»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 sorear on 25 June 2013.
00:01 virtualsue left 00:04 Ugator left 00:05 BenGoldberg left 00:06 BenGoldberg joined
brrt ehm, who maintains the perl6 advent blog? 00:07
gist.github.com/bdw/3f3928bed9428fa386cb <- blog entry for today
00:15 moritz left 00:16 moritz joined, bcode left, smash left, smash joined 00:17 bcode joined, pyrimidi_ joined 00:18 pyrimidi_ is now known as pyrimidine_
pyrimidine_ Ulti++ # getting an O'Reilly tweet 00:18
vendethiel ooh, amazing, brrt! I said I should contribute some to moarvm, nice start 00:20
brrt (tweets ftw :-)) 00:21
00:21 ka05 joined
brrt well, if it helps recruit people to moarvm, then i've succeeded :-) 00:21
but i don't know how to post it on the perl6advent blog (i don't have credentials for it, for one)
but whoever does has my full permission to post it and make changes 00:22
00:30 pyrimidine_ left 00:31 ab5tract joined 00:39 BenGoldberg left 00:40 BenGoldberg joined 00:44 BenGoldberg left 00:45 BenGoldberg joined 00:47 BenGoldberg left, BenGoldberg joined
colomon brrt: what's your e-mail? 00:54
raiph I've added a comment to brrt++'s gist with typo corrections 00:56
00:57 BenGoldberg left, BenGoldberg joined
brrt thanks raiph :-) 00:58
colomon: sent via pm
raiph: fixed 00:59
colomon brrt: sorry, got distracted by Michigan coaching news. ;) 01:04
brrt: you should have an invitation to join p6 advent waiting in your e-mail. 01:05
brrt np :-) 01:10
01:22 labster left 01:24 Brian joined 01:25 Brian is now known as Guest63206
brrt perl6advent.wordpress.com/?p=2733&a...eview=true 01:27
i've scheduled it for two hours from now 01:28
and i'm going to sleep
see y'all tomorrow
01:28 brrt left 01:43 prime left 01:46 konsolebox joined 01:50 Guest63206 left 01:54 chenryn joined 02:04 doublelel joined 02:06 grettis_ left, BenGoldberg left, BenGoldberg joined 02:11 abraxxa1 joined
abraxxa1 is www.perl6.org moved to the new server currently? it shows the wrong page... 02:12
02:13 abraxxa1 is now known as abraxxa_home 02:14 atroxaper joined 02:17 rmgk_ joined, rmgk is now known as Guest76406, rmgk_ is now known as rmgk 02:19 atroxaper left 02:21 Guest76406 left
TimToady indeed it does 02:25
02:31 BenGoldberg left 02:32 BenGoldberg joined 02:34 _4d47 left, telex left 02:36 telex joined 02:40 vendethiel left, doublelel left 02:43 vendethiel joined, prime joined 02:47 ab5tract left 02:49 BenGoldberg left, BenGoldberg joined 02:51 terrencehan joined 02:52 doublelel joined
timotimo i changed the title to start with "Day 18:" and put it into the 2014 category 02:58
i hope i didn't break it just by opening it in the wordpress editor 02:59
colomon brrt++ 03:00
timotimo damn 03:02
if i had been there, i could have told brrt how to run the graph spesh tool
(all it needs is for you to run update_ops.pl)
03:03 noganex joined
timotimo but it also tells you to in the comment at the beginning of the script :S 03:03
abraxxa_home i'm trying to compile rakudo with clang and found www.mail-archive.com/perl6-interna...45189.html
but it seems the option to pass the compiler and linker has changed
is this a --parrot-option ? 03:04
timotimo i wonder if it'd be disrespectful to write a little in-line comment about that? 03:05
03:06 noganex_ left, BenGoldberg left 03:07 terrencehan left 03:09 jack_rabbit joined
abraxxa_home seems to be --moar-option='--compiler clang' 03:12
timotimo right, we're suggesting people build with MoarVM by default nowadays 03:13
moarvm doesn't accept parrot options :)
abraxxa_home it's --compiler=clang, seems the help screen is slightly wrong there 03:15
03:16 Mso150 joined
timotimo oh, could be 03:16
timotimo doesn't know much about the build system
abraxxa_home i've added a moar-clang option to my rakudobrew 03:17
clang warns a *lot* 03:20
timotimo yes :( 03:21
abraxxa_home but it builds fine 03:23
as the all the perl6.org websites including modules.perl6.org doesn't work I can't install modules using panda
timotimo ooooh 03:33
abraxxa_home did you not know that? 03:37
maybe a vhost misconfiguration of apache
timotimo i didn't know that 03:38
i'm looking at www.p6c.org now (from the inside) and seeing a few difficulties
abraxxa_home all those urls go to Perl 6 Design Documents
timotimo yes
all the sites-enabled/ files aren't added to git, for example
abraxxa_home symlinks? 03:40
are you running Debian?
timotimo this is a debian system, aye
abraxxa_home yeah, i've read the posts about the new server
tried to come up with one at work too
timotimo which mod exactly would apache want to make the sites-enabled subdir work? 03:41
abraxxa_home no mod, the main config file just includes all those files 03:42
timotimo mhm
abraxxa_home most Debian packages are configured that way
even sudo since version 7
timotimo ok 03:43
in that case i don't really see what could be wrong :(
abraxxa_home check the main config file if the sites-enabled dir is really included
and file permissions
dir perms too
JimmyZ can take a look if he has permissions 03:44
abraxxa_home check the startup log for errors 03:45
timotimo the folders are r-x for others, the files are r-- for others
should work
03:46 atroxaper joined
raiph did you guys see irclog.perlgeek.de/perl6/2014-12-17#i_9821281? 03:47
timotimo is VirtualHost *:80 really correct? 03:48
JimmyZ yeah
raiph also, i just emailed Audrey (re perl6.org issues)
abraxxa_home yes
the server name should be configured for each vhost though
the VirtualHost directive just tells it to do routing by hostname 03:49
JimmyZ I don't know how au's action affects perl6.org
:)
raiph JimmyZ: did you see the dns change in my clog link? 03:50
abraxxa_home $ host modules.perl6.org
modules.perl6.org has address 213.95.82.53
is that what it should be?
