»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.
00:07 sudokode joined
[Coke] colomon: lots of "declared but not used" 00:09
colomon [Coke]: YES
er, yes 00:10
they seem to only happen the first time Niecza compiles the modules, I think.
[Coke] ah, only on initial compile?
ah.
error: failed files: "wedding.ly" 00:12
colomon huh? 00:13
[Coke] I compiled wedding.abc -> wedding.ly
colomon ah.
[Coke] I then ran wedding.ly through lilypond.
boom. 00:14
colomon and it failed?
can you post the wedding.ly you got in a gist?
colomon just checked, and wedding.abc -> wedding.ly does work in his Lilypond. 00:17
huh. though there are a lot of error messages cropping up in Lilypond's output window I never noticed before. 00:18
[Coke] feather.perl6.nl/~coke 00:19
GNU LilyPond 2.14.2
colomon has a sudden horrible feeling that Lilypond might be stupid enough to just display the old wedding.pdf file sitting around if generating a new one doesn't work....
[Coke] put up an "errors" in the same dir. 00:22
00:23 _jaldhar left, mikemol joined
colomon yes, I get the exact same wedding.ly file. 00:24
Arrrgh, so sorry for this, let me see what I can figure out here.
ooo, I think I see what's wrong with the wedding.ly file 00:25
[Coke] if it's any consolation, my son is aghast that he might have to use a text editor. ;) 00:26
colomon grump, my MBP is running slow tonight, too.
if you search and replace "" in the .ly file with " it should work fine.
now to track down the bug.
colomon must have broken it when he was working on ABC file export a few weeks back 00:27
00:30 tokuhirom left, tokuhirom joined
colomon [Coke]: just pushed a bugfix. 00:32
[Coke] \o/ 00:34
I'm going to tell Nathaniel how to file bug reports, btw. watch out. :)
colomon sweet! would be wonderful to have someone else poking at it. 00:35
00:36 _jaldhar joined
[Coke] is there a way to specify the output file? 00:37
nevermind, don't need it. 00:40
colomon you can do it with pipes... <wedding.abc >wedding.ly 00:42
I suppose I could add a -O option or something like that.
00:43 Telgalizer left
sorear o/ colomon, [Coke] 00:46
00:47 Exio joined 00:48 driador left, driador joined 00:49 ocharles left
colomon sorear, o/ 00:51
00:52 ocharles joined
[Coke] ugh, what a mess the wrapper script was. 01:03
but it works, he can now use TextEdit to edit an abc file, drop to terminal, and generate a pdf. 01:04
01:05 xinming_ left 01:06 xinming joined
[Coke] colomon: do you want to activate "issues" on ABC's github? 01:06
colomon Wynne:ABC colomon$ cat /Users/colomon/bin/abc2ly 01:07
mono /Users/colomon/tools/niecza/run/Niecza.exe -I/Users/colomon/tools/ABC/lib /
is the script I use. :)
Issues activated, I think. 01:08
(I mean, now they are.)
[Coke] it's messy because he's not me, and I can't install it.
gist.github.com/2985840 - colomon 01:10
colomon interesting. whenever I want a script that complicated, I always write it in perl. ;) 01:11
[Coke] I used to be a sysadmin, so I use the bash sometimes. 01:12
plus, one less thing to config for him.
bad enough I didn't make it a shiny icon he could click on. :)
colomon :) 01:20
01:24 scott_ joined, scott_ is now known as Guest20991 01:32 scott___ joined, raiph joined, Guest20991 left 01:34 BaggioKwok joined 01:35 fgomez left 01:42 tokuhirom left 01:44 Guest78621 left 01:54 fgomez joined 01:58 BaggioKwok left
colomon .u & 02:05
phenny U+0026 AMPERSAND (&)
02:09 _jaldhar left 02:11 _jaldhar joined 02:22 orafu left, orafu joined 03:02 Telgalizer joined 03:14 fgomez left 03:25 vlixes joined 03:34 raiph left
[Coke] # 06/24/2012 - rakudo++ (22459); niecza (91.12%); pugs (34.5%) 03:35
rakudo edging slightly forward.
03:42 Telgalizer left, Telgalizer joined 03:46 kaleem joined 03:48 jimmy1980 left 03:57 sisar left 03:58 sisar joined 04:16 adu joined 04:32 Telgalizer left 05:17 _jaldhar left 05:29 birdwindupbird joined 05:33 leprevost left 05:35 lue joined
lue blog post! rdstar.wordpress.com/2012/06/25/th...p-on-llvm/ 05:35
also, hello again o/
05:35 quietfanatic left
sorear LUE! 05:35
05:35 GlitchMr joined
sorear 's been ages! 05:36
lue hello sorear o/
sorear jump-hugs lue
lue XChat said "Loaded log from Wed Mar 28" O.o
05:39 odoacre joined
lue I feel I should mention modules.perl6.org is currently telling me no-one test, documents, or updates their modules. 05:41
*tests
moritz \o 05:42
lue hello moritz o/
05:43 aloha joined 05:45 fhelmberger joined
adu oOo 05:46
05:46 wtw joined, wtw left, wtw joined
lue hello adu o/ 05:48
adu lue: hey
who wrote the blog post? 05:49
lue the one I just linked to? Me. 05:50
adu good job! 05:51
I've also considered LLVM 05:52
lue thank you.
adu however, I was considering compiling PBC=>LLVM
lue That's sounds good, as well as simpler (but then I'd still have to compile CORE.setting to PBC if I'm not mistaken :P) 05:53
adu doesn't that happen already? 05:54
lue Once rakudo is put entirely on NQP, I don't think rakudo will need to compile to PBC anymore (esp. if NQP is ported to another backend) 05:56
It's been a while since I've tried compiling rakudo locally, but I'm fairly sure I'm unable to get past the PIR->PBC (or NQP->PIR?) stage for CORE.setting 05:58
adu oh right
bonsaikitten lue: works quite well for me, rakudo+nqp are really benign packages to build
adu my house might blow away in a tornado 05:59
just fyi
lue I have an incredibly old PowerBook laptop with 384 MiB RAM :) [oh, and 1GiB swap space]
bonsaikitten ouch
lue: you should upgrade to this decade ;)
lue adu: is the tornado close?
adu I don't know about tornado, but I hear wind right now, and my house is creaking, and Friday there was an neighborhood about 5 miles away that was completely destroyed 06:00
lue If you'd like to donate to the fund... :) (just kidding) hopefully soon I will upgrade. (I got this computer for free though, so I'm not complaining too loudly) 06:01
adu lue: do you have other computers?
you should set yourself some alerts/filters for computer shows in your area 06:02
lue adu: That's not good. (I get very nervous about high winds, because we live in a forested area with tall pine trees.)
adu but I still have power, for now :)
lue There are certainly other computers, but they all run Windows :) . I currently have a list of parts for a computer I can put together (~$800, or just under $600 sans non-computer parts (monitor etc.)) 06:03
adu lue: I've seen pictures of metal roofs penetrated by a blade of hay, wind can be serious 06:04
lue Yes, the power of wind is why I tend to feel paranoid on very windy days. 06:05
adu but this pic makes me feel so much better about natural disasters: wheeeeeeee.com/wp-content/uploads/2...003301.png 06:07
lue :) Can you imagine the horror if someone was in that chair, *doing things*‽ 06:09
adu lol
but I think pbc=>llvm might be useful independant of perl6 06:10
sorear lue: get a job :p you remember my old machine, right?
adu gives lue some cake on a plate
sorear lue: I just signed on with a tiny perl shop and one of the perks is a brand new macbook air 06:11
lue Not quite (but I imagine it was just as bad).
sorear 384mb ram, 2003 Pentium4 1 core, etc
lue Ooh, that's nice!
adu heh
my last 3 laptops have been macbooks
before that was a PowerPC, and before that was a Quadra 06:12
lue (sorear: At least your old machine had an x86 CPU) 06:13
sorear the only laptop I currently have is a Hitachi thingy from 1991
shachaf sorear: Oh, you replaced your old machine!
