21:34
jnthn Rakudo had this model since the nom days. 21:36
(As in, part of what nom did was this)
convergence++
[Coke] part of the announcement was sharing compiler/runtime - that was a more recent update, aye? 21:37
jnthn [Coke]: Maybe I'm misunderstanding, but I understood it as allowing the compiler to work against the real objects that will exist at runtime. 21:38
jnthn Having a good way to do that was one of the biggest architectural differences between b and nom. 21:39
[Coke] I'd expect you to understand it better than i. :)
FROGGS r: say "abc\0def"; say "abc\x00def" 21:51
p6eval rakudo e6e158: OUTPUT«abc␀def␤abc෯␤»
FROGGS ---------------------^ this should be the same, right?
sorear jnthn: I did that part earlier, what's happening now is that the compiler can treat them as real objects 21:52
FROGGS it shows \xEF for \x00
sorear jnthn: before, they had to be wrapped in special handles and accessed using dedicated opcodes
soon we'll be able to use real mthods
[Coke] r: say "abc{\x00}def" 21:53
p6eval rakudo e6e158: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&x00' called (line 1)␤»
[Coke] r: say "abc\x{00}def"
sorear r: say "abc\x[00]def"
p6eval rakudo e6e158: OUTPUT«===SORRY!===␤Method 'sorry' not found for invocant of class 'Perl6::QGrammar+{qq}+{stop}'␤»
rakudo e6e158: OUTPUT«abc␀def␤»
FROGGS ahh, I'm wrong, it is parsing the whole as hex
[Coke] thinks he found another error for FROGGS to fix!
sorear jnthn: wonderful rakudobug :D
jnthn sorear: Which one? 21:54
Huh...that's odd.
FROGGS the 'sorry' thing is fixed (in a pull request)
jnthn should probably be panic, yeah
masak I submitted that one an hour or so ago.
FROGGS right
jnthn Like I'm paying attention :P
FROGGS someone need to have a closer look at my pull request and should merge then ;o) 21:55
jnthn OK
FROGGS masak: but not to RT, right?
cant find it
jnthn I'm gonna sleep soon, but I can look tomorrow probably 21:56
FROGGS jnthn: no hurry & sleep well
(... soon)
masak FROGGS: I... I forgot. :/ but I did it just now. should be there now.
[Coke] wants a perl6 RT bot. 22:05
tadzik hm 22:05
[Coke] will update cjbot to do it if the aforementioned event issues ever get resolved.
(or tadzik can do it, that's fine! ;) 22:06
tadzik I'm megalazy these days :)
haha. My Panda is just installing MuEvent using p6prove to test it. So cool 22:07
masak :) 22:08
tadzik hrm. Does "Invalid operation on binary string" when dealing with recv()d data ring a bell to you? 22:13
masak yeah. 22:15
probably getting sub-character stuff.
tadzik so, Net::IRC fails due to that :) 22:20
masak :/ 22:26
masak 'night, #perl6 23:06
japhb_ Heh, even though sorear++ claimed he was going to do it *after* the v24 release, I note that the only two perl6-bench microbenchmarks in which niecza v22 and v24 perform differently are the while and for variants of the concat benchmark, where v24 is 13-15% faster. Slipped something in, did you sorear? ;-) 23:38
sorear japhb_: I don't know what you're measuring, concat's implementation hasn't changed since october 9 ... 23:42
2010
japhb_ Weird. 23:45
Now I wonder why the difference ....

Powered by Cro and the Raku® Programming Language.

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

Thank you!