JimmyZ raiph: yeah
timotimo the ServerName is configured, as are multiple ServerAlias directives
er, configs? options? 03:51
abraxxa_home httpd.apache.org/docs/2.4/en/vhost...based.html
03:51 Mso150 left
abraxxa_home timotimo: i'd reload apache and check the logs 03:52
timotimo i've reloaded it; which log file shoul di check? 03:53
JimmyZ /var/log/apache* ?
03:53 atroxaper left
timotimo yeah, not much to see there 03:54
abraxxa_home can you see my request coming from 62.178.122.70?
should have served modules.perl6.org 03:55
timotimo no accesses for 18th dec for modulse.perl6.org
abraxxa_home does httpd.conf include the sites-enabled dir? 03:56
timotimo your ip shows up in design.perl6.org, though
abraxxa_home where is that vhost configured?
in one of the sites-enabled files or the main httpd.conf? 03:57
timotimo yes, it does
JimmyZ it looks like perlcabal.org has bad configure...
raiph fwiw testers.perl6.org/ works
JimmyZ which make other configures fail
abraxxa_home JimmyZ: it redirects to perl6.org/ which also serves the same, wrong site as modules.perl6.org
timotimo oh 03:58
i think i know what's going on?
JimmyZ abraxxa_home: yeah
abraxxa_home $ host testers.perl6.org
testers.perl6.org has address 85.25.222.109
JimmyZ testers.perl6.org is different server
timotimo the virtualhosts for design and perlcabal have ip addresses
the others just have *
does that make them more specific, and thus prefered in the first step of VHost resolvment?
JimmyZ yeah, that is what it looks like 03:59
timotimo gimme a sec.
abraxxa_home i remember i had that problem when i was using apache (switched to nginx years ago)
maybe something with IPv6 enabled
03:59 noganex_ joined
timotimo that's it 03:59
please try it on your end 04:00
abraxxa_home the apache docs say: Name-based virtual hosts for the best-matching set of <virtualhost>s are processed in the order they appear in the configuration.
JimmyZ so perlcabal.org host's configure is bad
Is it?
abraxxa_home YAY!
timotimo abraxxa_home: yes, but only after IP-based virtual hosts have been processed
It is important to recognize that the first step in name-based virtual host resolution is IP-based resolution. Name-based virtual host resolution only chooses the most appropriate name-based virtual host after narrowing down the candidates to the best IP-based match. Using a wildcard (*) for the IP address in all of the VirtualHost directives makes this IP-based mapping irrelevant.
abraxxa_home got it 04:01
04:01 xinming left
abraxxa_home i'd recommend using * everywhere IF all sites should be accessible by all IPs only by their names 04:01
timotimo yeah
abraxxa_home then i had the same problem because it also occurred when mixing the two
timotimo moritz will have had some reason to have that in there; maybe it was copypasted from the previous config where these sites lived on separate servers 04:02
04:02 noganex left
abraxxa_home so why do i now get Could not find Shell::Command in any of: /home/abraxxa/.rakudobrew/moar-clang-2014.11/install/languages/perl6/lib, /home/abraxxa/.rakudobrew/moar-clang-2014.11/install/languages/perl6 04:02
ah, forgot to run rakudobrew build panda for the clang build, panda still found in path? 04:03
is there a useful benchmark for Perl6 so I can compare the gcc with the clang build? 04:05
04:05 rurban joined
timotimo we have perl6-bench, but there's not really a single benchmark that'd give you a good coverage of "everything" 04:05
abraxxa_home perl 6 test suite? 04:06
both builds fail NativeCall t/08-callbacks.t 04:07
timotimo 32bit?
abraxxa_home 64bit Ubuntu 14.10
where does panda extract the modules to run the tests? 04:08
it's test 3: *** Error in `/home/abraxxa/.rakudobrew/moar-2014.11/install/bin/moar': free(): invalid pointer: 0x00007ff23dfb0a58 *** 04:10
Aborted (core dumped)
timotimo in a .work folder 04:12
abraxxa_home ~/.panda-work
found out using panda look
default os compiler different from Perl 6 compiler maybe? 04:14
but then the regular moar build should succeed at least
clang fails with *** Error in `/home/abraxxa/.rakudobrew/moar-clang-2014.11/install/bin/moar': munmap_chunk(): invalid pointer: 0x00007fbe1e322a78 *** 04:16
i try to get some sleep
for whatever reason I wasn't able to today...
04:19 xinming joined
timotimo gnite! 04:19
abraxxa_home thanks! 04:20
have to do some dbic support ;) 04:21
04:25 perlpilot joined, dj_goku joined 04:31 perlpilot left 04:32 abraxxa_home left 04:35 rurban left 04:39 erkan left 04:42 erkan joined 04:46 erkan left 04:49 raiph left 04:50 jack_rabbit left 04:53 vendethiel left 04:54 vendethiel joined 04:55 tinyblak joined 04:59 chenryn left 05:03 kaare_ joined, jack_rabbit joined, anaeem1_ joined 05:04 anaeem1_ left, anaeem1_ joined 05:06 KCL_ joined 05:15 vendethiel left 05:19 vendethiel joined 05:22 aborazmeh joined 05:31 chenryn joined 05:38 anaeem1__ joined 05:39 anaeem1__ left, anaeem1_ left, anaeem1 joined 05:46 Chillance joined 05:47 Chillance left 05:59 kaare_ left 06:08 [Sno] left 06:18 nige joined 06:21 kaare_ joined 06:30 chenryn left 06:31 torbjorn left, aborazmeh left 06:32 torbjorn joined, daxim left, molaf left, daxim joined 06:33 molaf joined 06:41 konsolebox left 06:55 chenryn joined 07:04 mr-foobar left 07:08 hekmek joined 07:09 bjz joined, atroxaper joined 07:14 atroxaper left 07:32 Woodi joined 07:33 jack_rabbit left 07:43 [Sno] joined 07:46 jack_rabbit joined 07:49 Woodi left, Woodi joined 07:50 FROGGS joined
moritz \o 07:57
nwc10 o/ 07:59
moritz timotimo: the reason behind the odd virtualhost config was that I wanted separate IPv6 addresses for the separate vhosts
timotimo: I know it can be made to work, I just need to be more diligent testing it 08:00
timotimo: thanks for fixing it!