adu btw, Quadra=m86k
shachaf remembers the horror stories.
sorear if it were a year older it would be older than me.
shachaf: not delivered yet.
shachaf: a few days yet.
but it's been promised to me in writing
adu lue: why SFML? 06:14
shachaf Ah.
sorear quatra = 680*4*0
bonsaikitten sorear: I don't see the MBA as perk ;) 06:15
lue adu: 1) C++ 2) I hope I had a better reason that just C++ :)
bonsaikitten: me too, being a linuxer, but at least it's free!
bonsaikitten lue: but ... it's unusable :( 06:16
06:16 GlitchMr left
bonsaikitten lue: can I interest you in free prostate exams? ;) 06:16
adu that's terrible! 06:17
lue (why the SFML tutorials (as well as many graphical programming ones) is curious)
sorear it runs Unix, that's all I really insist on
lue Well, I'd personally s/OS X/Linux/ if possible (worked on this laptop :P)
sorear yeah, I'll try dual-booting 06:18
but I won't whine if it fails
adu wtf is code.google.com/p/wxgo/ 06:19
lue My one problem with OS X (which wasn't what spurred me to install Linux though) was that I had a difficult time doing programming things (e.g. XCode)
[Mind you, this was long before I even started using python, IIRC] 06:20
adu lue: XCode was designed to sell tickets to workshops at ADC
bonsaikitten my biggest problem with macbooks is that their peripherals suck badly
adu or WWDC
lue adu: wxgo appears to be a port of wxwidgets to the Go language. 06:21
adu why isn't it.
sorear ok, everyone: *I* didn't start this OS holy war. Drop it. 06:22
adu sorear: what? war?
sorear next time this comes up, I'll tell people I have a brand X laptop
lue adu: smells like vaporware to me (someone probably started the page and got bored with the idea 2 days later)
adu sorear: I'm not interested in oswars 06:23
sorear: I'm also a mac user, but it doesn't mean I have to like Apple
06:23 Unik joined
bonsaikitten sorear: not about OS ;) 06:23
lue I do have to say that OS X is at least Unix and comes with the normal terminal stuff I'm now used to, like emacs (all helpful when writing up papers on the school's Macs, and insisting on using XeLaTeX :P) 06:24
adu my usual procedure involved deleting emacs, installing macports, then installing emacs
Unik rakudo: class A { has $.aa; method get_aa handles Real { return $.aa; } }; 06:25
p6eval rakudo 7d394e: OUTPUT«===SORRY!===␤Cannot call 'trait_mod:<handles>'; none of these signatures match:␤:(Attribute:D $target, $thunk)␤␤»
sorear I forgot to greek the brand on my company machine and now everyone is talking about OS X
adu lue: were you also looking into a GUI lib for perl6? 06:26
Unik hi everyone
was method delegation scrapped from the docs?
lue adu: through zavolaj, of course (I'd be coding it myself though).
sorear Unik: which docs?
Unik or am I doing something wrong? (probably)
synopses, I suppose
or is it just NYI? 06:27
sorear S12:1413
=head1 Delegation is still in the synopses
(do you know how to handle Sxx:yyyy synopsis links?) 06:28
Unik no, but found it anyway. thanks
so my example above is flawed somehow? 06:29
sorear this does not follow.
Unik perl6: class A { has $.aa; method get_aa handles Real { return $.aa; } };
p6eval rakudo 7d394e: OUTPUT«===SORRY!===␤Cannot call 'trait_mod:<handles>'; none of these signatures match:␤:(Attribute:D $target, $thunk)␤␤»
..niecza v18-9-g95a1946: OUTPUT«===SORRY!===␤␤Action method trait_mod:handles not yet implemented at /tmp/IgyFskkKv0 line 1:␤------> { has $.aa; method get_aa handles Real ⏏{ return $.aa; } };␤␤Unhandled trait 0 for this context at /tmp/IgyFskkKv0 lin…
sorear it's in the docs but that's not the same as being implemented 06:30
HOWEVER, your example IS flawed
'handles' only works on attributes
lue bye all o/
sorear r: class { has $.aa handles Real; }
p6eval rakudo 7d394e: ( no output )
sorear that's the correct syntax
Unik method select_tail handles <wag hang> {...} is straight from S12, no?
06:31 sisar left, kaleem left
Unik at the bottom of delegation section 06:31
sorear I guess that isn't implemented 06:36
06:37 sisar joined
Unik that's fine, i nearly lost my mind trying to spot an error in my code 06:37
also, S12 doesn't say if you can use rw methods in this way 06:39
sorear you ought to be able to, I can't think of any interaction 06:40
moritz you can do 'handles' on a method? 06:45
if you call it, does it act on the return value of the method? 06:46
or is it just an alias?
Unik S12 says it passes calls to return value of the method 06:50
but it's only mentioned once without too much detail 06:51
sisar moritz: link to Synopsis on perlgeek.de/blog-en/perl-5-to-6/16-enums.html is broken 06:59
07:00 vlixes left
moritz sisar: thanks 07:00
07:03 bbkr joined
moritz sisar: fixed 07:03
sisar :)
bbkr hi from Silicon Valley o/
sorear hello, bbkr 07:04
moritz \o bbkr
sorear say hi to TimToady for me :)
dalek ecza: 07900b5 | sorear++ | lib/CORE.setting:
Handle 0-arg [min] and [max], fixes #71
 07:08
masak morning, #perl6
sisar morning masak 07:09
sorear morning masak.
07:11 bbkr left
pmichaud rn: say [max] (Any, Any, 2); 07:13
p6eval rakudo 7d394e, niecza v18-9-g95a1946: OUTPUT«2␤»
07:14 wamba joined 07:17 brrt joined
jnthn morning, #perl6 07:19
sisar moritz: link to Synopsis on perlgeek.de/blog-en/perl-5-to-6/17-unicode.html is broken. Btw, nice post !
diakopter ahoy
sisar morning jnthn
sorear good morning diakopter. jnthn.
pmichaud good morning, briefly
(about to head off to sleep :)
dalek ast: 4ec2969 | pmichaud++ | S03-metaops/reduce.t:
Add some tests for argumentless [min] and [max].
 07:26
masak .oO( and don't call us Briefly ) :P 07:28
"Delegation" gets a green for Rakudo in perl6.org/compilers/features -- and yet Rakudo doesn't have method delegation. is that correct? 07:29
07:30 kresike joined
kresike good morning all you happy perl6 people 07:30
jnthn meeting & 07:31
pmichaud masak: I don't think we even have tests for method delegation yet. 07:33
and given that it gets only two lines of S12; I'm not sure that pushes it into yellow. You can of course disagree. :) 07:34
sorear n: $*moo = 5 07:55
p6eval niecza v18-9-g95a1946: OUTPUT«Unhandled exception: Writing to readonly scalar␤ at /tmp/VnLIF8lUMt line 1 (mainline @ 2) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3917 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3918 (module-CORE @ 562) ␤ at /home/p6eval/niecza/lib…
dalek kudo/nom: 53fc0ae | pmichaud++ | src/core/Bool.pm:
Restore [^^] and [xor].
 08:13
ast: 65b73e9 | pmichaud++ | S03-metaops/reduce.t:
[S03]: Unfudge [^^] tests for rakudo.
