»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by moritz on 3 May 2013.
timotimo_ 71.10.146.50:3001/project/Bailador - in these pages, the date column is titled "module". who should i highlight about this? (or maybe even: where's the repo on github?) 00:03
colomon that would be me.
timotimo_ you are spartacus? 00:04
colomon timotimo_: no, I am colomon
timotimo_ oh
colomon you're looking at output from github.com/colomon/SmokeResults
running on the Linux box in my basement
timotimo_ that's why it reacts kind of slowly! 00:05
at first i thought it was built in perl6, but then i saw it's dancer
colomon using a debug Dancer install, too.
It's actually a pretty fast Linux box, but it's using the Dancer development server rather than a real web server
timotimo_ probably got a slow-ish internet connection 00:06
00:09 benabik left 00:14 snearch joined 00:19 rurban joined 00:23 ggoebel joined 00:29 btyler left 00:39 tomyan left 00:46 benabik joined 00:47 telex left 00:51 snearch left 00:52 census joined 00:54 telex joined 01:14 anuby joined
census hi! 01:19
01:21 havenwood joined 01:26 Teratogen left 01:31 census left 01:32 btyler joined 01:36 prevost left
dalek p: 6c24e71 | coke++ | tools/lib/NQP/Configure.pm:
Quote args passed to parrot's Configure.
 01:37
p: 1ff01d2 | coke++ | tools/lib/NQP/Configure.pm:
Allow spaces in --prefix path
[Coke] whoever does the next star release - please upgrade to the latest parrot and nqp to get some build fixes for spaces. 01:39
(spaces in build & install paths) 01:40
01:44 JimmyZ joined
[Coke] needs RT help. 01:48
geekosaur what kind of help? 01:49
[Coke] I want to search for all perl6 tickets that do not contain the tag "testneeded" 01:50
I can only search for tickets that contain the tag. Trying to reverse the sense of the search just shows all the tickets. 01:51
geekosaur urgh. I don't have sufficient access to rt.perl.org to do normal searches, as far as I can tell, so I can't see how tags are defined to see how to specify a not properly :/ 01:52
[Coke] Do you have an rt.perl.org login?
geekosaur no 01:55
that's why I don;t have access, random schlubs don't get to use their search form as far as I can tell
[Coke] rt-- 02:01
02:09 lustlife` joined
[Coke] .ask masak if he can provide an RT search that shows tickets that are NOT marked testneeded. 02:10
yoleaux [Coke]: I'll pass your message to masak.
02:10 lustlife joined 02:14 yoleaux left, SamuraiJack joined 02:18 yoleaux joined, ChanServ sets mode: +v yoleaux
dalek ast: 14f35c6 | (Solomon Foster)++ | S02-types/bag.t:
Fix bag.t test to use .Bag instead of .bag.
 02:28
ecza: b15a162 | (Solomon Foster)++ | lib/CORE.setting:
Kill non-standard .set and .bag methods, update Bag.new-from-pairs so that it doesn't issue warnings.
[Coke] r: my $x = :(Int); $x.^methods>>.say 02:31
camelia rakudo b2072f: OUTPUT«perl␤returns␤count␤<anon>␤<anon>␤ACCEPTS␤params␤arity␤<anon>␤<anon>␤» 02:32
dalek ast: c399710 | (Solomon Foster)++ | S02-types/ (2 files):
Translate the last batch of .Bag tests to .KeyBag tests.
 02:35
ast: b320135 | (Solomon Foster)++ | S02-types/keyset.t:
Translate the last batch of .Set tests to .KeySet tests.
 02:40
ecza: 6314dd4 | (Solomon Foster)++ | lib/CORE.setting:
Update KeyBag.new-from-pairs so that it doesn't issue warnings.
 02:41
02:55 wamiks joined 02:58 thou joined 03:00 stevan_ joined 03:06 preflex_ joined, ChanServ sets mode: +v preflex_ 03:07 preflex left, preflex_ is now known as preflex
sorear o/ #perl6 03:09
[Coke] ~~ 03:11
03:11 raiph joined, cognominal left 03:31 labster joined
labster hiya, #perl6 03:33
I am alive and back from YAPC:NA 03:34
diakopter hi :)
[Coke] nr: say first * > 5, 1..3 03:36
camelia niecza v24-70-g6314dd4: OUTPUT«Nil␤»
..rakudo b2072f: OUTPUT«No values matched␤ in method gist at src/gen/CORE.setting:10161␤ in method gist at src/gen/CORE.setting:893␤ in sub say at src/gen/CORE.setting:11047␤ in block at /tmp/nJ5_Ga8uxI:1␤␤»
sorear Hi labster 03:48
So am I :D
diakopter :D 03:49
labster good to hear
diakopter you scal folks 03:50
03:50 havenwood left
labster Scala folks? 03:51
sorear socal, presumably
labster maybe he was typing with a lisp 03:52
sorear or he acidentally left out a letter 03:58
labster my puns are failing tonight
sorear trip went well? 04:01
04:02 atroxaper joined
labster plane started 1 hour late, landed 1.5 hours late. :( 04:03
both of my 787 Dreamliner flights had technical problems that delayed the flight.
sorear mixes words from adjacent lines D:
"plane crashed 1 hour late!"
04:07 rurban left 04:14 sazzq joined 04:15 rurban joined
benabik 787? Technical problems? Shocking... 04:16
04:19 btyler left 04:26 Psyche^ joined 04:27 btyler joined 04:29 mrallen1 joined 04:30 Patterner left, Psyche^ is now known as Patterner 04:33 aindilis joined 04:42 mrallen1 left 04:46 rurban left 04:52 hlin_ joined 04:56 atroxaper left 04:59 kaare_ joined 05:06 sazzq left
diakopter heh, lisp 05:17
05:22 cognominal joined 05:25 SamuraiJack left 05:29 xenoterracide left 05:46 rurban joined 05:50 rurban left 05:57 btyler left
moritz \o 06:07
diakopter o/
sorear o! 06:12
nwc10 !o
06:16 estrabd left 06:17 FROGGS[mobile] joined
FROGGS[mobile] I feel homeish 06:17
06:17 cognominal left
moritz welcome backish :-) 06:18
FROGGS[mobile] :o)
sorear o/ whereever you are, FROGGS[mobile]
FROGGS[mobile] in a train in berlin 06:19
moritz: you have still dry feet?
hi sorear
moritz FROGGS[mobile]: yes, I'm blessedly high up here, compared to any water around me 06:21
FROGGS[mobile] good to know 06:22
quiet some ppl that I know got evacuated
sorear is moritz in a plane? 06:25
moritz no
sorear oh, he wasn't travelling from yapc
floods?
www.abc.net.au/news/2013-06-08/germ...de/4742236 ? 06:26
moritz aye 06:27
p4.focus.de/img/gen/E/7/HBE7VHnS_Px...02+0+0.jpg # 100km from my home
FROGGS[mobile] and I guess the flood is away already? 06:29
sorear eesh, that does not look fun 06:30
is 100km the closest it is to you?
moritz sorear: no, just where the worst of it is
jnthn made it home safely
yoleaux 8 Jun 2013 23:16Z <sorear> jnthn: Would significant changes to ResumeState.java be acceptable? It *looks* currently unused...