timotimo++
(increased bus number)++
08:02 anaeem1 left, anaeem1 joined 08:04 zakharyas joined
JimmyZ \o/ 08:07
08:12 darutoko joined 08:13 brrt joined
brrt timotitmo++ :-) 08:13
(not my best day in typing names) 08:14
timotimo++
moritz brrt: use tab completion for nick names :-) 08:15
08:17 hekmek left
brrt (does that work in erc :-o) 08:17
08:20 nige left
moritz brrt: I've never used erc, but I'd be surprised if it didn't 08:22
brrt it does :-)
08:23 brrt left, brrt joined 08:24 Alina-malina left 08:25 Alina-malina joined 08:31 rindolf joined 08:34 KCL_ left 08:35 [particle]1 left 08:48 jack_rabbit left
mathw good $localtime o/ 08:52
08:53 ptc_p6 joined 08:56 atroxaper joined 08:57 Isp-sec joined
mathw SpaceX are going to test out landing the first stage of a Falcon 9 on a 300x100ft barge on their next launch 08:57
that is pretty damned awesome
if they can do it
they reckon it's only got a 50% chance of success though
08:57 virtualsue joined 08:59 jack_rabbit joined 09:00 ptc_p6 left 09:01 ptc_p6 joined
mathw woops 09:01
that was all in the wrong channel, I apologise
09:02 atroxaper left
FROGGS mathw: np, many people here are interested in that :o) 09:02
09:04 virtualsue left 09:09 molaf_ joined
lizmat good *, #perl6! 09:10
yoleaux 17 Dec 2014 22:08Z <jnthn> lizmat: I'll take care of the MoarVM release early afternoon; hope that fits well if your relesae planning...
lizmat related PSA: I will start the release process in about 4 hours from now 09:11
09:11 molaf left
lizmat backlogs 09:12
brrt \o lizmat 09:13
lizmat brrt o/
brrt++ # nice blogpost
09:14 jack_rabbit_ joined
moritz should read 09:14
09:15 [particle] joined
dalek kudo/nom: 0385d0f | lizmat++ | docs/ChangeLog:
ChangeLog tweaks
09:15
09:17 jack_rabbit left 09:18 rurban joined
dalek kudo/nom: c6d87c3 | lizmat++ | docs/release_guide.pod:
This month's release will be called 'Cologne'

In appreciation of this week's very successful Niederrhein.PM meeting.
09:18
09:20 bjz_ joined, bjz left
brrt thanks :-) 09:26
09:28 denis_boyun_ joined 09:29 dakkar joined 09:31 fhelmberger joined
lizmat afk for a few hours& 09:34
09:34 brrt left, brrt joined 09:39 mvuets joined 09:40 virtualsue joined
tadzik mathw: this is the most right channel for that that I know of 09:43
#perl6 is all about rocket launches :)
09:48 virtualsue left 09:51 pecastro joined
mathw Rocket.new(size => Size::ReallyReallyBig); 09:56
10:06 tinyblak_ joined 10:08 denis_boyun_ left 10:10 tinyblak left, spider-mario joined 10:11 spider-mario left 10:12 spider-mario joined 10:19 jack_rabbit_ left 10:24 xinming left 10:25 xinming joined 10:30 atroxaper joined 10:31 tinyblak_ left, tinyblak joined 10:33 telex left 10:34 tinyblak_ joined, telex joined 10:36 tinyblak_ left, khisanth_ joined, atroxaper left, tinyblak_ joined 10:37 tinybla__ joined, tinyblak left 10:38 Khisanth left 10:40 tinyblak_ left 10:43 jack_rabbit_ joined 10:48 virtualsue joined
masak good noon, #perl6 10:50
brrt++ # perl6advent.wordpress.com/2014/12/1...d-curious/ 10:51
10:51 ssqq joined
masak who wants to do tomorrow? we have a free slot. 10:51
masak pictures someone reading this and going "oh, it's my turn! time to volunteer!"
brrt thanks masak :-) 10:52
ssqq nqp-m: ok((?nqp::getlex("&say") == 1), "&say is global routine");
camelia nqp-moarvm: OUTPUT«ok 1 - &say is global routine␤»
brrt although 215 mov instructions in such a simple routine... that shocked me, really 10:53
ssqq nqp-m: my $sub-name := '&say'; ok((?nqp::getlex($sub-name) == 1), "&say is global routine");
camelia nqp-moarvm: OUTPUT«expected QAST constant; didn't get one␤ at gen/moar/stage2/QAST.nqp:5402 (/home/camelia/rakudo-inst-2/languages/nqp/lib/QAST.moarvm:as_mast_constant:6)␤ from gen/moar/stage2/QAST.nqp:5390 (/home/camelia/rakudo-inst-2/languages/nqp/lib/QAST.moarvm:as_mast_c…»
10:58 chenryn left 11:03 jack_rabbit_ left
hahainternet o/ masak 11:06
11:12 ab5tract joined
masak ahoj. 11:18
more specifically, we have four slots to fill before the Advent calendar is complete. 11:19
11:19 Isp-sec left
masak I could maybe do one more, but I'd like to see two more slots filled in first... :) 11:19
as usual, it would be kind of nice to have something a little extra for the 24 slot. 11:20
11:28 Ugator joined, jack_rabbit_ joined 11:31 ab5tract left 11:34 atroxaper joined, bjz_ left 11:35 bjz joined 11:38 atroxaper left 11:39 rurban left 11:44 denis_boyun joined
dalek kudo/nom: 1c88e82 | lizmat++ | docs/ChangeLog:
Mention fixing of OS X flakiness
11:56
moritz masak: I'm considering doing a sleepsort for day 23
masak: 'cause we've had sorting stuff on two 23rd days before 11:57
11:57 andreos joined
lizmat lunch& 11:58
12:00 perturbation joined
andreos how to apply a function to an array as it were a list of arguments? 12:03
masak moritz: go for it! :) 12:04
andreos: f(|@array)
andreos m: sub first($car, *@cdr){$car}; say first( | "asdfgh".comb)
camelia rakudo-moar c6d87c: OUTPUT«a␤»
andreos thanks
12:08 ab5tract joined 12:22 jack_rabbit_ is now known as jack_rabbit 12:30 ab5tract left 12:36 ssqq left, ssqq joined 12:48 pdcawley joined 12:50 rurban joined 12:57 atroxaper joined
jnthn Afternoon, #perl6 13:01
hahainternet hey jnthn
moritz \o
hahainternet bringing up a topic from a few days ago, did any progress get made regarding unions/fixed allocs in nativecall? i still haven't had chance to go through the nqp pdf and learn the infrastructure well enough to be of any use 13:02
13:03 atroxaper left 13:04 atroxaper joined
nwc10 good UGT heresy, jnthn 13:05
dalek : 1ce52da | lizmat++ | misc/perl6advent-2014/schedule:
Add title of brrt's post
13:08
kudo/nom: 2698598 | lizmat++ | src/core/Mu.pm:
Remove support for bless *, deprecated 1 year ago
13:09
kudo/nom: a71ed88 | jnthn++ | docs/ChangeLog:
Some missing ChangeLog entries.