08:16 kurahaupo left 08:20 Gesh joined 08:23 dakkar joined 08:25 fgomez joined
sorear sleep& 08:27
08:30 grondilu joined 08:34 grondilu left 08:36 sudokode left
moritz I think the REPL and backtraces don't interact well 08:37
which is the ultimate reason for RT #113776 08:42
08:50 kaleem joined
dalek Heuristic branch merge: pushed 17 commits to rakudo/ex-hardening by moritz 09:08
09:16 adu left 09:20 fgomez left, fgomez joined
dalek kudo/nom: b9a332b | moritz++ | src/core/Backtrace.pm:
improve robustness of internal error reporting
 09:26
kudo/nom: b68a884 | moritz++ | src/core/Backtrace.pm:
be more tight about :D in invocant
kudo/nom: 57d251f | moritz++ | src/core/Backtrace.pm:
fix previous commit
kudo/nom: 2fdf4bc | moritz++ | / (11 files):
Merge branch 'nom' into ex-hardening
kudo/nom: 50774a3 | moritz++ | src/core/Backtrace.pm:
[Backtrace] it is fine to generate empty backtraces

this seems to happen only on the REPL. Fixes RT #113776
kudo/nom: 094e41b | moritz++ | src/core/Backtrace.pm:
Merge branch 'ex-hardening' into nom
09:32 krunen left, sudokode joined 09:36 brrt left
moritz r: <a b c>[Any] 09:39
p6eval rakudo 53fc0a: OUTPUT«use of uninitialized variable $pos of type Any in numeric context in block <anon> at /tmp/Hn3qA2mQFn:1␤␤No such method 'Int' for invocant of type 'Any'␤ in method at_pos at src/gen/CORE.setting:5323␤ in method postcircumfix:<[ ]> at src/gen/CORE.setting:1208␤ i…
moritz r: <a b c>[+Any]
p6eval rakudo 53fc0a: OUTPUT«use of uninitialized value of type Any in numeric context in block <anon> at /tmp/EPChfeI9F8:1␤␤»
moritz one could argue that those two should do the same 09:40
n: <a b c>[Any]
p6eval niecza v18-10-g07900b5: OUTPUT«Use of uninitialized value in numeric context␤ at /home/p6eval/niecza/lib/CORE.setting line 1262 (warn @ 5) ␤ at /home/p6eval/niecza/lib/CORE.setting line 296 (Any.Numeric @ 8) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /home/p6eval/niecza/lib/CORE.s…
09:41 krunen joined 09:42 brrt joined 09:45 daxim joined 09:47 wamba left 09:49 Unik left 10:00 SHODAN left 10:01 SHODAN joined 10:10 wamba joined 10:16 scott___ left 10:21 ab5tract joined 10:45 GlitchMr joined 10:55 wamba left 11:02 brrt left 11:08 mtk left 11:16 lue left 11:19 mtk joined 11:28 lue joined 11:35 lue left 11:36 JimmyZ joined
colomon Has it been announced where the next YAPC::NA will be? 11:38
daxim colomon, answer from MagNET #yapc: 11:48
<Yaakov> No, there is no announcement. Bids are being reviewed.
11:48 lue joined
colomon daxim++ 11:48
11:48 skids left
colomon hopes it is someplace he can easily drive to.... 11:49
11:55 leprevost joined
colomon Columbus was actually ideal for me -- an easy drive and I could stay with my brother-in-law. :) 11:56
11:56 spider-mario joined 11:58 tokuhirom joined 11:59 mucker joined
pmichaud (yapc) TPF extended the deadline this year for yapc submissions 12:08
(good morning, Perl6)
kresike hello pmichaud 12:13
12:14 mucker left
pmichaud hello kresike 12:14
[Coke] pmichaud: you didn't get much sleep. 12:15
kresike you certainly don't sleep much
[Coke] heh
kresike :)
pmichaud I may head back to sleep in a bit; have to take $kid to $school 12:16
12:19 _jaldhar joined 12:28 odoacre left
[Coke] kids are done, here. 12:28
pmichaud well, $school is really summer activities 12:29
kresike no kids here yet, thouugh that will change in about three weeks :)
though*
[Coke] ooh, good luck. 12:30
kresike thanks
jnthn home 12:31
masak home 12:32
dalek ast: ea70649 | pmichaud++ | S05-grammar/protoregex.t:
Add tests for RT #113590 (double angles in protoregex names).
12:35 Khisanth left 12:52 atrodo joined
dalek ast: b61ccd1 | pmichaud++ | S03-smartmatch/any-bool.t:
Update boolean smartmatch tests for RT #77080.
 12:54
masak pmichaud: yes, I disagree to the extent that having that box be green (1) tells outsiders "it's done now", and then they are disappoint when it isn't, and also (2) tells us "it's done now", so we don't go looking for more spec to implement. 12:55
12:55 PacoAir joined
pmichaud masak: I'm fine with switching it to yellow. 12:56
It just means that jnthn++ will likely implement method handles to switch it back to green again. :)
masak that sounds like a fine deal by me :)
jnthn :P
jnthn had totally forgotten that bit of the spec existed... 12:57
pmichaud it's only 2 lines in S12. :)
dalek ecs: 26b64e4 | pmichaud++ | S03-operators.pod:
S03: C<true> should be C<so>
atures: f5f9c89 | (Carl Mäsak)++ | features.json:
switch delegation/rakudo to yellow
 12:58
pmichaud ☹ 13:00
13:05 lue left
masak fine with it... but sad? 13:06
pmichaud always a little sad to see green->yellow, even if I know it's "correct" 13:07
masak nod
[Coke] pmichaud: perhaps more cowbell would help.
.u bell 13:08
phenny U+237E BELL SYMBOL (⍾)
PerlJam Easier would have been to remove the two lines from the spec ;-)
(good morning btw)
[Coke] ⍾ - ... Someone needs to make rog-utf8-e 13:09
13:13 Patterner left 13:15 Psyche^ joined, Psyche^ is now known as Patterner 13:17 _jaldhar left 13:18 skids joined, _jaldhar joined, lue joined 13:19 Crooky joined
Crooky Hai. Are the perl6 developer in any way involved with the perl5 developers? 13:19
related*
PerlJam Crooky: kinda not really
Crooky: Many of ther Perl 6 devs are also Perl 5 people in that they use Perl 5 regularly 13:20
pmichaud we're involved at high levels; we usually aren't involved at low-level code levels.
[Coke] doesn't think we have anyone actively committing (or even commenting) on both perls.
Crooky why perl6? And what makes it important?
masak Crooky: p5 and p6 devs keep in touch and talk a lot at conferences. 13:21
Crooky: Perl 6 is a redesign of the Perl ideas embedded in Perl 5. it has some advantages.
pmichaud and even outside of conferences; rjbs++ has been a regular here lately :)
13:21 driador left, driador joined 13:31 c1sung joined 13:41 brrt joined
PerlJam lue++ 13:46
dalek ast: 9e82fe0 | pmichaud++ | S02-names-vars/perl.t:
Test default Mu.perl (resolves RT #62002).
13:48 fgomez left
masak wow, there's 195 nicks in here right now. 13:49
13:49 Vlavv left
frettled That means that the activity is at less than 3% in an hour. ;) 13:50
13:51 sisar left
masak why does it mean that? 13:51
frettled Only 6 nicks said anything in the past hour. :) 13:52
6/195 ≈ 3%
flussence 7! 13:53
masak ah.
frettled I'm not making a value judgment, though.
flussence: darnit, 8?
A whopping 4%!
masak hehe.
frettled Given the technical level of some of the discussions here, I don't think that's bad. I think it's very good that so many listen or pretend listening. 13:54
masak well, I don't believe lurkers are a sign of a systemic error or anything.
quite the opposite.
sometimes I'm concerned about the argument "I don't dare say anything, because obviously all you people are rocket surgeons from the fifth dimension, and I'm just a lowly coder". 13:55
but I don't know what to do about it. it's always been like that here. 13:56
it's hard to fix people's confidence :)
jnthn Rocket...surgery? :P
masak it ain't exactly brain science, you know.
kresike masak, your perfect english makes things even worse :P
PerlJam heh 13:57
masak kresike: nice try, Frenchie boy.
we're not switching the channel over to French. 13:58
PerlJam honestly, many times #perl6 does speak in another language that looks like english but doesn't grok the same
kresike by the way, after watching a few of jnthn's presentations about rakudo development, it seems a little simpler to me too :)
masak aye. jnthn++ has that effect on people.
kresike masak, make that hungarian :)
PerlJam (I believe I just minted a new use of the work grok in that sentence :)
masak oh!
my bad.