FROGGS[mobile] jnthn: \o/ 06:31
sorear jnthn! \o/
jnthn sorear: Current ResumeState.java was my first guess at how it may need to look; change it as you need.
sorear: It's certainly not in use yet.
sorear jnthn: k
glad to hear you're home safe 06:32
I am too :D
FROGGS[mobile] obviously *g*
jnthn yay :)
It was an uneventful journey.
Just long. 06:33
(5-6 hour layover at Dulles)
sorear FROGGS[mobile]: not so obviously. my layover in phoenix had excellent wifi and (unlike austin) doesn't block outbound ssh
FROGGS[mobile] I was sitting between two fat ppl -.-
jnthn FROGGS[mobile]: ugh
sorear I could be overnighting there with a cancelled flgiht 06:34
moritz jnthn: it has 'dull' in the name :-)
FROGGS[mobile] so I had their ellbows in my ribs for 8h+
jnthn moritz: Yeah...well, it did at least have free wifi...that mostly worked, if you sat in the right places
moritz FROGGS[mobile]: so that's a good motivation to get a window seat next time, then you'll have at most one ellbow in your ribs 06:35
FROGGS[mobile] right
jnthn always tries to get those 06:36
sorear muses about teratomas
FROGGS[mobile] I guess sitting near the vails has advantages too (getting the laptop) 06:39
sorear reads more on method handles 06:41
jnthn OK, my current uptime is somewhere around 22 hours, given I can't sleep on planes... 06:42
So I'm gonna go take a short nap :)
06:42 tomyan joined
moritz jnthn: good reboot :-) 06:43
sorear
.oO( my uptime is more like 22 or 14 years, since I only sleep-mode at night, not complete shutdown )
diakopter seems a problem 06:45
06:45 Guest1337 left 06:46 rurban joined 06:52 rurban left, JimmyZ left
sorear attempts to press Scala into service as a Java REPL, despite knowing zero scala 06:55
labster good luck sorear 06:56
07:07 rindolf joined 07:08 JimmyZ joined 07:10 konundra left 07:18 JimmyZ left 07:20 tomyan_ joined 07:22 tomyan left 07:25 JimmyZ joined 07:28 Blahhh joined 07:30 stevan__ joined 07:31 JimmyZ left 07:33 LordVorp joined 07:35 awwaiid_ joined, Tene_ joined, Tene_ left, Tene_ joined, zakalwe_ joined, cxreg2 joined, xinming joined, pjcj joined 07:36 __sri joined 07:38 _sri left, awwaiid left, Tene left, stevan_ left, xinming_ left, zakalwe left, s1n left, LV left, cxreg left, pjcj_ left, ivan``_ left, ivan`` joined
labster r: sub foo { X::IO::Dir.new(path=>'a', os-error=>'b').fail }; my @a = foo(); 1; 07:46
camelia rakudo b2072f: ( no output )
labster r: sub foo { X::IO::Dir.new(path=>'a', os-error=>'b').fail }; my @a := foo(); 1;
camelia rakudo b2072f: OUTPUT«Failed to get the directory contents of 'a': b␤ in method sink at src/gen/CORE.setting:10169␤ in method BUILDALL at src/gen/CORE.setting:801␤ in method bless at src/gen/CORE.setting:743␤ in method new at src/gen/CORE.setting:728␤ in method new at src/gen/CORE.…
labster should binding cause the failure to get thrown?
FROGGS[mobile] binding is lazy 07:47
try to say @a
ohh
err
nvm
07:48 s1n joined 07:49 rurban joined 07:51 dmol joined
sorear invocation of method handles is... a lot more complicated than I though 07:51
FROGGS[mobile] :/ 07:52
07:54 JimmyZ joined 07:56 rurban left, tomyan_ left 08:00 lustlife` left, atroxaper joined 08:07 Tene_ is now known as Tene 08:11 kingbeast joined, kingbeast left 08:12 kingbeast joined 08:17 risou is now known as risou_awy 08:21 tomyan joined 08:27 flaviusb joined 08:31 flaviusb left 08:46 thou left 08:47 flaviusb joined 08:51 flaviusb left 08:54 rurban joined 08:55 Blahhh left 08:58 rurban left 09:05 raiph left 09:06 flaviusb joined 09:10 FROGGS[mobile] left, flaviusb left 09:23 PacoAir joined 09:26 flaviusb joined 09:44 cognominal joined
sorear jnthn: around? 09:51
nwc10 is this where we have a sweepstake on how much nap jnthn needs after 22 hours of uptime? 09:52
sorear maybe 09:53
nwc10 3 so far, base on his last comment in #moarvm
sorear "nap" in sorearish implies <= 2h 09:54
09:55 rurban joined 09:57 spider-mario joined 10:00 rurban left 10:02 hlin_ left 10:04 hypolin left
nwc10 anyway, I'm going to guess 3 more hours :-) 10:05
and likely be completely wrong
10:12 woosley left, anuby left
sorear .ask jnthn Have you an opinion on S04's concept of a lexotic gather/take? 10:30
yoleaux sorear: I'll pass your message to jnthn.
sorear sleep&
10:31 census joined 10:36 census left 10:51 moukeddar joined 10:57 rurban joined
tadzik labster | both of my 787 Dreamliner flights had technical problems that delayed the flight. 10:59
heh, that souns familiar :D
mine's doors broke just before takeoff 11:01
delayed by around 30 minutes. The difficulty was visible through the window, but the announcement said that it's delayed because it arrived late. Yeah. 11:02
11:05 rurban left
jnthn Hm. The nap left me feeling like I need more sleep rather than refreshed. :/ 11:07
yoleaux 10:30Z <sorear> jnthn: Have you an opinion on S04's concept of a lexotic gather/take?
colomon sounds like most of the p6ers have made it home safely by now!
jnthn (Which of course, I should not do, otherwise I won't sleep at all tonight...)
.tell sorear lexotic gather/take will need to be supported at some point, though Rakudo doesn't do that yet today so if we don't have it from the start it's not a regression compared to Rakudo on Parrot. 11:09
yoleaux jnthn: I'll pass your message to sorear.
tadzik r: sub foo { die "no"; LEAVE { say "leaving" } }; foo 11:16
camelia rakudo b2072f: OUTPUT«no␤ in sub foo at /tmp/Y1eClAKYJX:1␤ in block at /tmp/Y1eClAKYJX:1␤␤»
11:24 tgt joined 11:31 kaare_ left, kaare__ joined 11:35 kaare__ left
tadzik oh 11:42
LEAVE {...} at every block exit time (even stack unwinds from exceptions)
so, LEAVE is wrong here?
r: sub foo { LEAVE { say "leaving" }; die "NO" }; foo 11:43
camelia rakudo b2072f: OUTPUT«NO␤ in sub foo at /tmp/pS_UKiac76:1␤ in block at /tmp/pS_UKiac76:1␤␤»
tadzik it causes Panda::Common::indir() to be broken 11:45
I think
jnthn tadzik: I think the issue is that the global exception handler doesn't cause the LEAVEs to be triggered, somehow... 11:47
tadzik apparently we don't have a spectest for that 11:48
jnthn No; think there's an RT, though. 11:50
12:02 rurban joined
tadzik I don't see one 12:04
I'll report it, and come up with a spectest
12:06 rurban left 12:15 tgt left 12:16 tomyan left
Ulti jnthn: get a jawbone up so that you can take power naps jawbone.com/up#features 12:24
12:26 flaviusb left 12:36 birdwindupbird joined
jnthn std: 'abc' ~~ /a $<a>=[\w+]/; say :$<a> 12:46
camelia std 6348f35: OUTPUT«ok 00:00 45m␤»
dalek ecs: de6df31 | (Elizabeth Mattijsen)++ | S02-bits.pod:
Minor reorganisation in order of :delete combinations + some :exists combinations
 12:52
13:02 rurban joined 13:07 rurban left 13:09 wamiks left 13:10 kaare__ joined
dalek kudo/nom: d25808b | jnthn++ | src/Perl6/ (2 files):
Implement :$<foo> colonpair syntax.