13:13
lizmat jnthn: the {} zen slice listifies, according to the code ? 13:14
jnthn I thought we said .hash? 13:16
*sigh*
lizmat well, when I tried that, I got quite a few spectest failures
jnthn arrgh
lizmat let me double check that again in a mo
jnthn Well, .list is certainly wrong
That's worse than it wsa before.
So we should back that out if we can't .hash. 13:17
lizmat ok, will do
13:18 Rounin joined
jnthn lismat++ 13:18
uh, lizmat
This train internet is so crappy today Im typing blind half the time... :/ 13:19
dalek kudo/nom: e84b99a | lizmat++ | docs/release_guide.pod:
Add a step about removing DEPRECATED code
jnthn Will work on MoarVM release. 13:20
lizmat ok, cool :-)
13:21 anaeem1 left 13:25 aborazmeh joined, aborazmeh left, aborazmeh joined 13:29 perturbation left
m_athias jnthn: mosh.mit.edu/ makes that a waaaay better 13:30
13:30 andreos left
jnthn m_athias: ooh, nice... I should give it a try :) 13:32
colomon Net::DNS and Net::XMPP failed tests overnight. 13:33
13:34 ssqq left
lizmat jnthn: with {} returning SELF.hash, I get errors in t/spec/S32-str/numeric.t 13:36
m: my $str = ":10<8_>"; +$str
camelia rakudo-moar 1c88e8: OUTPUT«Unhandled exception: Cannot convert string to number: malformed ':10' style radix number, expecting '>' after the body in ':10<8⏏_>' (indicated by ⏏)␤ at <unknown>:1 (/home/camelia/rakudo-inst-1/languages/perl6/runtime/CORE.setting.moarvm:throw:42…»
lizmat becomes:
Odd number of elements found where hash initializer expected
jnthn That looks like a failure in constructing the error message... 13:37
Oh, or maybe we're losing a container?
lizmat not sure yet
it's a rather large test file :-)
jnthn What happens if we do: 13:38
nwc10 jnthn: maybe if the train stood still, the internet would get better? :-) 13:39
jnthn multi sub postcircumfix:<{ }>( \SELF ) is rw { SELF.hash }
multi sub postcircumfix:<{ }>(Associative \SELF ) is rw { SELF }
lizmat trying 13:40
13:40 Alina-malina left 13:41 Alina-malina joined
abraxxa tadzik: are you interessted in a patch for rakudobrew that lets you set the compiler for moar? 13:43
tadzik abraxxa: I'd rather have something generic that gets passed along to Configure.pl 13:44
so one can make that --moar-option=different-compiler or whatever
lizmat jnthn: same errors in that test
running full spectest now
abraxxa tadzik: i did it by cloning moar to moar-clang in line 24
i wanted a generic, additional option first but then realized its moar specific 13:45
tadzik well, rakudo's Configure.pl manages to have backend specific options
we could get away with something like --configure-opts 13:46
abraxxa when multiple impls are brewed you should be able to pass different things to them..
tadzik that too
but rakudo's Configure.pl is capable of that, afaik 13:47
there's --moar-option, --parrot-option...
abraxxa yeah, so rakudobrew should be able to pass different things to them 13:48
tadzik I don't think rakudo breaks if you don --backends=moar --parrot-option=foo
abraxxa ah, i see
tadzik s/don/do/
lizmat jnthn: more breakage that way
tadzik if it does, I'm willing to go and change that :P
but I don't see why it would
lizmat looks like I need to specify candidates for all named parameters as well: Unexpected named parameter 'exists' passed 13:49
tadzik so: I don't think that rakudobrew should be concerned as to how to pass different options to different backends
just have rakudo do it
abraxxa ok
clone + github pull request?
tadzik yes please :) 13:50
abraxxa s/clone/fork/
tadzik I might be slow to respond, feel free to bother me here
lizmat m: my %h; %h{}:exists # jnthn: fails now
camelia ( no output )
[Coke] double checks, and the release in jan occurs the week before work travel. whew.
jnthn :exists on a zen slice?! 13:51
But yeah...dammit, the Associative beats the :exists.
Just back out the .list for now, and the ChangeLog entry then. We'll have to take it in the next release.
lizmat ok 13:53
dalek kudo/nom: 1b0e064 | lizmat++ | src/core/hash_slice.pm:
Revert "Make foo[] and foo{} consistent"

We will need to revisit that for a future release.
13:54
abraxxa tadzik: can we please use a module for cli arg handling? 14:02
tadzik abraxxa: I'd much rather not :| 14:03
oh, unless it's core
abraxxa is there one? i just looked in 5.20.1 but can't find one
masak doesn't Perl 6 already have quite a lot of arg handling in the language?
abraxxa i thought one very old is core, don't remember its name
masak: i uses perl 5 to build perl 6
masak I see. 14:04
tadzik abraxxa: "Getopt::Long was first released with perl 5"
abraxxa so why isn't it listed here? metacpan.org/release/perl 14:05
the corelist is a lie!