PerlJam: medial 'grok'? nah, old :P
PerlJam oh, perhaps I just don't see it often
(ever)
kresike but don't try learning it, I know by experience it's one of the harder languages ... 13:59
masak PerlJam: you know it originally means "to drink", right? :P
PerlJam aye
masak kresike: that's what they all say.
[Coke] is not sure he's share water with any of you weirdos.
13:59 tyatpi joined
[Coke] *he'd 13:59
masak kresike: do you know the tongue-twister with the sailing boat? the one with lots of "hajo" in it?
PerlJam kresike: Perl 6 is not one of the harder languages ;)
jnthn [Coke]: To drink, or to swim in? :)
masak [Coke]: *lol* 14:00
kresike masak, Ok, go ahead then, maybe I'll have something I can help you with :)
PerlJam [Coke]: as long as the water isn't yellow and has been boiled, you're probably safe
kresike masak, you mean: vitorlás hajó
masak jnthn: [Coke]++ is referring to the work of fiction that introduced "grok". www.amazon.com/Stranger-Strange-Lan...441790348/
jnthn oh!
I'm so uncultured :) 14:01
masak kresike: yes, but there's a whole sentence with just repeated haj? all the way.
jnthn: no, you're just behind on your Heinlein ;)
kresike PerlJam, compared to hungarian, it's a breeze :)
masak, I don't think I know the right answer to that :( 14:02
PerlJam masak: I specifically read that book because of the word "grok" many years ago
haj? == buffalo?
;)
masak PerlJam: I think I knew about the word, but read the book for other reasons.
PerlJam: it's way out of date in some ways (even though it was a bit ahead of its time back then), but it's still an eerie read. 14:03
14:03 sisar joined 14:06 leprevost left
dalek p: 5a8707b | pmichaud++ | src/HLL/Compiler.pm:
Add -V/--verbose-config option to HLL::Compiler (RT #62462).
 14:13
14:15 fgomez joined
dalek kudo/nom: 185f8fc | pmichaud++ | tools/build/NQP_REVISION:
Bump NQP_REVISION to get -V option (RT #62462).
 14:27
14:33 JimmyZ left, JimmyZ joined 14:35 wtw left 14:36 adu joined 14:41 felher left 14:46 tokuhirom left
sisar rn: say any(2,3)>>+<<all(1, 2); 14:50
p6eval niecza v18-10-g07900b5: OUTPUT«all(any(3, 4), any(4, 5))␤»
..rakudo 094e41: OUTPUT«any(all(3, 4), all(4, 5))␤»
sisar honestly, i don't know what I was expecting ^
harmil -3 was the correct answer 14:51
pmichaud well, since neither any(2,3) nor all(1,2) are lists, they end up acting just like scalars (or lists with one element)
rn: say any(2,3) + all(1,2)
p6eval niecza v18-10-g07900b5: OUTPUT«all(any(3, 4), any(4, 5))␤»
..rakudo 094e41: OUTPUT«any(all(3, 4), all(4, 5))␤»
harmil Which seems like the correct lazy behavior... 14:52
pmichaud rn: say all(1,2) + any(2,3) 14:53
p6eval rakudo 094e41, niecza v18-10-g07900b5: OUTPUT«all(any(3, 4), any(4, 5))␤»
pmichaud rn: say any(2,3) + all(1,2)
p6eval niecza v18-10-g07900b5: OUTPUT«all(any(3, 4), any(4, 5))␤»
..rakudo 094e41: OUTPUT«any(all(3, 4), all(4, 5))␤»
pmichaud senses a rakudobug.
masak how so?
jnthn I think the current auto-threader just threads on the first junction it finds. 14:54
masak oh!
that's wrong, yes.
jnthn S09 has some funny wording on doing something else.
masak submits rakudobug
jnthn It's already in RT.
masak S09 says "disjunctions outermost, yes".
jnthn Yeah.
masak ok.
masak unsubmits it
jnthn The junction auto-threader is also slow.
It's still the "make something work enough" cut from the very early nom days.
masak jnthn: it's not so funny (as in weird), if you consider how && an || nest in terms of precedence. 14:55
and&
jnthn masak: Yeah, I know the reason.
masak and*
pmichaud leftmost 'all' or 'none', followed by leftmost 'one' or 'any'
masak nod.
jnthn It's just an edge case Rakudo doesn't implement yet.
14:56 felher joined
sisar rn: say (any(2,3))++; #Rakudo leaks internal(?) error 14:58
p6eval rakudo 094e41: OUTPUT«No such method 'STORE' for invocant of type 'Int'␤ in sub AUTOTHREAD_METHOD at src/gen/CORE.setting:1829␤ in <anon> at src/gen/Metamodel.pm:2296␤ in sub postfix:<++> at src/gen/CORE.setting:1341␤ in block <anon> at /tmp/Anddle7G7z:1␤␤»
..niecza v18-10-g07900b5: OUTPUT«Unhandled exception: cannot increment a value of type Junction␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /home/p6eval/niecza/lib/CORE.setting line 272 (Mu.succ @ 9) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /tmp/q8xcTWtSBc line 1 (mainline @ 3) ␤ …
pmichaud doesn't look internal to me
sisar but yeha, LTA erroe 14:59
but yeah, LTA error
pmichaud rn: say 2++
p6eval niecza v18-10-g07900b5: OUTPUT«Unhandled exception: Writing to readonly scalar␤ at /tmp/obkhqvBI4v line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3917 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3918 (module-CORE @ 562) ␤ at /home/p6eval/niecza/li…
..rakudo 094e41: OUTPUT«Cannot assign to a non-container␤ in sub postfix:<++> at src/gen/CORE.setting:2770␤ in block <anon> at /tmp/7l3RmlxCH_:1␤␤»
moritz well
the proper thing would be to fail the dispatch to postfix:<++>
pmichaud moritz: why?
moritz after all postfix:<++> has an rw argument
and rw-ness should fail at binding time 15:00
erm
pmichaud r: my $a = 2; my $b = 3; any($a, $b)++; say $a; say $b;
p6eval rakudo 094e41: OUTPUT«No such method 'STORE' for invocant of type 'Int'␤ in sub AUTOTHREAD_METHOD at src/gen/CORE.setting:1829␤ in <anon> at src/gen/Metamodel.pm:2296␤ in sub postfix:<++> at src/gen/CORE.setting:1341␤ in block <anon> at /tmp/dht_TpIRbU:1␤␤»
moritz binding a non-container to an rw-paremeter should fail at binding time
jnthn thinks moritz is right here
pmichaud okay, if we cannot have junctions of containers, then yes.
15:00 _jaldhar left
jnthn I think the junction construction is decontainerizing. 15:01
masak could "fail at binding time" mean we can detect it at compile time at some point in the future?
pmichaud r: my $a = 2; my $b = 3; any(\$a, \$b)++; say $a; say $b;
p6eval rakudo 094e41: OUTPUT«No such method 'succ' for invocant of type 'Capture'␤ in sub AUTOTHREAD_METHOD at src/gen/CORE.setting:1829␤ in <anon> at src/gen/Metamodel.pm:2296␤ in sub postfix:<++> at src/gen/CORE.setting:1341␤ in block <anon> at /tmp/pfVcYA13t5:1␤␤»
masak r: say "OH HAI"; 5++
p6eval rakudo 094e41: OUTPUT«OH HAI␤Cannot assign to a non-container␤ in sub postfix:<++> at src/gen/CORE.setting:2770␤ in block <anon> at /tmp/mocoU1kHOU:1␤␤»
jnthn That last one is probably catchable at compile time in le optimizer. 15:02
pmichaud l'optimizer :-)
jnthn d'oh :P
pmichaud so, does that mean that is rw parameters never autothread, ooc? 15:03
that feels.... wrongish.
jnthn pmichaud: I think they can auto-thread.
pmichaud: But we statically know 5 is not rw :)
pmichaud I'm talking about the (2,3)++ case 15:04
er, any(2,3)++
jnthn Ah
I think ti'd auto-thread
15:04 birdwindupbird left
jnthn Probably already does 15:04
15:04 thou joined
jnthn Then fails to bind. 15:05
pmichaud actually, I think the problem is that it *is* binding
jnthn Oh.
pmichaud thus we get to the STORE issue
jnthn Oh, it is auto-threading
But then the "is rw" doesn't cause a bind failure yet.