 13:19
kudo/nom: f415ec0 | (Elizabeth Mattijsen)++ | src/core/Any.pm:
:exists? :k doesn't make sense, remove it.
 13:29
kudo/nom: e20d626 | (Elizabeth Mattijsen)++ | src/core/Any.pm:
Normalize candidate order for better readability
jnthn found 3 RTs to tag testneeded as well as tagging a 4th testneeded by doing :$<foo> :) 13:30
13:31 rindolf left
dalek ast: e48b1c5 | jnthn++ | S04-statements/terminator.t:
Correct quoting of { } in a test.

Since { } in a string is a closure, it caused the eval to get invalid code, causing the test to fail in Rakudo even though it should pass.		 13:36
13:37 vm joined
jnthn std: my &foo; sub foo { }; 13:38
camelia std 6348f35: OUTPUT«===SORRY!===␤Illegal redeclaration of routine 'foo' (see line 1) at /tmp/BnQEHRaeMW line 1:␤------> my &foo; sub foo⏏ { };␤Check failed␤FAILED 00:00 45m␤»
jnthn r: my &foo; sub foo { };
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Redeclaration of routine foo␤at /tmp/5Az3XW4y9z:1␤------> my &foo; sub foo { }⏏;␤ expecting any of:␤ statement list␤ horizontal whitespace␤»
13:41 moukeddar left 13:45 vm left
dalek ast: ef55c9b | jnthn++ | S32-exceptions/misc.t:
Unskip a now-passing test.
 13:46
kudo/nom: 7567877 | jnthn++ | src/Perl6/Actions.nqp:
Fix LTA error on 'my &foo; multi foo() { }'.
 13:47
13:47 konundra joined 13:53 woolfy left 13:54 lizmat left 13:55 prevost joined 13:59 lizmat joined 14:00 woolfy joined 14:03 rurban joined 14:07 rurban left 14:08 rurban joined 14:13 rindolf joined
masak oh hai, #perl6 14:16
yoleaux 02:10Z <[Coke]> masak: if he can provide an RT search that shows tickets that are NOT marked testneeded.
masak [Coke]: "Edit search" > "Tag" > "isn't" > "testneeded" > "Add and search". 14:17
dalek mentaler: 9f7ecd1 | tadzik++ | smoker:
Panda is now fixed. Adjust
 14:18
nda: be1030f | tadzik++ | lib/Panda (5 files):
Don't leak out non-Panda exceptions.

This fixes emmentaler. It's also a part of my experiment to use Failures where I usually used Exceptions. Please bare with me.		 14:19
tadzik colomon: there, I got this
masak tadzik: "bare with me" means "get nekkid with me". did you mean "bear with me"? :P
tadzik oh man 14:20
jnthn :D
colomon tadzik++
dalek nda: 4dc0da5 | tadzik++ | lib/Panda (5 files):
Don't leak out non-Panda exceptions.

This fixes emmentaler. It's also a part of my experiment to use Failures where I usually used Exceptions. Please bear with me.
colomon tadzik: no one would ever use "bare with me" with that meaning, though. :)
tadzik there, no one noticed :D
colomon tadzik: I'll give it a test in a few minutes.
masak retconned!
tadzik I may have broken your git history, if you're very fast ;) 14:21
moritz the logs will know.
jnthn It'll go down in history
14:21 spider-mario left
tadzik the logs always remember. Like eleghants 14:21
masak .oO( no-one is that fast. taking off your shirt takes at least a few seconds )
jnthn The day tadzik said "bare with me"
nwc10 masak: where is "Tag" ?
nwc10 is trying to follow those instructions
masak nwc10: it's hard to describe the exact position. some ways down, near the bottom.
moritz wonders why tadzik wants a bear with him 14:22
masak moritz: exit left, chased by a bare. ;)
nwc10 masak: I don't see it at all on rt.perl.org/rt3/Search/Build.html (for me, logged in) and it's not in the page source
masak nwc10: I see it. 14:23
we must have different views of that form.
it's the third one from the bottom in the "Add Criteria" pane.
nwc10 I only have time worked, time estimated, time left
masak maybe the controls depend on how much admin you have?
nwc10 maybe. 14:24
masak anyway, if people don't have the full toolset, that's another way I can understand that they don't get along with RT...
nwc10 if so, that might explain why [Coke] can't manage to do it
masak (though it also doesn't make me more sympathetic or prone to defent RT)
defend*
14:25 woolfy left
masak rn: class A { has $.name = (state $ = "a")++ }; say A.new.name for ^3 14:27
camelia rakudo b2072f, niecza v24-70-g6314dd4: OUTPUT«a␤b␤c␤»
masak I like that solution? does anyone know a better one?
14:27 census joined
masak s/\?/./ 14:27
jnthn can't think of one right away 14:30
nwc10 masak: if you build the search for "not tagged testneeded" does it end up with a URL that anyone else can use?
timotimo_ hello everyone. what's new? :) 14:31
masak timotimo_! \o/
14:31 ajr joined
lizmat masak: fwiw, that seems to need some syntactic sugar, like "has state $.name = 'a'" or "state has $.name = 'a'" 14:31
masak nwc10: let's test. hold on.
nwc10 timotimo_: what's new since when?
14:31 ajr is now known as Guest87187
jnthn timotimo_: 5-6 extra RTs are now "testneeded" ;) 14:32
14:32 Guest87187 is now known as ajr_
masak lizmat: both "has" and "state" are declarators, so that feels strange to me. a bit like if you'd write "my our". 14:32
timotimo_ nwc10: about the last 24 hours 14:33
nwc10 is this like christmas tree IP packets? :-)
moritz lizmat: if you want a state 'has', you should really use a class variable
nwc10 timotimo_: sorear fixed all the remaining test failures for NQP-JVM. Is that more or less than 24 hours ago?
timotimo_ i think i've heard that, but it may have been newer
masak nwc10: rt.perl.org/rt3/Search/Results.htm...stors__%3C
lizmat then why does masak want to do it like that?
jnthn masak: I think pmichaud++ fixed the issue in rt.perl.org/rt3/Ticket/Display.html?id=88128 - I'm guessing it's a dupe ticket. Something to look at and maybe an easy close if you have a moment. :) 14:34
masak moritz: please show what you mean.
jnthn: ok.