Ulti m: say !???
camelia rakudo-moar 1c88e8: OUTPUT«Stub code executed in block <unit> at /tmp/56EnEzsE7K:1␤␤True␤»
Ulti m: my $var = !!???????; say $var.uc; 14:07
camelia rakudo-moar 1c88e8: OUTPUT«===SORRY!=== Error while compiling /tmp/tK8Fs1anto␤Two terms in a row␤at /tmp/tK8Fs1anto:1␤------> my $var = !!???????⏏; say $var.uc;␤ expecting any of:␤ argument list␤ postfix␤ i…»
Ulti m: my $var = !!???; say $var.uc;
camelia rakudo-moar 1c88e8: OUTPUT«Stub code executed in block <unit> at /tmp/0neLrfGtlV:1␤␤FALSE␤»
Ulti heh there is some weird stuff around there
I'm guessing the last ?? are a ternary
which is why it doesnt like the stub 14:08
lizmat m: !!!! # similar issue
camelia rakudo-moar 1c88e8: OUTPUT«===SORRY!=== Error while compiling /tmp/z9XzkdGBnH␤Negation metaoperator not followed by valid infix␤at /tmp/z9XzkdGBnH:1␤------> !!!!⏏ # similar issue␤ expecting any of:␤ argument list␤ …»
lizmat m: say !!!! # similar issue
camelia rakudo-moar 1c88e8: OUTPUT«===SORRY!=== Error while compiling /tmp/foUqmZtV8x␤Negation metaoperator not followed by valid infix␤at /tmp/foUqmZtV8x:1␤------> say !!!!⏏ # similar issue␤ expecting any of:␤ argument list␤ …»
14:09 sven_123_ joined 14:12 sven_123_ left
lizmat jnthn: re: 19f661f3b4092548cd64284c52f2609af3057b43 14:16
wouldn't it make sense to add a method to Mu with that list of names, and call that
and thus make it overridable for a class ?
TimToady: I guess 5a06ace5c5c458a78313e32d7235e428486bbe23 is really an API change, so should be emphasized in the release message ? 14:18
jnthn: re, 7432d6202ede8569cac13ecbfae78afafc915fa4 , wouldn't that be worthy of a advent post ? 14:19
abraxxa tadzik: can the dirname of the installed rakuda be set with rakudobrew? 14:23
tadzik abraxxa: nope
maybe it should be :)
abraxxa would it be easier to extend perlbrew? 14:24
tadzik I don't think so
abraxxa i'm not very happy with my code but it works
tadzik I once looked into making a plenv plugin, but it looked like more trouble than it was worth
abraxxa the error messages are a bit misleading, talk about gcc although i requested clang
jnthn lizmat: (method on Mu), no, that feels like mixing levels to me...
abraxxa probing whether your compiler thinks that it is gcc Can't compile simple gcc probe, so something is badly wrong at build/probe.pm line 92.
jnthn lizmat: Dunno about a post on that...I mean, it's quite esoteric. :) 14:27
lizmat well, it could serve as documentation, as the documentation atm is quite minimal :-)
moritz documentation belongs on doc.perl6.org
14:27 raiph joined
masak hard to argue with that. 14:27
lizmat moritz: I meant, until we have documentation :-) 14:28
and it would be a bit more hard core than most of the other advent posts :-)
but maybe too hard core :-)
jnthn Yeah, it's also a question of when I'd have time to write it.
abraxxa tadzik: you may look now github.com/abraxxa/rakudobrew/comm...ba27c73d11 14:29
the previous commit is a whitespace cleanup that will be in the pull request as well
jnthn Today was my last day of teaching. Tomorrow I need to take care of some christmas shopping and prep for travel to UK. Saturday I'm traveling there. And I'm so exhausted from the last weeks the first thing I'll probably do is get sick. :/
abraxxa tadzik: i have a meeting now, will read the scrollback afterwards
masak jnthn: I suggest you skip over getting sick this year. 14:30
lizmat jnthn: ok, gotcha
14:30 FROGGS left
lizmat masak: that's not how that works, unfortunately 14:30
:-(
masak hey, it was just a suggestion :P
jnthn masak: I never plan to, it's just a depressingly common pattern. The tireder I am when I go for Christmas break, the more likely I am to quickly catch something or other.
abraxxa it's a usual pattern that the body backs down when the stress goes away 14:31
jnthn Right.
moritz abraxxa: aye; my father used to have problems with his teeth each summer vacation for several years in a row 14:32
14:32 spider-mario left, anaeem1 joined
masak it's not so strange, I think. stress keeps you going. 14:33
evo devo explanations are fraught, I guess. but maybe stress helped people survive. and when they were safe, they slowed down and got sick.
people who managed to suspend getting sick were more fit and spread through the gene pool. 14:34
moritz it makes a lot of sense if you think of being sick as "maintenance downtime" 14:35
lizmat "health" debt
moritz but classic medicine teaches us it's all those bad viruses and bacteria that make us sick, and that's kinda an incompatible view
lizmat: aye
masak it's probably a mix of that, and the fact that parasites always win.
that is, the body can hold them off, but it's a losing battle. 14:36
lizmat also: meeting people you haven't seen in a while, means new germs :-)
masak also, winter. means recycling stale indoor air.
jnthn Yeah, 'cus people can't hack the cold, dammit. :P
lizmat personally, I notice this when meeting people with young kids (fresh kindergarten germs!) who I haven't seen for a while :-) 14:37
masak kids are excellent disease vectors.
brrt thing about stress is, we know from various sources that it is 'bad', but it is one of these long-term-biology things that we can't quite get our head arround 14:39
otherwise, winter being sickness season is mostly people recycling stale air yeah :-) 14:40
14:40 aborazmeh left
Ulti can someone explain what's going on here 14:41
m: say !!1 14:42
camelia rakudo-moar 1b0e06: OUTPUT«True␤»
Ulti m: say !!!1
camelia rakudo-moar 1b0e06: OUTPUT«===SORRY!=== Error while compiling /tmp/4FKXCtEKkW␤Two terms in a row␤at /tmp/4FKXCtEKkW:1␤------> say !!!⏏1␤ expecting any of:␤ argument list␤ postfix␤ infix stopper␤ in…»
masak that one's correct.