I think we had some reason for delaying adding that...I forget what it was.
It's probably not a reason any more. Guess there's a way to find out... :) 15:06
sisar I'm sorry, i don't know enough about this issue to file a bug report (or if there is even an issue or not), maybe masak++ can do it...
sisar -> dinner
masak I'm not sure yet if it is a bug either. 15:07
[Coke] is sad at www.youtube.com/user/yapcna/videos...y=pmichaud 15:09
pmichaud I'm pretty sure the post-autothreading calls should fail the binding
jnthn Yes 15:10
But it's not anything to do with auto-threading really
r: sub foo($x is rw) { }; foo(5)
p6eval rakudo 094e41: ( no output )
jnthn That's the real issue.
pmichaud [Coke]: oh, yay, at least the videos are starting to appear
r: 5++
p6eval rakudo 094e41: OUTPUT«Cannot assign to a non-container␤ in sub postfix:<++> at src/gen/CORE.setting:2770␤ in block <anon> at /tmp/BQ_3JlegfF:1␤␤»
pmichaud oh, I get it. okay.
well, not completely, but okay. I don't understand why 5++ and any(2,3)++ give different errors 15:11
PerlJam [Coke]: sad that the videos aren't all tagged properly? I'm pretty sure I saw pmichaud's list talk on there the other day.
pmichaud PerlJam: those are unofficial videos 15:12
[Coke] ah. s/pm/m// !
pmichaud Coke is looking at the yapcna official channel
15:13 Khisanth joined
kresike bye all 15:13
15:13 mucker joined, kresike left
[Coke] there is one patrick video, but this one is only 49s. :) 15:14
PerlJam ah
pmichaud mithaldu's videos were "unauthorized" -- see the end of blogs.perl.org/users/mithaldu/2012/...utube.html
PerlJam just searched for "YAPC 2012" and doesn't care if they are "official" or not :)
daxim official and complete is youtube.com/yapcna 15:15
pmichaud well, I don't really care about officialness either; but I'm pretty sure mithaldu won't be posting more, so that means we have to wait for the official yapc::na ones to come out, and since none of my talks are out yet, I am sad (as Coke++ is)
daxim youtu.be/SuY8FOIb7I0 # Introducing Camelia 15:16
15:16 Dzinosky joined
pmichaud daxim: official, yes; complete, I hope not. 15:16
15:17 Dzinosky left
pmichaud as in, I hope what they have up there now isn't "complete" :) 15:17
15:18 JimmyZ left
daxim completer than just the last day, what mithaldu got 15:19
pmichaud okay, I managed to close 11 RT tickets in the past six hours; time for a break :) 15:20
jnthn whoa
pmichaud++
frettled pmichaud++ good going, man! 15:22
pmichaud that even includes about 3 hours of sleep :)
masak pmichaud++ 15:24
moritz pmichaud++ 15:25
15:28 JimmyZ joined 15:32 kaare_ joined, JimmyZ left 15:39 lue left
rjbs pmichaud++ # and if you just need a break from p6, we have some tickets you could close in p5 for a break :) 15:39
[Coke] I am happy to see leurs (sp?) and keenan slogging through that queue. 15:41
moritz le urs? 15:43
[Coke] with bees in his mouth. 15:44
hurm. that may be slightly too esoteric. ;) 15:45
Jesse Luehrs.
++
15:45 sirhc joined
moritz oops, seems I've broken the layout on perl6.org/community/ 15:51
15:51 lue joined
masak you broke the community! o.O 15:52
anyway, moritz++ for improving things in general. 15:53
dalek href="https://perl6.org:">perl6.org: 375a7e5 | moritz++ | source/community/index.html:
[community] unbreak markup
moritz in general, but not in particular :-) 15:54
masak you'd probably get more ++es if I knew the particulars :) 15:55
masak .oO( unbreak markup. say you'll love me again... ) 15:56
[Coke] *groan* 15:57
masak requested: a lightning talk about pop love songs as memetic viruses thriving on people's neurochemical pathways.
masak .oO( this song is *exactly* what happened to me: yes, yes, he walked out the door! ) 15:58
16:02 brrt left
moritz nr: my $x = 2; $x = ($x + 2 / $x) / 2 for 1..10; say $x 16:04
p6eval niecza v18-10-g07900b5: OUTPUT«1.4142135623730949␤»
..rakudo 185f8f: OUTPUT«1.41421356237309␤»
moritz nr: my $x = 2; $x = ($x + 2 / $x) / 2 for 1..10; say $x.WHAT
p6eval rakudo 185f8f, niecza v18-10-g07900b5: OUTPUT«Num()␤»
moritz nr: my $x = 2; $x = ($x + 2 / $x) / 2 for 1..5; say $x.WHAT
p6eval rakudo 185f8f, niecza v18-10-g07900b5: OUTPUT«Rat()␤»
16:07 cognominal left, cognominal joined
masak that looks right to me. 16:09
moritz aye 16:10
dalek c: cfa8f0e | moritz++ | lib/Rat.pod:
[Rat] explain overflow of denominator and fallback to Num
 16:11
16:11 adu left
sorear good * #perl6 16:16
16:17 cognominal left 16:23 sisar left 16:24 vlixes joined
moritz \o sorear 16:25
dalek c: f6edbb6 | moritz++ | lib/Rat.pod:
[Rat] some more trivia
 16:29
16:32 kaleem_ joined, sisar joined 16:34 kaleem left 16:38 moukeddar joined, moukeddar left
dalek c: 25a4b5f | moritz++ | lib/Rational.pod:
Rational
 16:39
c: 6d2c1d8 | moritz++ | lib/Rational.pod:
[Rational] method signatures; method new
 16:45
[Coke] moritz: mind a docpatch for perl6/doc that updates the README to explain why all the docs look empty in github? 16:53
(I think this points to a larger confusing issue for POD tools that find six.pod files in the wild.) 16:54
moritz [Coke]: I don't mind, no 16:55
dalek c: 7bf22c9 | moritz++ | README:
[README] add help wanted section
moritz hm, maybe we should switch to .pod6 as extension 16:56
16:56 dakkar left 17:11 kaleem_ left 17:12 kaleem joined 17:18 Gesh left
[Coke] I would suggest it, but imagine it would be shot down. 17:21
17:36 vlixes left 17:39 daxim left 17:41 lue left 17:50 moukeddar_ joined 17:51 moukeddar_ left
masak nah. knock yourselves out with extensions. :) 17:54
17:54 lue joined
PerlJam moritz: nah, the suffix should be .p6d 17:56
</bikeshedding> :)
timotimo i'm for pōd instead 17:57
it looks like a combination of pod and p6d, because the upper line of the 6 is still there 17:58
17:59 birdwindupbird joined
PerlJam timotimo: clearly your suggestion is best. 17:59
timotimo i know, but thanks :]
masak so much agreement 18:00
[Coke] .рбд
geekosaur rabid pods?