TimToady we are trying very hard not to take a nap here in Manchester...well, really Sale
masak is on a train, and so has plenty of moments :)
jnthn
.oO( is everything cheap in the Sale? :) )
nwc10 masak: I can't use that, but I'm not sure if it's getting chewed in the paste. What happens with a link shortener?
TimToady part of that was taking a walk in Worthington Park :)
jnthn :D
masak jnthn: I was going with "you are the Sale of the Earth" :P 14:35
lizmat we are about to close the suitcases and be on our way to Houston, then Amsterdam
masak TimToady: walking in Worthington Park made you almost fall asleep? :P
TimToady afterwards it did :)
nwc10 masak: actually, I think that it got truncated. Last bit I see is %3Csmall%3E__Requestors__%3C 14:36
TimToady but I think we got a suffient amount of sunshine to persuade our circadia that they were rythmic
*cient
*rhyth 14:37
hoelzro jnthn: ah, there you are! would you be adverse to reverting the commit 721f127a83bdfef9f9f2e6dd46a7850899cb2858 in nqp?
masak nwc10: ok, making tinyurl
jnthn hoelzro: I can't Just Revert It, otherwise we break the nqp-jvm build. 14:38
hoelzro I see
is there a way we can restores its functionality for just Parrot, then?
jnthn (due to the pir::)
Yeah, wrap the thing in a #?if parrot ... #?endif I guess.
hoelzro ok, cool
jnthn Hm, duck mapping went away, right? So rt.perl.org/rt3/Ticket/Display.html?id=102276 can be rejected? 14:39
masak nwc10: tinyurl.com/notestneeded
nwc10 masak: I have a page titled "Found 665 tickets" 14:40
masak that's the one.
nwc10 oh, *interesting*, if I got to edit search I now have "Tag", "Platform" and "Patch Status" as dropdowns
TimToady hyper doesn't need duckmapping, but there's still a .duckmap method specced
nwc10 and if I go "new search" they disappear. 14:41
masak nwc10: might be you were running RT in logged-out view.
nwc10 nope, I'm logged in
masak or something.
nwc10 .tell [Coke] Try this: < masak> nwc10: tinyurl.com/notestneeded 14:42
yoleaux nwc10: I'll pass your message to [Coke].
masak moritz: what's a Perl 6 "class variable", and how would I use it to do what I wanted (increment letter with each new instance, provide accessor)? 14:43
nwc10 and I think [Coke] needs to ask perlbug-admin at perl.org what's with the funkiness
14:44 census left
moritz r: class A { our $.c = 0; method new(|) { ++$.c; self.bless(*) } };A.new for ^5; say A.c 14:45
camelia rakudo b2072f: OUTPUT«5␤»
moritz masak: ^^ 14:46
I should have said ↑↑ :-)
masak moritz: that would not give each instance a unique $.c
dalek kudo/nom: 6b1a671 | jnthn++ | src/Perl6/Grammar.nqp:
Start checking reduce for fiddliness.
 14:47
masak we seem to be speaking at cross-usecases.
lizmat good * #perl6, we're on our way now! 14:48
14:48 lizmat left
masak moritz: anyway, I had to abandon the refactor that included the 'has = state' thing, because it turned out that I wanted the count to be scoped not globally, but to another bigger thing which I have not reified into a class. (it's just a lexical scope right now.) 14:49
(bigger thing than the class I was writing, not bigger than global scope) :)
14:50 FROGGS[mobile] joined
masak jnthn: merged #117831 into #88128. (pmichaud's fix of infix:<**> capturing separators) 14:53
14:53 pecastro joined
jnthn hm, and that ticket claims to have tests... 14:54
...but it's still not closed.
masak maybe the merge reopened it? :/
yes, looks like. resolving ticket. 14:55
jnthn masak++
masak I think I assumed that RT would prefer the latest state-change. but it might be it's conservative in some way and preserves the most unresolved state, or something. 14:56
dalek kudo/nom: b2e8375 | jnthn++ | src/Perl6/ (2 files):
Eliminate a now-unrequired hack.

This as a side-effect fixes RT#111734.		 15:00
hoelzro yay! NQP pull request: github.com/perl6/nqp/pull/102
dalek p: 831cb3a | (Rob Hoelz)++ | src/NQP/World.nqp:
Restore dynext option processing for Parrot
 15:03
p: 2e79dae | jonathan++ | src/NQP/World.nqp:
Merge pull request #102 from hoelzro/master

Restore --dynext processing for Parrot
hoelzro \o/
thanks jnthn 15:04
masak r: constant FOO = q:to /EOF/;␤ OH HAI␤ EOF␤␤say FOO
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Premature heredoc consumption␤»
masak what in the world does "Premature heredoc consumption" mean? :(
is it unreasonable to want to put a heredoc string in a constant? I don't think so. 15:05
colomon std: constant FOO = q:to /EOF/;␤ OH HAI␤ EOF␤␤say FOO 15:07
camelia std 6348f35: OUTPUT«===SORRY!===␤(Possible runaway string from line 2)␤Couldn't find delimiter for heredoc␤ at /tmp/oNSPLuODMp line 5 (EOF):␤------> say FOO⏏<EOL>␤Parse failed␤FAILED 00:00 42m␤»
jnthn masak: It means that constants are evaluated immediately 15:08
masak: It's the BEGIN block / heredoc thing in disguise.
masak jnthn: I fail to understand. I know that constants evaluate immediately. but my heredocs are constant; they aren't qq strings, they're q strings.
jnthn masak: The problem being that expression after the = up to the ; is evaluated right away at statement end 15:09
masak: The heredoc itself hasn't been parsed at that point.
masak ah.
jnthn r: BEGIN say q:to /EOF/;␤ OH HAI␤ EOF␤␤say FOO 15:10
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Premature heredoc consumption␤»
masak I understand now.
jnthn Same as there.
masak still dissatisfied with that semantics.
jnthn Well, not sure what we can do about it.
Given
masak feels like Perl 6 isn't doing enough for me :)
we know that there is *going* to be a string there, in a very short while.
jnthn r: (BEGIN say q:to /EOF/; say 42)␤ OH HAI␤ EOF␤␤say FOO
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Premature heredoc consumption␤»
jnthn There the BEGIN isn't even a top-level statement. 15:11
Well, feel free to play with it to see if you can work out a sane way to make it work
jnthn isn't sure how
masak oki. gotcha.
jnthn I mean, I'd be happy enough with it working. I just don't see how to get there.
masak yeah, I'm not saying it's an easy problem to solve. I'm saying it feels like it could work. 15:12
masak submits rakudobug
dalek ast: 1415743 | jnthn++ | S12-introspection/methods.t:
Extra .^methods tests.
15:16 kaare__ left
dalek ast: b4d71e8 | jnthn++ | S12-introspection/methods.t:
Test for RT #108968.
 15:16
jnthn r: try { CATCH { ~$! }; die } 15:19
camelia rakudo b2072f: OUTPUT«Nominal type check failed for parameter ''; expected Any but got Mu instead␤ in block at /tmp/D4IDicmIrw:1␤ in block at /tmp/D4IDicmIrw:1␤␤»
15:21 raiph joined 15:24 kaare_ joined
jnthn masak: That constants one is a dupe, it seems... 15:29
masak: with rt.perl.org/rt3/Ticket/Display.html?id=117853
masak ok, merging. jnthn++ 15:32
15:32 prevost left 15:33 thou joined
dalek kudo/nom: 3f49bbd | coke++ | src/core/Any.pm:
first should return Nil on not found.