Ulti why can you do two lots of negation but not more?
masak because `!!!` is a term.
jnthn !!! parsers as a term
masak m: say !!(!1)
camelia rakudo-moar 1b0e06: OUTPUT«False␤»
masak m: say !! !1
camelia rakudo-moar 1b0e06: OUTPUT«False␤»
masak m: say ! ! ! 1
camelia rakudo-moar 1b0e06: OUTPUT«False␤»
Ulti wait so what is !!! ?
masak m: !!! "argh!" 14:43
camelia rakudo-moar 1b0e06: OUTPUT«argh!␤ in block <unit> at /tmp/B9YH6N1R8a:1␤␤»
masak a synonym of "die"
Ulti o___O
jnthn m: omg!!!
camelia rakudo-moar 1b0e06: OUTPUT«===SORRY!=== Error while compiling /tmp/O1mLZCcGGv␤Two terms in a row␤at /tmp/O1mLZCcGGv:1␤------> omg⏏!!!␤ expecting any of:␤ argument list␤ postfix␤ infix stopper␤ infi…»
jnthn Aww :)
m: omg !!!
camelia rakudo-moar 1b0e06: OUTPUT«===SORRY!=== Error while compiling /tmp/8XuV8hG_7y␤Undeclared routine:␤ omg used at line 1␤␤»
Ulti and ???
?
jnthn hah :0
:)
Ulti: !!! means "die", ??? means "warn", ... means "fail"
Ulti aha
jnthn Different levels of stubbiness
JimmyZ m: ... 14:44
camelia rakudo-moar 1b0e06: OUTPUT«Unhandled exception: Stub code executed␤ at <unknown>:1 (/home/camelia/rakudo-inst-2/languages/perl6/runtime/CORE.setting.moarvm:throw:4294967295)␤ from src/gen/m-CORE.setting:13847 (/home/camelia/rakudo-inst-2/languages/perl6/runtime/CORE.setting.…»
JimmyZ m: !!!
camelia rakudo-moar 1b0e06: OUTPUT«Stub code executed␤ in block <unit> at /tmp/kBkNhKngqx:1␤␤»
masak <!!! ??? ...> isn't particularly strangely consistent with anything else.
Ulti so who ordered !!! and ??? 14:45
masak except perhaps very loosely in that I agree with the respective urgencies of those three punctuation marks.
14:45 FROGGS joined
Ulti given ?? !! is a thing too that is some serious semantic overload 14:45
masak Ulti: welcome to Perl 6.
Ulti ;___;
masak we have Serious Semantic Overload Syndrome.
:D
lizmat well, fwiw, I would think that most people would use ...
Ulti it's especially confusing when the warn is evaluated as false 14:46
lizmat and that you actually wouldn't really need !!! and ???
masak I'd like to recommend skillsmatter.com/skillscasts/6088-...guage-ever to people who enjoy deadpan reverse psychology crossed with language design.
14:46 ptc_p6 left
Ulti masak: thats how I got to this point :D 14:46
I was just replying to someone on the reddit comments for that talk who called out Perl 6 14:47
but I think I mostly agree with them :P
masak lizmat: I only use `...`. and only in roles that I want to act a bit like interfaces.
JimmyZ m: ??!!!
camelia rakudo-moar 1b0e06: OUTPUT«Stub code executed␤ in block <unit> at /tmp/AoU5KvFgk6:1␤␤»
JimmyZ std: ??!!!
camelia std 76ccee1: OUTPUT«===SORRY!===␤Expecting a term, but found either infix ?? or redundant prefix ?␤ (to suppress this message, please use space between ? ?) at /tmp/pjvwTWB7xC line 1:␤------> ??⏏!!!␤Parse failed␤FAILED 00:00 135m␤»
Ulti m: !!???
camelia rakudo-moar 1b0e06: OUTPUT«Stub code executed in block <unit> at /tmp/Dt5fAinAj1:1␤␤»
JimmyZ look, LTA
14:47 oetiker left
masak I think the best part of the above video was that the presenter got cut off at the end, shouting: "PLEASE DON'T WRITE A CO" :P 14:48
Ulti m: say "wow" unless !!???
camelia rakudo-moar 1b0e06: OUTPUT«Stub code executed in block <unit> at /tmp/iKxRpctn2b:1␤␤wow␤»
14:48 oetiker joined
dalek kudo/nom: 3d390fb | lizmat++ | src/core/Distro.pm:
Make sure we always have a $*DISTRO.release
14:48
Ulti masak lol, the language looked almost reasonable just like all awful languages :) its reasonable enough to actually use for real but make devs cry 14:49
the problem with every programming language is that regardless of how high level you are, you still have to talk on the computers terms about what you want 14:50
masak Ulti: you should have a look at Haskell. :) 14:52
jnthn masak: So he can talk on mathematicians terms instead? :P 14:53
masak well, it's a counterexample, is my point. 14:54
dalek kudo/nom: 57faf7a | lizmat++ | docs/ChangeLog:
Remove mention of {} change (there is none yet)
15:01 doublelel left 15:02 kaare_ left 15:18 doublelel joined 15:20 PZt left 15:22 FROGGS left 15:26 virtualsue left
dalek kudo/nom: f8267f2 | lizmat++ | src/core/Str.pm:
Streamline :nth(*-1)
15:26
15:26 virtualsue joined 15:29 kaare_ joined
Ulti masak all that happens with really different languages is they attempt to simulate a different computer and then get you to write for that instead :P 15:32
PROLOG being a good example
masak yes. probably the "imperative" execution model is the only really intuitive one to humans. do this, then do that. 15:33
but declarative sublanguages will always have a place in specific domains. that includes Prolog. 15:34
includes things like Forth and Haskell as well.
brrt that is so counterintuitive, and i'm not sure if it doesn't include some 'survivor bias'
as in, everybody who could grok imperative became a programmer, and those who could not did not 15:35
vendethiel backlogs again
masak brrt: I'm not sure that weakens the argument in any way.
brrt: "X is just intuitive because the survivors consider it intuitive" -- well, yeah.
same reason sunsets are beautiful, I guess. 15:36
brrt well, the counterargument is that because computer languages are imperative people who would've liked another paradigm better (declarative) have already left
i'm not saying that it's true. i like my imperative programs 15:37
masak there is still hope, though. my argument only goes for what becomes mainstream and the main slang.
as I pointed out, there is still room for more exotic runtime models at the fringes.
and that's something to cherish and exploit. 15:38
brrt and embrace and assimilate if you're perl
:-)
geekosaur there is some anecdata for beginners finding declarative languages easier to comprehend than imperative; I'm not aware of any serious exploration of that, though
masak geekosaur: would be interesting to find one. 15:39
I mean, an evo devo explanation for *that* would be that imperative looks like stories by the campfire.