masak general question to the room: in writing up the blog post for p6cc2011/t5, I'm desperately trying to remember the URL of a PDF paper from the 1980s about counting unrooted 4-trees using functional programming. does anyone remember sending this URL to me in the first place? (through privmsg, IIRC) 18:06
18:11 snearch joined 18:12 am0c left
masak nvm, moritz++ dug it up from our privmsgs. 18:17
it's here, for people who can't wait for the t5 post: www.cs.wright.edu/~tkprasad/courses...turner.pdf
PerlJam resists a pun with K&R C 18:20
18:27 kaleem left, fhelmberger left 18:35 Crooky left 18:39 lue left 18:40 mustapha joined
masak it seems a cute language, really. "a collection of equations by means of which the user attaches names to various objects in KRC's universe of discourse". sounds like Haskell :) 18:49
masak only now noticed that the definition of KRC is basically recursive
18:50 Gesh joined 18:51 lue joined 18:54 Gesh left, Gesh joined 18:59 Gesh left 19:00 GlitchMr left, Chillance joined
sisar hm, this is the error i get when trying to install LWP::Simple using panda: gist.github.com/2990527 . Looks like Panda failed to recognize that MIME::Base64 is a dependency. 19:00
tadzik looks 19:01
oh, I see
grep -i base ~/.panda/state will probably tell you, that it's installed, while it isn't really
so your ~/.panda and ~/.perl6 are probably out of sync
sisar tadzik: right. What should I do ? 19:02
arnsholt masak: If KRC is Kent Recursive Calculator, it's Haskell's grandmother IIRC 19:03
tadzik sisar: you can manually install MIME::Base64
arnsholt KRC begat Miranda begat Haskell
sisar tadzik: ok
tadzik or just purge/manually update your state file, to be sure it works in the future
19:03 brrt joined
arnsholt (Ref. Wadler's Why Calculating is Better than Scheming) 19:04
sisar adding a "plan x;" to a .t file is compulsory or recommended ? 19:09
tadzik phenny: en pl "compulsory"?
phenny tadzik: "obowiązkowy" (en to pl, translate.google.com)
tadzik recommended 19:10
you can also add 'done;' at the very end
sisar why is it recommended ? why is 'done;' recommended ? 19:11
19:11 pupoque is now known as Gesh
tadzik having either of those is good to have, for if you don't have one, TAP harness cannot know whether the tests are finished, or simply the script has crashed 19:11
geekosaur (1) documentation, and so test frameworks can tell how many tests you are supposed to have; (2)... that 19:12
19:20 vmspb joined 19:24 bluescreen10 joined 19:25 driador left
dalek p/toqast: 4c86f84 | jnthn++ | src/QAST/Node.nqp:
$!name can't be str if we put it here, it turns out. May be better to move it into various other nodes where it can be, since it's just Regex with specialer needs.
 19:33
p/toqast: 20a37a9 | jnthn++ | src/Q (3 files):
Various fixes so we can use native types on QAST::Regex attributes.
masak arnsholt: ooh! that makes a lot of sense, then. 19:42
19:48 driador joined, alvis joined
dalek kudo/toqast: 841e783 | jnthn++ | src/QPerl6/ (2 files):
Translate method calls to QAST. Again, a few complicated ones left over (please don't 'help' by changing the leftovers!)
 19:52
masak "(please don't 'help' by changing the leftovers!)" -- I picture a frustrated chef somewhere. 19:57
tadzik :P
masak "Madre mia, they again switch the leftovers... for chicken feathers! Why, why?" 19:58
jnthn Yeah, one of them will result in a bloody hard to find bug if dealt with wrongly...
20:04 mustapha left 20:23 sisar left
eiro tadzik, jnthn i don't find your pass id: did you sent it to me ? 20:24
hello all 20:25
BooK, also :)
masak hi eiro
eiro masak, hi. i'm working on the perl6 introduction (what is rakudo, rakudo-star, parrot, ...) 20:26
any slides to steal ?
20:27 kaare_ left
tadzik eiro: no, I did not; TBH, I fell a little bit uneasy about it :/ 20:28
eiro: is it too late already?
jnthn
.oO( Do they validate the numbers we give them... )
 20:29
eiro tadzik, your choice :)
tadzik, no but i prefer to ask :)
jnthn eiro: jnthn.net/articles.shtml is my slides page
eiro jnthn, i don't think so: feel free to give a fake :)
jnthn, thx
tadzik eiro: if I wanted to buy a local SIM to use the internet, how do the prices look like? 20:30
eiro i really don't know about those kind of things :(
tadzik BooK?
eiro tadzik, i ask on the fpw list right now 20:31
20:31 pochi joined
tadzik okay, thanks 20:31
20:36 sisar joined
eiro tadzik, jnthn masak i hope you'll be there to back me up at the perl6 intro: i'll let a long time for free questions 20:37
tadzik eiro: as long as the language barrier won't stop us :)
masak language barrier? 20:38
phenny: en fr "language barrier?"?
phenny masak: "barrière de la langue ?" (en to fr, translate.google.com)
masak phenny: "barri?re de la langue?"? 20:39
phenny masak: "barrels ? re de la langue ?" (es to en, translate.google.com)
sisar masak: the type hierarchy/tree diagram you have on strangelyconsistent.org/blog/june-24-2011-types, is partial. Is the complete diagram of type hierarchy available somewhere?
masak phenny: fr "barri?re de la langue?"?
phenny masak: "Barri ? re the language ?" (fr to en, translate.google.com)
masak oh!
phenny: fr "barrière de la langue?"?
phenny masak: "language barrier ?" (fr to en, translate.google.com)
masak sisar: no. feel free to create one.
eiro tadzik, i'll translate if needed. the thing is i don't know what kind of question can be asked 20:40
sisar masak: ok. I thought the spec would(should?) have one.
tadzik eiro: we'll do the best we can
masak by the way, I again want to advertise my plans for July to re-visit strangelyconsistent.org/blog/june-3...venture-ii and re-build it from scratch in 31 simple steps, with tests. 20:41
tadzik, eiro: I'm not worried, at all.
some of us know school French. and there are plenty of anglophones in the room.
(for those who don't know, an anglophone is a special phone where you call for translation help. France is full of them) 20:42
sisar masak: i think thats the only part of your blog series I havne't gone through ! I mean the last two post, the ones with the game.
masak turns out the game was far too big for two posts.
(and hence took far too long to get out the door, too)
but I must say I'm proud of the image in strangelyconsistent.org/blog/june-2...dventure-i 20:43
I wish I didn't have to make such images using Inkscape. I'd like a DSL for them.
eiro masak, "is learning perl6 worth now" ? are you ready for that ? 20:45
PerlJam masak: dot is kinda close
masak: but ... you could always make a DSL yourself (and blog about it! :-) 20:46
dalek kudo/toqast: 27f9453 | jnthn++ | src/QPerl6/ (2 files):
Translate a bunch of calls to QAST.
masak PerlJam: I already have plans. there's a gist if you're interested. 20:47
PerlJam: the idea has infected many things; my blog; some $work stuff. I expect to do it this summer. 20:48
eiro: could you please be more specific? is this a talk I'm supposed to make? do I need slides?
eiro masak, no: i mean: 20:49
PerlJam masak: I'm certainly curious, but I wouldn't say "interested" because I don't want to add to my list-of-things-undone
20:50 snearch left
eiro one of the first talk is "intro to perl6". I have a slot of 45mn to introduce all the strange keywords, the history of perl6, what the community expect from it and what the users can expert right now 20:50
the truth is: i'm really not sure about my answers
masak PerlJam: oh, I'm not dragging you into anything. but it certainly would be fun to drag the idea through your expertise and see what sticks. 20:51
masak digs up the gist
sisar gist.github.com/2991123 ... how the hell did Rakudo reach line no. 5 ?! There are only three lines in the code !
eiro it would be very easy to me if i know you're around and handle those kind of questions
sisar r: { 20:52
p6eval rakudo 185f8f: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 2, near ""␤»
masak PerlJam: gist.github.com/1751911
sorear sisar: rakudo used to complain about exceptions in the setting around like 1 bln
20:52 driador left
arnsholt sisar: Rakudo has been known to have problems counting lines 20:52
eiro i mean: the running joke "perl6 is the langage of Duke Nukem Forever running on HURD", someone will tell us that... i know it
masak eiro: I'll certainly be ready for such questions. 20:53
eiro or: perl5 is fine, why perl6 ?
masak, ok :)
masak eiro: I remember when jnthn and I came to YACP::Russia in Moscow, and I got all those questions.
sisar sorear, arnsholt: oh, ok.
tadzik . o O ( "Oh, I haven't noticed you uploading DNF to our ecosystem" )
masak eiro: I feel I'm pretty good at handling them.
eiro \o/ glad to read that :)
jnthn Nobody has ever actually asked me a question involving HURD and Duke Nukem at a Perl conference yet, fwiw. :) 20:54
PerlJam eiro: "Why is it called Perl 6 if it's a different language?" :)
masak the more I see Perl 6 emerging, the more I feel it was worth the wait. 20:55
sisar r: use v6;␤␤{ 20:57
p6eval rakudo 185f8f: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 4, near ""␤»
PerlJam btw, people occasionally come to #perl asking about #perl6. This seems to be happening more often lately.
sisar huh, here it is line 4, but i locally get line 5
PerlJam Maybe that was the YAPC effect?