RT #118141		 15:40
jnthn shop; bbiab 15:42
dalek kudo/nom: ff94231 | jnthn++ | src/Perl6/Metamodel/MultipleInheritance.nqp:
Fix self-inheritance detection.
 15:43
15:44 prevost joined
[Coke] masak: that doesn't work. 15:45
yoleaux 14:42Z <nwc10> [Coke]: Try this: < masak> nwc10: tinyurl.com/notestneeded
nwc10 *think* of the devil. (ish)
[Coke] masak: if you pick that via the build a search screen, you get about 660-665 tickets - that list includes, e.g. 73104, which is marked testneeded. 15:47
nwc10 [Coke]: oh, interesting. I can't even *build* that query via the search screen
15:48 rurban left
[Coke] Queue = 'perl6' AND ( Status = 'new' OR Status = 'open' ) AND 'CF.{Tag}' != 'testneeded' 15:48
doesn't work with != - doesn't work with "not like" 15:49
moritz note that you have to be logged in for most RT search features (you probably are; justin case)
15:50 AndChat|356841 joined
[Coke] ... Ah. I thought this complaint of mine sounded familiar. I already opened this ticket with robert in 2011. 15:51
Queue = 'perl6' AND ( Status = 'open' OR Status = 'new' OR Status = 'stalled' ) AND 'CF.perl6.{Tag}' NOT LIKE 'testneeded' 15:53
Note the ".perl6."
15:53 FROGGS[mobile] left
[Coke] which never ever ever would show up via the interface. you have to know to add it by hand when editing the query in advanced mode. 15:53
I must have updated my "notdone" query at some point in the past 2 years forgetting this. :( 15:54
16:02 rurban joined 16:04 AndChat|356841 left, tomyan joined, FROGGS[mobile] joined, rurban left, FROGGS[mobile] left 16:05 FROGGS[mobile] joined, rurban joined 16:07 JimmyZ left
[Coke] is it ok if NaN.Rat returns itself? (rather than something that is actually a Rat?) 16:13
colomon I'd say yes (at least for now). Is that Num's NaN you're talking about? 16:15
colomon is still tempted to think that NaN should be a type of its own. 16:16
[Coke] Yah. I'll just get this test passing for now. 16:18
16:19 FROGGS[mobile] left
dalek kudo/nom: 5075c3b | thundergnat++ | tools/build/gen-version.pl:
Change quoting to capture the version correctly under Windows.

Windows Command.exe does not understand single quoted parameters so perl6 -v returned no version number.		 16:19
kudo/nom: 58f5ae6 | moritz++ | tools/build/gen-version.pl:
Merge pull request #162 from thundergnat/patch-1

Change quoting to capture the version correctly under Windows.
moritz we could encode NaN rats as 0/0, Inf as 1/0 and -Inf as -1/0 16:20
colomon moritz: true. but we don't have Int NaN or Inf either, do we? 16:27
dalek kudo/nom: a503510 | coke++ | src/core/Num.pm:
NaN.Rat -> NaN

RT #77820		 16:29
16:29 rurban left
dalek ast: 97d8539 | coke++ | S32-num/rat.t:
rakudo unfudge
 16:30
flussence r: my \FOO = q:to /EOF/;␤ OH HAI␤ EOF␤␤say FOO; FOO ~= '...'; 16:35
camelia rakudo b2072f: OUTPUT«OH HAI␤␤Cannot modify an immutable value␤ in block at src/gen/CORE.setting:13979␤ in block at /tmp/VHU39ugLa9:5␤␤»
flussence masak: is that close enough for you? :)
[Coke] Ugh. the search is still returning *some* tickets that are marked testneeded. 16:38
e.g. 76400
(*#&$#$> 16:39
16:45 PacoAir left
[Coke] opening new ticket with rt admins. 16:46
masak - things like this make it very frustrating to use RT.
moritz does anybody know a non-annoying bug tracker? :-) 16:49
16:49 dmol left
nwc10 they are *all* made of software, aren't they? This isn't a good start. 16:50
colomon most of the bug trackers I know track both annoying and non-annoying bugs.
moritz hardware bug trackers... 16:51
at $work we have a highly patched and customized OTRS installation, which is somewhat usable
(after 5 years of patching, that is...)
nwc10 but so forked, that it's un-upgradeable? 16:52
moritz the 3.0 to 3.1 upgrade was several man weeks worth of work 16:53
3.1 to 3.2 was less painful
and now upstream changed to git (used to be cvs)
so we'll likely upgrade more often and in smaller steps, keeping them less painful 16:54
and maybe push some contributions upstream
16:56 konundra left 16:57 rurban joined 17:02 birdwindupbird left
[Coke] r: 5. 17:04
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Confused␤at /tmp/ogJuVhXpa8:1␤------> 5.⏏<EOL>␤ expecting any of:␤ dotty method or postfix␤»
17:11 tgt joined
diakopter r: 5.␤ 17:15
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Decimal point must be followed by digit␤at /tmp/zVobw9UfLN:1␤------> 5.⏏<EOL>␤Decimal point must be followed by digit␤at /tmp/zVobw9UfLN:1␤------> 5.⏏<EOL>␤Unsupported use of . to concatenate stri…
diakopter [Coke]: the difference in the repl is just the newline 17:16
nwc10 wonders where mberends is 17:19
tadzik he's not often here these days 17:20
17:21 ajr_ left, rurban left
timotimo_ i'm thinking about subtitling my perl6 talk in english for y'all (and te rest of the interwebs) 17:21
nwc10 I wonder if he even knows about MoarVM yet 17:22
or the state of the JVM work
timotimo_ what's the occasion to wonder about him? 17:23
17:23 rurban joined
nwc10 nothing particularly special *today* 17:23
[Coke] diakopter: seems like we shouldn't act differently if the program has a trailing newline. 17:24
17:24 raiph left 17:25 atroxaper left 17:28 rurban left, tgt left 17:45 spider-mario joined
[Coke] rt.perl.org/rt3/Ticket/Display.html?id=105004 - i disagree that saying nothing else derives from it means that nothing else CAN derive from it. 17:48
r: class A is Whatever { ... }
camelia rakudo b2072f: OUTPUT«===SORRY!===␤The following packages were stubbed but not defined:␤ A␤at /tmp/ZH75OXcek5:1␤------> class A is Whatever { ... }⏏<EOL>␤ expecting any of:␤ argument list␤ postfix␤ statement end␤ sta… 17:49
[Coke] r: class A is Whatever { }
camelia rakudo b2072f: ( no output )
17:50 SHODAN left 17:51 tgt joined 17:52 SHODAN joined 17:53 tgt left 17:55 census joined
jnthn [Coke]: agree 18:00
colomon tadzik: ping? 18:09
tadzik colomon: pong?
colomon tadzik: is there some standard way to update a panda install?
tadzik hmm
colomon I'm trying to do it and not succeeding at all.
tadzik other than rebootstrap.pl?