Ulti I dont get why prolog isnt more popular in robotics, it makes writing control systems quite natural human(X) :- dontkill(X)
masak "and then they did a GO TO and handled the exceptional condition!" 15:40
geekosaur Ulti: prolog doesn't compile down to embedded processors well
15:40 raiph left
Ulti it could if the embedded processor could execute prolog :P 15:40
geekosaur there's also the usual problem with real time control vs. garbage collection
rurban more the exponential nature of backtracking 15:41
geekosaur that too, yes
rurban new-style SMT solvers are usually better than prolog
geekosaur "real time" is the real constraint, in any case 15:42
rurban like z3
15:42 FROGGS joined
geekosaur with memory usage still something of a close second 15:42
brrt wonders what masak expands evo devo to 15:43
rurban I'm thinking to add a solver into p2 core (minisat), because I'd need it for compiler optimizations anyway
but simple prolog|ml like declarative matchers and unifiers don't need that much, and are expressible with simple = or =~ and varioables on the rhs 15:44
dalek p: 4394713 | lizmat++ | tools/build/MOAR_REVISION:
Bump MOAR_REVISION to release 2014.12
rurban and _ on the lhs 15:45
15:46 treehug88 joined
brrt (mosh is udp based. that's clever) 15:48
abraxxa tadzik: if the code is ok i'll send the pull request 15:50
15:50 raiph joined
hoelzro morning #perl6 15:54
lizmat hoelzro o/ 15:55
brrt \o hoelzro 15:56
hoelzro o/ lizmat, brrt 15:57
m: class Foo {} ; my $f = Foo.new; say $f[0] 16:00
camelia rakudo-moar 57faf7: OUTPUT«Foo.new()␤»
hoelzro when did that start behaving that way?
PerlJam which part? 16:01
m: class Foo {} ; my $f = Foo.new; say $f;
camelia rakudo-moar 57faf7: OUTPUT«Foo.new()␤»
PerlJam m: class Foo {} ; my $f = Foo.new; say $f[0][0][0]; 16:02
camelia rakudo-moar 57faf7: OUTPUT«Foo.new()␤»
TuxCM YOU DID VERY WELL THIS MONTH: (tiny flood)
i686 i686 x68_64 x68_64 x68_64
20141007 20141014 20141105 20141111 20141218
------------------ -------- --------- --------- -------- --------
Text::CSV::Easy_XS 0.022 0.021 0.037 0.018 0.020
Text::CSV::Easy_PP 0.018 0.022 0.028 0.016 0.033
Text::CSV_XS 0.038 0.039 0.046 0.039 0.042
Text::CSV_PP 0.651 0.535 0.533 0.516 0.534
Pegex::CSV 1.947 1.420 1.387 1.371 1.401
csv.pl 12.543 8.437 9.152 9.130 8.195
test.pl 256.985 201.621 71.847 72.221 58.765
test-t.pl - - - 72.024 58.673
the last three are perl6
note the drop from 257 to 58 seconds! amazing
masak :) 16:04
16:04 abraxxa left
hoelzro PerlJam: why does postcircumfix:<[ ]> work on a non-Positional? 16:04
masak brrt: en.wikipedia.org/wiki/Evolutionary...al_biology
16:04 Sqirrel left
dalek kudo/nom: 4dfc29f | lizmat++ | docs/ChangeLog:
Some more ChangeLog updates
16:06
16:07 FROGGS left
PerlJam hoelzro: we've treated an item as a "list of 1" for quite a while now. As to why, I'm not sure I can explain it right off. 16:07
hoelzro see, my @list = $item makes sense to me 16:08
$item[0] does not
and I feel that the former has been valid for as long as I remember, but the latter is more recent
lizmat hoelzro: not only [0] works always, you can call .end on anytthing
m: say 42.end 16:09
camelia rakudo-moar 57faf7: OUTPUT«0␤»
lizmat so you can always do something like:
m: my $a = 42; for 0 .. $a.end { say $a[$_] }
camelia rakudo-moar 57faf7: OUTPUT«42␤»
hoelzro that seems...odd to me 16:10
dalek kudo/nom: d01c70d | lizmat++ | docs/ChangeLog:
Mention we have more signals on Moar
16:11
lizmat hoelzro: if $a is returned from a sub, maybe it becomes clearer?
16:12 gtodd joined
brrt ah thanks :-) 16:14
i thought something like that
hoelzro lizmat: well, if I'm returning from a sub, and I'm assigning to a positional container, I expect the item to be assigned to $container[0] 16:16
I just don't know of a circumstance where $a[0] is useful to have if $a isn't a Positional 16:17
16:18 mr-foobar joined
lizmat m: my @a=^10; say @a[0].WHAT; say @a[0,1].WHAT 16:18
camelia rakudo-moar 57faf7: OUTPUT«(Int)␤(Parcel)␤»
lizmat m: my @i = 0; my @a=^10; say @a[@i].WHAT 16:19
camelia rakudo-moar 57faf7: OUTPUT«(Parcel)␤»
16:21 brrt left
dalek kudo/nom: bfbef0b | lizmat++ | docs/announce/2014.12.md:
Add 2014.12 announcement
16:22
lizmat ^^^ comments welcome!
dalek p: b7c3fd9 | lizmat++ | VERSION:
bump VERSION to 2014.12
16:24
16:24 hekmek joined 16:25 denis_boyun left
hoelzro lizmat++ # release 16:27
16:29 jluis left 16:31 zakharyas left 16:32 atroxaper left 16:37 virtualsue left
PerlJam lizmat: should the announcement have a link to parrot.org too? I don't know if we've ever done that before, but it struck me as odd that we'd have a link for moarvm and not one for parrot 16:40
lizmat good point, will add 16:41
16:50 virtualsue joined
lizmat ok, I'm not getting NQP built from a tar-file: 16:52
In file included from 3rdparty/libuv/include/uv-unix.h:55: 16:53
3rdparty/libuv/include/uv-darwin.h:26:11: fatal error: 'mach/mach.h' file not
found
# include <mach/mach.h>
oddly enough, there is no 3rdparty/libuv directory
this seems to be a feature of Xcode 6.1.1 16:54
rurban on darwin? 16:55
lizmat yup
Woodi hallo today :)
rurban libuv's configure normally catches those include path for me, macport gcc and xcode cc 16:56
lizmat also for Xcode 6.1.1 ?