20:57 tyatpi left
jnthn r: use v6;␤␤{␤ 20:57
p6eval rakudo 185f8f: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 5, near ""␤»
jnthn It looks on the last line for it :) 20:58
20:58 driador joined
sisar jnthn: hmm 20:58
PerlJam r: {
p6eval rakudo 185f8f: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 2, near ""␤»
20:59 tyatpi joined
pmichaud (YAPC effect) -- I've seen a number of blog posts from yapc::na attendees saying "hmmm, I need to take a / another look at Perl 6" :-) 20:59
tadzik that's awesome :)
pmichaud I'm really hoping the yapcna youtube channel gets my lightning talk up soon :) 21:00
PerlJam pmichaud: What was your lightning talk about?
pmichaud I wonder if I can send a message saying "I'll make a donation to TPF if you do my lightning talk next :-)"
PerlJam: rakudo speed improvement since 2011.01
PerlJam oh, that's right, you showed the timing here 21:01
PerlJam enjoyed mdk's Perl and CPAN 21:02
21:02 skids left
masak pmichaud: ooh! please market it hard as "The YAPC effect", too! :D 21:02
"dude, did you hear about the YAPC effect? apparently YAPC makes people want to check out Perl 6!" 21:03
tadzik dude, did you hear about YAPC effect effect? Apparently YAPC effect makes people want to check out Perl 6!
PerlJam tadzik++
tadzik dude, did you hear about YAPC effect effect effect? AppaMaximum recursion depth exceeded
PerlJam masak: "YAPC effect" gives new meaning to "butterfly effect" :) 21:04
masak PerlJam: only if checking out Perl 6 generates more YAPCs somehow and leads to a positive feedback gain, ending with a grey-goo situation. 21:05
sisar so i'm still trying to understand OOP both in general and in Perl6. I wrote this: gist.github.com/2991221, but it fails. Can someone explain? Also notice the wrong line no. in Rakudo's error. 21:06
PerlJam waits for TEDx YAPC
jnthn More YAPCs wouldn't be a bad thing. :)
jnthn waits for YAPC::Antarctic
That'd be so cool!
masak jnthn: only you would like it :)
tadzik www.myfacewhen.net/uploads/3890-oh-you-dog.jpg oh you
PerlJam sisar: you need to make your attrs rw if you want to change their values
jnthn masak: Oh come on...I can't be the ONLY person here who likes cold weather! 21:07
tadzik we didn't want to tell you
that reminds me of this old joke 21:08
masak :P
PerlJam (though that error message could give you some more hints as to the problem)
PerlJam afk
pmichaud jnthn: I like cold weather... remember how I was enjoying Oslo in April?
sisar PerlJam: I thought 'attributes' are something like private variables. But I don't have any private variables there.
tadzik Boss to employee: "Do you like warm vodka?" "No" "How about sweaty women?" "No" "Great, then you get your vacation in december"
masak jnthn: you do know that all they eat at YAPC::Antarctic is variations on frozen fish?
jnthn Eww. 21:09
How sill-y.
masak :P
phenny: sv en "sill"?
phenny masak: "herring" (sv to en, translate.google.com)
tadzik :)
pmichaud cold would be nice. 38.8 degC here right now.
masak tadzik: there's something about that joke that feels like it could be improved. :) 21:10
21:10 birdwindupbird left
tadzik masak: what is it? 21:10
masak tadzik: it doesn't quite have that shift-in-perspective at the end that a joke of that category has.
pmichaud oops, now 38.9 degC :-)
tadzik hmm
masak I mean, it's surprising, but somehow not enough, or not in quite the right way.
jnthn tadzik: Vacation in December works. Go to where it's too darn hot the rest of the year. :)
Like Texas :P 21:11
masak I took a vacation in December last year :)
tadzik iirc there's 20 degrees celsius in Egypt in January
masak went before the tourist season too, so we had the beach mostly to ourselves.
tadzik both air and water temp
masak: maybe it loses something it translation, but it's still not the best of jokes I know 21:12
21:12 bruges_ left
sisar um, can someone have a look at my gist a few line back. PerlJam told me somthing but he went afk before I could ask him something. 21:12
tadzik sisar: if you want to modify the attributes they need to be "is rw"'d 21:13
sisar tadzik:I thought 'attributes' are something like private variables. But I don't have any private variables there. 21:14
pmichaud 'attributes' are anything declared with 'has'
an attribute declared with a dot gets a public accessor method
masak tadzik: it reminded me of this old joke. gist.github.com/2991296
21:14 bruges joined
pmichaud an attribute declared "is rw" gets a public accessor method that can modify the attribute 21:14
masak pmichaud: well, except that you can go 'has method', IIRC :)
tadzik masak: haha, nice one :) 21:16
sisar pmichaud: "$.d is rw" makes it public... that means other parts of the program can modify $d ?
pmichaud sisar: "$.d" makes it 'public', in the sense of being visible. 21:17
sisar realizes he needs to understand OOP in general :|
pmichaud sisar: "is rw" makes it publicly modifiable, in the sense of being assignable
masak sisar++ # curious in ways that help 21:18
pmichaud has $!d; # $!d can only be accessed from within the methods of the class
21:18 brrt left
pmichaud has $.d; # creates $!d, but also creates a method 'd' that allows outsiders to obtain the value of $!d 21:19
21:19 drbean left
pmichaud has $.d is rw; # creates $!d, and also creates a method 'd' that allows outsiders to modify the value of $!d 21:19
tadzik maybe there's some material for the blog post, "OOP as Perl 6 encourages it"
pmichaud oops, gotta run for a bit 21:20
afk
sisar tadzik: +100 to that. I would really appreciate it if someone wrote that. Unfortunately, masak++'s strangelyconsistent.org/blog/june-2...11-classes is very superficial. 21:22
And en.wikibooks.org/wiki/Perl_6_Progra...Attributes is kinda confusing, well atleast for me.
tadzik sisar: maybe that should just go into the Perl 6 Book 21:23
sisar tadzik: yeah I forgot to mention the Perl6 book, its Classes chapter is not newbie friendly. Given that it is not aimed towards newbies, that is acceptable. 21:25
21:26 mucker left
masak still, we should maybe aim to make it more newbie friendly. 21:27
sisar masak: +1 21:28
21:29 vmspb left 21:30 kurahaupo joined
jnthn I guess there's a difference between being a Perl 6 newbie and an OOP newbie too 21:35
sisar oh, and the completely wrong line number in the gist. Should I file a bug report ? (After golfin it.) 21:38
jnthn Yeah 21:39
sisar will do
masak \o/ 21:40
sisar: before I forget: it's nice to have you on board.
sisar :) 21:41
lue hello #perl6 o/ 21:45
jnthn o/ lue
dalek p/toqast: 702c1b7 | jnthn++ | / (2 files):
Add a QAST::CompUnit, which will be involved in SC handling and a few other things we have one of per compilation unit.