panda install panda should work too
colomon pandabrew/bin/pandabrew panda install panda===SORRY!=== 18:10
Missing or wrong version of dependency 'lib/Shell/Command.pm'
tadzik bah
rebootstrap is the way to go, then
colomon that's after I did pandabrew rebootstrap
tadzik oh
well, crap
maybe pandabrew is broken too :/
colomon which seemed to work but didn't leave me with an update-to-date panda
tadzik I broke all the things, apparently
colomon trying a fresh init of pandabrew 18:12
==> Bootstrapping Panda 18:13
===SORRY!===
Missing or wrong version of dependency 'lib/Shell/Command.pm'
timotimo_ delete all blib folders? :S
colomon down below the rakudo directory, or somewhere else? 18:14
timotimo_ good question
sorear good * #perl6
yoleaux 11:09Z <jnthn> sorear: lexotic gather/take will need to be supported at some point, though Rakudo doesn't do that yet today so if we don't have it from the start it's not a regression compared to Rakudo on Parrot.
labster good morning, sorear, * 18:15
tadzik colomon: it may be that I broke it :( 18:16
don't have time to look at it right now
colomon tadzik: I warn you that I may fix it. ;) 18:20
18:23 rurban joined 18:25 rurban1 joined 18:28 rurban left
colomon is finding this bug VERY frustrating 18:29
18:29 raiph joined 18:36 kingbeast left
masak [Coke]: agreed. what you're experiencing with RT is inexcusably bad, and I have no interest in defending it. 18:41
18:53 census left, benabik left 18:54 benabik joined
TimToady has no interest in staying away any longer... 18:59
*awake, grr
18:59 PacoAir joined
timotimo_ stay away from sleep? :) 18:59
19:05 Guest1337 joined 19:08 dmol joined 19:12 tomyan left
masak I found okmij.org/ftp/ML/generalization.html very interesting, and others here might too. it draws parallels between HM type checking and GC dependency tracking. 19:15
timotimo_ what's HM mean? 19:17
19:18 skids left
masak timotimo_: en.wikipedia.org/wiki/Hindley%E2%80%93Milner 19:20
19:20 skids joined 19:24 tomyan joined
sorear okmij! \o/ 19:25
hi masak
19:26 rindolf left
masak hejhej, sorear ;) 19:27
19:27 rindolf joined
sorear oh, it's not just hehe? 19:33
sorear is not super up to speed on swedish phonology
19:39 Guest1337 left
sorear starts putting togeter a gist of design decisions for jnthn 19:39
19:46 census joined 19:47 kaare_ left, tomyan left
masak sorear: all Swedish vowels are monophthongs. 19:48
jnthn
.oO( even in Skane? )
 19:49
masak hm, no :)
jnthn ;-)
masak ...but I was talking about *Swedish*... :P 19:50
19:50 raiph left 19:51 raiph joined
sorear jnthn: about how much time do I have to finish this and get you to read it while you're still awake? 19:51
jnthn sorear: tbh, if it needs me to think, it'd be better for me to look at it when I wake up tomorrow 19:52
sorear: I'm physically awake at the moment but have the concentration of a goldfish... :)
sorear tehee 19:55
arnsholt was lucky and got some melatonin on the flight back 19:57
I'm a bit weird today, but not as bad as I feared, by any means
sorear jnthn: ok, well, let me just ask the one thing that I'm sort of blocking on: I want to expose saved states to NQP/QAST as a SixModelObject, ideally directly invocable but I'll settle for needing a nqp::cont_invoke op. what should I use as a model for this? do I need to do anything with HLLConfig/KnowHOWBootstrapper?
arnsholt: \o/ safe and sound at home in/near Oslo? 19:58
jnthn sorear: KnowHOWMethods is probably more useful 19:59
arnsholt Yeah, arrived yesterday morning (or afternoon, US time)
colomon o/ 20:00
jnthn sorear: Shows how to get a CodeRef that points to something of your choice.
arnsholt The flight wasn't too boring either, since I slept almost the entire time =D
\o colomon
jnthn sorear: Currently, all invokable things are either a CodeRef directly, or something with an InvocationSpec in the STable that knows how to produce one. 20:02
(Which may be by looking up an attribute, or saying "oh, call this thing instead") 20:03
20:05 mrallen1 joined
sorear so where would I create the STable for the saved states? 20:05
jnthn Create...what? :) 20:06
STables are always created as part of creating a new type, never "directly"
nqp::newtype is the op typically used to do that.
I'm not sure I follow what you're trying to do, though.
Oh, you want a different type that captures the saved states? 20:07
If so, and you want whatever types in there, it may be easiest to create a different REPR, and go with an nqp::continvoke op that knows what to do with it. 20:10
sorear if the saved states are to be exposed to NQP, presumably they need STables, ya?
20:10 tomyan joined
jnthn Well, they need to be in some kind of SixModelObject 20:10
And that has to have a type, and s-tables are per type. 20:11
sorear I don't really care much what they look like as long as they can be stored and passed around by NQP code
jnthn OK
colomon .ask tadzik I deleted rakudo and reinstalled it from scratch, and now panda is happily working again.
yoleaux colomon: I'll pass your message to tadzik.
jnthn Then create a REPR that looks as you wish, then see where BOOTCode is created and create a BOOTCont the same way.
sorear so presumably I need one STable for all my saved state objects
OK
jnthn (The BOOTCode setup is the step that creates the STable) 20:12
20:19 PacoAir left 20:22 pecastro left 20:23 census left
tadzik yoleaux: pass me my message 20:30
yoleaux 20:11Z <colomon> tadzik: I deleted rakudo and reinstalled it from scratch, and now panda is happily working again.
tadzik colomon: well, good to hear :)
colomon tadzik: you've still got a sort of bug in there: just got No such method 'stage' for invocant of type 'X::IO::Unlink'
during the smoke test
tadzik fffuu
colomon: what module does that? 20:31
colomon Yarn
tadzik or, near what module does that happen?
hmm
I did a full smoke before I commited my changes
sorear tadzik! o/ 20:32
colomon tadzik: related issue: you changed my smoker patch so that it once again does not catch exceptions other than X::Panda. Only now that's implicit instead of explicit.
tadzik sorear: hey :) 20:34
colomon: I wanted it to be implicit
or, well, no
I wanted it to crash smoker
so that we may notice if anything else is broken
colomon tadzik: ah. well, this certainly does that. 20:35
tadzik yep :)
I need to attend to some $university stuff, I'll look at it tomorrow maybe
sorear nqp: (-> $x { say($x+1) })(2) 20:36
camelia nqp: OUTPUT«3␤»
sorear nqp-jvm: (-> $x { say($x+1) })(2)
camelia nqp-jvm: OUTPUT«3␤»
lue Is there any particular reason why I can't do augment class A { supersede method B { } } ? (this would be to, say, modify the messages from Perl 6's X:: classes) 20:45
timotimo_ are you going to work on X::Offensive? :D 20:46
fwiw, i think you can does a role onto the class, can't you?
masak lue: just 'method B' should work fine, if you ask me.