tadzik hey lue 21:46
masak rn: say "Hello, {"Life Universe Everything".words>>.substr(0,1).join.lc}!" 21:48
p6eval rakudo 185f8f, niecza v18-10-g07900b5: OUTPUT«Hello, lue!␤»
lue :)
masak rn: say "Hello, {"Life, the Universe, and Everything".words>>.substr[0, 2, 4](0,1).join.lc}!" 21:49
p6eval niecza v18-10-g07900b5: OUTPUT«Unhandled exception: No value for parameter '$start' in 'Cool.substr'␤ at /home/p6eval/niecza/lib/CORE.setting line 0 (Cool.substr @ 1) ␤ at /tmp/EBwnWWmiJe line 1 (ANON @ 2) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3051 (hyperunary @ 66) ␤ at /h…
..rakudo 185f8f: OUTPUT«Not enough positional parameters passed; got 1 but expected between 2 and 3␤ in method substr at src/gen/CORE.setting:3500␤ in method dispatch:<hyper> at src/gen/CORE.setting:872␤ in block <anon> at /tmp/n5xH9pdE_m:1␤␤»
masak oops. my bad.
rn: say "Hello, {"Life, the Universe, and Everything".words[0, 2, 4]>>.substr(0,1).join.lc}!"
p6eval rakudo 185f8f, niecza v18-10-g07900b5: OUTPUT«Hello, lue!␤»
21:55 ggoebel left, Entonian joined 21:57 Entonian left
harmil I don't want to be pushy, but what can I do to get github.com/rakudo/rakudo/pull/71 merged? 21:58
masak harmil++ # reminder 21:59
22:00 ggoebel joined
jnthn I'm a bit uneasy about the we-know-it-won't-be-called DESTROY going in. The buffering bits look fine...have they been added or are they already in the spec, ooc? 22:00
masak was gonna say the same about DESTROY 22:01
I'm a bit confused, also. the two commits look identical in what they change.
anyone else seeing this?
harmil masak: that's because of how I manage my repo
It's a no-ff merge from my topic 22:02
Once you pull it and do a git log —oneline —graph —decorate it will look sane
masak ok.
harmil About the specness of my changes
masak harmil: I believe a merge is fine. we don't have a policy against merge commits in Rakudo. 22:03
harmil no, it's not in the spec exactly that way. As someone pointed out, the spec for IO is still in need of some updating for the modern world. All I did in the meanwhile was to mock up what Parrot does with buffering
However, flush is definitely in the spec as I did it (coincidentally) 22:04
jnthn harmil: That's fine...sometimes the implementations should lead the spec, and IO is one of those. Was just curious. :)
22:05 kurahaupo left
jnthn I'm a bit distracted right now, and about to go get some rest, but I can merge it in the morning (minus the DESTROY) 22:05
harmil If you feel like reverting the DESTROY addition before merging it in, I'm fine with that
jnthn OK
22:06 ggoebel left
harmil Though, keep in mind that that means that any IO that sets full buffering without explicitly flushing before exit will throw away all of its work. 22:06
Then again, that's the reality whether we have an impotent DESTROY method or not, right now. 22:07
22:08 bluescreen10 left
tadzik do we pass the spectests after these changes? 22:08
22:11 spider-mario_ joined, spider-mario_ left, ggoebel joined, spider-mario left
pmichaud looks at pull request 71 22:12
jnthn Uh-oh :D
22:13 lichtkind joined
pmichaud I'd remove the Bool::True from method flush() 22:14
jnthn pmichaud: Why? 22:15
pmichaud: $!PIO is a Parrot object
lichtkind cabecon++ 22:16
masak right. indecent exposure, no wait, that's not the right expression.
jnthn pmichaud: If we don't explicitly return something we risk leaking a Parrot thingy into Perl 6 space.
masak: We...haven't tended to call the problem that. :P
pmichaud jnthn: at least p6booleanize the return from the Parrot thingy
but always returning True is... wrong.
masak ah, yes. 22:17
that's better.
pmichaud $!PIO.flush ought to be returning whether the flush succeeded or no.
22:17 PacoAir left
jnthn Hm, unless it throws...in which case we need to catch it and turn it into a fail 22:17
pmichaud looking. 22:18
It does throw. 22:19
22:19 PZt left
pmichaud anyway, I'd prefer to see a p6booleanize of the return from $!PIO.flush(). We should at least acknowledge that flush can return something other than True. 22:19
(whether it actually does or not.)
we still don't have a good guideline for use of _ versus - in method names :-/ 22:20
jnthn Time for some sleep, 'night o/ 22:22
tadzik e. 22:23
o/
lichtkind \o
pmichaud harmil: I wonder if there's a better name than 'buffer_type'
masak the underscore indicates that it's low-level, no? 22:27
pmichaud masak: looking
masak: yeah, that seems to fit. In which case this should be buffer-type and not buffer_type 22:28
masak aye. 22:29
pmichaud I also wonder if buffertypes/adverbs should be arguments to open()
22:30 am0c joined 22:33 whiteknight joined 22:34 tokuhirom joined 22:38 cognominal joined
masak 'night, #perl6 22:43
22:43 skids joined
pmichaud I'd like us (#perl6) to have a discussion about how to handle some of the more exploratory parts of the spec while still retaining a commitment to some sense of stability for users. 22:45
22:47 flussence left, flussence_ joined, flussence_ is now known as flussence 22:51 thou left 22:53 ab5tract left 22:55 libertyprime left, libertyprime joined 22:58 am0c left 23:02 cognominal left 23:04 cognominal joined 23:07 Gesh left 23:11 Vlavv joined 23:13 eric joined
harmil re: Bool::True — I was just emulating other calls in IO 23:16
I don't think flush returns non-true. I think it's either True or an exception, but I could be wrong (speaking of Parrot, here) 23:17
pmichaud: yes, the buffer_type call is a direct Parrot mock-up. If you have a better name, it would be perfectly acceptable to re-name it (though I'm getting closer and closer to wanting an IO::Parrot from which Rakudo derives its IO and which non-Rakudo implementations replace with an exception-generating stub)…. just a thought 23:19
Yeah, now that I actually say that, I can't think of a good reason it shouldn't be the case. It's kind of like having the POSIX library in Perl 5. You don't usually want POSIX::write semantics, but if you do, they're there... 23:21
Would anyone be averse to an IO::Parrot?
pmichaud I think an IO::Parrot is an excellent idea. 23:22
I have less concern about stability/deprecation for something that exists as IO::Parrot 23:23
s/less/much less/
harmil yeah, that makes sense
I have a hard time getting my head around the long-term implications of adding interfaces in the core. I know I should be cognizant of it, but that's not terribly easy for me. 23:24
pmichaud re: Bool::True -- yes, I looked at some of the other calls in IO and they're much the same (for better or worse). So the pull request is at least consistent with the rest of IO in that respect.
we all have a hard time with the longer-term implications. :-)
I think I prefer buffer-mode or buffering-mode to buffer_type 23:27
23:27 Chillance left, Chillance joined
pmichaud Indeed, setvbuf talks about "mode" instead of "type", as does Perl 5's documentation on setvbuf 23:28
"type" tends to be an overused noun; kind of like "user" and "data".
23:29 ponbiki left 23:30 simcop2387 left, mikemol left 23:31 ponbiki joined, simcop2387 joined, Pleiades` left 23:32 sorear left, sorear joined 23:33 Pleiades` joined 23:35 Maddingue left 23:36 Maddingue joined 23:37 Guest81317 left, infosophy joined 23:38 xinming left, xinming joined 23:39 pochi left, pochi joined 23:40 ponbiki left, sjohnson joined, rhr left 23:41 ponbiki joined 23:43 ashleydev left 23:44 ashleydev joined
diakopter phenny: ask moritz I'd love to see a histogram of weekly #perl6 activity (the number of messages in a given time slice (1hr? 15min?), averaged over the past N weeks, for the whole week). Does your irclog db support such a thing? 23:44
phenny diakopter: I'll pass that on when moritz is around.
23:46 rhr joined
cxreg so, is there any effort so far to publish a set of common grammars for re-use? like, for example, SQL 23:53
does that make sense?
tadzik sounds sensemaking to me 23:56
I think the best way would be to publish it as a module
as we have URI for example
cxreg yeah, i was just looking at that one
but rather than bundle it with code
just publish the grammars
(code too, but separately) 23:57
tadzik grammars are code :)
23:59 Chillance 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!