no 'supersede' necessary. 20:47
lue r: use MONKEY_TYPING; class A { method B { say "hi" } }; A.new.B; { augment class A { method B { say "O" } }; A.new.B; }
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Package 'A' already has a method 'B' (did you mean to declare a multi-method?)␤at /tmp/sjoBWUoKn5:1␤------> ␤»
timotimo_ r: role offend { sub message { "you are stupid if you think that's going to work" } }; X::Redeclaration does offend; eval q{my &a; multi a { };};
camelia rakudo b2072f: OUTPUT«Cannot use 'does' operator with a type object.␤ in sub infix:<does> at src/gen/CORE.setting:13624␤ in sub infix:<does> at src/gen/CORE.setting:13606␤ in block at /tmp/Qtrw3PI93X:1␤␤»
timotimo_ ah, so that doesn't work.
labster timotimo_'s classes don't follow the CoC. 20:48
timotimo_ r: use MONKEY_TYPING; role offend { sub message { "you are stupid if you think that's going to work" } }; augment class X::Redeclaration { does offend }; eval q{my &a; multi a { };};
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Undeclared routine:␤ does used at line 1␤␤»
timotimo_ isn't there a way to put roles into the class body?
r: use MONKEY_TYPING; role offend { sub message { "you are stupid if you think that's going to work" } }; augment class X::Redeclaration does offend { }; eval q{my &a; multi a { };};
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Method 'is_dispatcher' not found for invocant of class 'Any'␤»
masak lue: that may or may not be correct behavior. I'm not sure. 20:49
timotimo_ oh, whoops. i recongize wree that comes from. jnthn just improved LTA for self-deriving classes or something and added that check
lue frequently finds wanting much better supersede and augment keywords 20:50
*finds himself
.oO(This makes it really hard to create things like X::Condescending and X::NothingsWrongWhyDoYouAsk)
 20:51
labster I never saw a real answer to my question last night: should the error get thrown here? 20:53
r: sub foo { X::IO::Dir.new(path=>'a', os-error=>'b').fail }; my @a := foo(); 1;
camelia rakudo b2072f: OUTPUT«Failed to get the directory contents of 'a': b␤ in method sink at src/gen/CORE.setting:10169␤ in method BUILDALL at src/gen/CORE.setting:801␤ in method bless at src/gen/CORE.setting:743␤ in method new at src/gen/CORE.setting:728␤ in method new at src/gen/CORE.…
timotimo_ there's a bug about that feature
pm also highlighted that in his parallelism talk
20:57 twigel left 21:00 mrallen1 left
masak hm, it'd be good if $str.subst in void context warned... :) 21:00
timotimo_ as in: "subst is not mutative"? 21:01
colomon perhaps it was called for it's side effects. 21:03
colomon should be careful with sarcasm...
labster r: sub foo { X::IO::Dir.new(path=>'a', os-error=>'b').fail }; my $a := foo(); 1; 21:05
camelia rakudo b2072f: ( no output )
21:06 lue left
labster okay, I'm calling the array binding thing a rakudobug 21:07
21:08 lue joined 21:10 Guest1337 joined
timotimo_ r: say "foo 'bar'" ~~ s/"'"/"\'"/ 21:10
camelia rakudo b2072f: OUTPUT«Cannot modify an immutable value␤ in sub infix:<=> at src/gen/CORE.setting:13602␤ in block at /tmp/sD0100r5kH:1␤␤»
timotimo_ er ...
r: given "foo 'bar'" { say s/"'"/"\'"/ }
camelia rakudo b2072f: OUTPUT«Cannot modify an immutable value␤ in sub infix:<=> at src/gen/CORE.setting:13602␤ in block at /tmp/h64heunBN2:1␤␤»
timotimo_ someone's missing an "is copy" or something? 21:11
jnthn No; use .subst for this case
21:12 rurban1 left
timotimo_ does it with trans instead 21:12
hm, that's not right
jnthn sleep & 21:17
21:18 lizmat joined
lizmat *sigh* we missed our plane, so we will have to enjoy Texas for one more day 21:21
tadzik aw
lizmat torrential rains caused the highway to flood, so it took us 1 hour for the last 3 miles to the airport
enough to miss our window with about 5 minutes :-(
labster the last mile is always the slowest 21:22
lizmat so I guess I'll be hacking some more today
labster lizmat++
tadzik the rain flooded subawy here in warsaw today
tnij.org/vzmu
err, tnij.org/vzmu clickable
lizmat get an early sleep so that we're better on track when we get back in NL] 21:23
at the deepest point, there was about 35 cm of water on the highway… our Dodge could take that, but slowly :-) 21:24
anyways, first some dinner& 21:25
21:26 spider-mario left
masak oh noes not water in the subway :( 21:28
timotimo_ is there a module that properly does shell-quoting? >_<
because i totally escaped ' inside '' quotes as \' and when it didn't work i tried to escape ( and ) with \es ... 21:29
sorear proper shell quoting...for which shell? :/
timotimo_ well, if you use the "shell" command, you get /bin/sh, no? how many systems have /bin/sh as csh or zsh or fish? 21:30
sorear my /bin/sh is bash 21:31
also doesn't it use $SHELL ?
timotimo_ don't think so
sh: nosuchcommand: command not found
my $SHELL is /bin/zsh, though
sorear tadzik: did you take that picture? 21:32
geekosaur I can't imagine \e ever being right for shell quoting
tadzik sorear: nope 21:33
sorear: for what is worth, I can only assume it's authentic, and taken today
but it seems to come from tvn, which is a news network 21:34
21:39 woolfy joined 21:48 raiph left
labster I just read S22. There's .JIB files and CPAN6? o.O 21:51
tadzik ...yeah.. 21:53
flussence that file is mostly conjecture and it hasn't been touched for 2.5 years
eternaleye Sadly, nobody seems to have taken a liking to the cut of .JIB :P
21:54 dmol left
flussence and there is a CPAN6, but I've never seen any activity from it in all the time I've been here 21:54
lue also, the CPAN6.org site is very old. "CPAN6 is not even usable on the moment (as is Perl6)" indeed.
labster more like 4 years ago 21:55
lue
.oO(Panda + modules.perl6.org is a better CPAN6 than CPAN6)
 21:56
labster www.theregister.co.uk/2013/06/08/fa...sus_cloud/ Finally, my meteorology experience is useful in cloud computing. 21:59
masak :) 22:00
22:03 lustlife left
sorear at what point do we send a mail to MARKOV and politely ask him to remove the site? :/ 22:06
22:08 tgt joined
masak won't be long now, I think. 22:08
at least remove the formulation that Perl 6 isn't usable at the moment. 22:09
22:12 rurban joined 22:25 census joined, raiph joined
Tene timotimo_: \ is not a significant character inside '' quotes. The ONLY meaningful character inside '' quotes is ' which means "end quote". You escape ' as '\'' 22:27
timotimo_ no 22:29
you escape ' as ''
Tene No, that just parses as two adjacent quoted strings.
lue S02:4519 : "Only range operations will be compromised by an unknown foreign collation order, such as a system that sorts "delta" after "gamma"." But... isn't delta after gamma in the Greek alphabet?
Tene Try it and see. 22:30
timotimo_ oh, it must be zsh-specific. 22:31
but! 22:32
Tene Yes, that's not sh quoting.
timotimo_ echo 'foo\'bar' will end the line and give me a prompt for another line, so that didn't work
(in sh)
and echo 'foo\'bar'' will give me foo\bar
lue Am I right in thinking that should be "'delta' before 'gamma'" instead? 22:33
Tene Yes, because as I said, \ is NOT an escape character in '' quotes. There are no escape chars in '' quotes. That is three tokens, 'foo\' bar '
In '' quotes, all characters are literal except for ' which only means "end quote" and nothing else.
colomon en.wikipedia.org/wiki/Greek_alphabet -- looks like it to me, lue
timotimo_ oh yikes 22:34
Tene OH, you read my initial example as the entire quote, not as just the escape
timotimo_ yes, i did. fail me :) 22:35
Tene For example "don't" would be 'don'\''t'
timotimo_ good example.
Tene That makes way more sense now. Sorry about that
22:35 rurban left 22:36 tomyan left
masak that 'delta' comes after 'gamma' feels perfectly natural to me. probably because I ingested a chunk of Russian at an early impressionable age :) 22:39
masak .oO( a be ve ge de... )
dalek ecs: d259755 | lue++ | S02-bits.pod:
[S02] Correct minor thinko.

  Delta does, in fact, come after gamma normally, not vice versa.		 22:40
lue Me too, because I for some reason have memorized the first four letters of the greek alphabet: alpha beta gamma delta 22:41
geekosaur ...epsilon zeta eta...
masak it does match the collation order of the Cyrillic script quite closely. 22:43
Juerd ...theta iota kappa lambda...
masak: Same for the latin script. 22:44
masak troo.
except for that "g d" thing.
Juerd ABDEIKLMNOPRSTU
phi => v could work too. 22:45
masak hey, where did G go?
lue Also, does anyone around know about the (perlcabal.org|feather.perl6.nl)/syn spec generator? I don't think it's working (S02/Literals/Interpolation Rules/Twigils still contains a duplicate of the first Twigils section (whereas in the spec Twigils #2 it's appropriately whittled down)
Juerd masak: Not listed because it's not in the same relative position in both alphabets :)
masak ah :) 22:46
sorear timotimo_: as far as I'm concerned, if you ever want to programmatically shell-quote something, you're doing it wrong.
22:46 rindolf left
timotimo_ i could have ported ruby-parrot (what was that called again?) to use 6model, so it could be used with rakudo again ... 22:47
Tene Cardinal 22:48
timotimo_ that one
Tene I started on that once; I wonder if the branch is still there.
timotimo_ is that like 1 year or maybe 2 years? 22:49
22:49 tgt left
Tene I don't understand the question. 22:49
timotimo_ ago* 22:50
Tene No idea.
timotimo_: You can find it on maybe this issue, or one linked from it: github.com/cardinal/cardinal/issues/41 22:55
sorear heya Tene 22:56
Tene sup sorear
sorear I got JNQP passing all tests! so excited.
Tene :D
sorear currently working on delimited continuations for NQP
*JNQP 22:57
will use that to implement GatherIter later
masak \o/
22:57 tomyan joined
Tene Hmm. I wonder if there's anything in the rakudo repo still written by me. 22:57
22:59 dalek left, census left
sorear BinGOs++ 22:59
22:59 tomyan left 23:00 dalek joined, ChanServ sets mode: +v dalek, census joined
diakopter BinGOs++ 23:02
masak Tene: it seems that most of your code was PIR, which has been mostly reworked into NQP code.
diakopter sorear: have you tried using clang on macosx to build moarvm?
23:03 tomyan joined, Kelder left
diakopter I'm curious how quickly it compiles it compared to gcc 23:03
sorear: are you using stack saving for the continuations? 23:04
sorear diakopter: i've not tried to build moarvm 23:05
however, my cc is clang, so when I do ... :) 23:06
23:06 tomyan left 23:07 yoleaux left
sorear diakopter: (stack saving) "it's complicated" 23:09
diakopter well it's just I was discussing it with jnthn Friday night and that's how he wanted it to be done in jnqp 23:12
sorear: ^
sorear diakopter: I'm following jnthn's plan fairly closely; we discussed that on sunday, I think 23:13
23:13 yoleaux joined, ChanServ sets mode: +v yoleaux
diakopter sorear: today? 23:16
sorear diakopter: hackathon day 2 of 4; a week ago 23:17
23:20 Ayiko joined
diakopter sorear: by "it's complicated" do you mean you don't want to talk about it? 23:20
.. because it didn't answer my question, and I still wanted to know the answer
sorear diakopter: I thought "I'm following jnthn's plan fairly closely" would be a good enough answer 23:22
it's hard to explain, and if jnthn already explained it directly to you, I don't want to duplicate that
23:25 Blahhh joined 23:26 Blahhh left 23:27 Blahhh joined, Blahhh left
dalek kudo/nom: d4e9ae8 | (Elizabeth Mattijsen)++ | src/core/Any.pm:
Make {}:kv {}:p {}:k {}:v behave according to spec
 23:28
23:28 Blahhh joined, Blahhh left 23:30 raiph left
lizmat masak, sorear: I'll contact markov when I'm back in NL 23:30
23:32 wbill joined, wbill left
lue for the record, my only concern with CPAN6 is the "Perl 6 is not useable" remark. Everything else is fine by me. 23:32
masak +1
lizmat++ 23:33
23:34 rurban joined, wbill joined, wbill left 23:35 wbill joined 23:36 wbill left 23:37 wbill joined, wbill left 23:38 wbill joined, rurban left, rurban joined 23:39 pecastro joined
dalek ast: 2f1a6eb | (Elizabeth Mattijsen)++ | S32-hash/exists-adverb.t:
Unfudge {}:exists tests that now work
 23:41
23:51 Chillance left
colomon .say tadzik I hacked around the error so I could get the rest of the smoke test run in. There is something interesting going on here... Several tests are reaching the "tests failing" stage even though their prereqs are failing *their* tests. DateTime::Utils is an example. 23:52
.say tadzik When I try to install using panda they detect the problem and fail in the prereq stage as expected. 23:53
.ask tadzik I hacked around the error so I could get the rest of the smoke test run in. There is something interesting going on here... Several tests are reaching the "tests failing" stage even though their prereqs are failing *their* tests. DateTime::Utils is an example. 23:54
yoleaux colomon: I'll pass your message to tadzik.
colomon .ask member:tadzik When I try to install using panda they detect the problem and fail in the prereq stage as expected.
yoleaux colomon: What kind of a name is "member:tadzik"?!
colomon .ask tadzik When I try to install using panda they detect the problem and fail in the prereq stage as expected.
yoleaux colomon: I'll pass your message to tadzik.
lizmat colomon: fwiw, I think yoleaux only keeps one message per user to a another user 23:56
.ask colomon: one
yoleaux lizmat: What kind of a name is "colomon:"?!
lizmat .ask colomon one 23:57
yoleaux lizmat: I'll pass your message to colomon.
lizmat .ask colomon two
yoleaux lizmat: I'll pass your message to colomon.
colomon o/
yoleaux 23:57Z <lizmat> colomon: one
23:57Z <lizmat> colomon: two
lizmat I guess I was wrong
:-)
colomon lizmat++ # science

Powered by Cro and the Raku® Programming Language.

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

Thank you!