geekosaur wonders if precompiled headers are in use somewhere 16:57
japhb lizmat: Has 2014.12 already been cut, or still in progress?
lizmat the VERSION has been updated, but not yet tagged 16:58
japhb OK, thanks
rurban cannot try now, my wife has my laptop
lizmat japhb: any suggestions ?
japhb lizmat: suggestions for what? The Xcode problem? (Because I know almost zero about Xcode, except that my little bit of interaction with it was not a positive experience). 17:00
lizmat ah, ok
found a solution: adding a symlink in Moarvm/3rdparty/libuv/include to /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.9.sdk/usr/include/mach
Woodi some days ago I was in our family bakery's office and was thinking about that imperative vs declarative languages - do being declarative means that on some low level there need to be programmed imperative actions have something done ? and then: knoc*knock There is no sugar... -- was this a "declarative" approach to a problem ? 17:01
masak Woodi: "declarative" to me means that you write a specification, not instructions. it's kind of defined by what it isn't. 17:04
Woodi so html is declarative... but still something need to render it and on some level imperative things are executing ? 17:07
masak I'm not sure HTML qualifies in the first place. not if we're talking programming languages. HTML is markup. 17:08
a good example might be the class/has/method/submethod keywords in Perl 6. they're declarative: they're a specification of a class, not instructions on how to put the class together. 17:09
17:09 virtualsue left
masak SQL might qualify as declarative as well. 17:09
17:10 doublelel left
masak still no takers for tomorrow's advent post? 17:10
many of the APIs for promises are declarative. they free you of the minutia of handling things in sequential order, and can turn "pyramids of doom" into nice linear structures. 17:12
dalek p: 8de8f7b | lizmat++ | README.pod:
Add my fix to building on OS X

In the hope it will be helpful. I don't see another fix yet. And I'm apparently the only person suffering from this (so far)
masak that's a curious case, because these declarative APIs are found in the middle of an imperative language. 17:13
lizmat masak: I'm too busy with the release
masak lizmat: I understand completely. I was thinking somebody who isn't releasing Rakudo today ;)
17:15 dj_goku left
lizmat I have a NQP release tarball ready to upload 17:16
can someone do that ?
dalek kudo/nom: 412e5f5 | lizmat++ | tools/build/NQP_REVISION:
[release] bump NQP revision
17:20
17:20 mvuets left
dalek kudo/nom: 76eb31b | lizmat++ | VERSION:
[release] bump VERSION
17:20
masak PerlJam, moritz: ping 17:23
Woodi lizmat: is it R* maybe ?
masak ^ can either of you upload a tarball for lizmat?
Woodi: no, just R.
lizmat but since we have fixed the OS X flakiness, maybe we should do a R* again :-) 17:24
but after this
:-)
Woodi just arrive home and now wondering: Is pressing Play on cd-player is imperative ?
17:25 ab5tract joined
lizmat
.oO( it is if the player is functional ? )
17:25
Woodi lizmat: just after last R* was nice performance in R, if my memory is correct... would be nice to have one 17:26
it do play songs, if you ask about that :) 17:27
bot how would look like declarative cd-player ? here are buttons and CD. EOD. ? :) 17:28
17:29 njmurphy left
Woodi ow, another one... few days ago I found that Freenet is part of Dark Net! scary... 17:30
Freenode I mean
17:32 khisanth_ is now known as Khisanth, sorear left 17:33 atroxaper joined, sorear joined
dalek kudo/nom: 1f301e6 | lizmat++ | docs/announce/2014.12.md:
Add reference to parrot.org, PerlJam++
17:34
kudo/nom: e487d72 | lizmat++ | docs/announce/2014.12.md:
Add troubleshooting section
17:38
masak I seem to be able to do an advent blog post tomorrow.
but I would prefer it if someone else signed up. then my promise kicks in and I can do day-after-tomorrow. 17:39
lizmat dinner& 17:40
masak either way, I can work on a draft tonight. 17:41
17:41 atroxaper left 17:45 fhelmberger left
moritz masak: ping 17:45
masak: erm, pong
masak lizmat would like to upload an nqp. 17:46
17:47 telex left
moritz then she should do that :-) 17:47
there's an ssh pubkey [email@hidden.address] in the upload user 17:48
masak ooh
17:48 telex joined
masak :) 17:48
moritz, strict but fair BOFH :P
moritz masak: as is an cmasak@gm... key 17:49
I mean, I can upload it too, when I know where the tarball is
or add more keys
masak seems lizmat is away for dinner right now. I guess she had missed (as I did) that there are pubkeys in the upload user ;) 17:52
dalek p: e089718 | moritz++ | docs/release_guide.pod:
release guide: update the list of known uploaders
17:53
p: e3dcf53 | moritz++ | docs/release_guide.pod:
Oops, no need to list FROGGS twice
17:54
kudo/nom: 0a12e17 | moritz++ | docs/release_guide.pod:
release guide: update list of known uploaders
17:55
17:57 tinybla__ left 17:58 tinyblak joined, anaeem1 left 17:59 ptc_p6 joined 18:00 anaeem1_ joined, dakkar left, tinyblak left 18:01 tinyblak joined 18:04 rurban left 18:06 anaeem1_ left 18:07 anaeem1_ joined, anaeem1_ left, ab5tract left 18:09 grettis joined 18:19 virtualsue joined 18:24 tinyblak left, tinyblak joined 18:26 dylanwh joined 18:27 lucas_ joined
lucas_ release day! \o/ 18:27
I like updating my tarballs. Thanks in advance!
18:29 lucas_ left 18:40 atroxaper joined 18:45 mvuets joined, atroxaper left 18:50 denis_boyun joined 18:58 denis_boyun left 18:59 DarthGandalf left, treehug88 left
PerlJam masak: pong. If lizmat has issues uploading, I'll be aroundish and able to upload (FYI) 18:59
masak good, good 19:00
moritz informs me it should be no problem. we'll see.
19:02 DarthGandalf joined, zakharyas joined, treehug88 joined 19:06 raiph left
lizmat thanks PerlJam++ for uploading NQP 19:07
masak he hasn't yet.
lizmat I think he has ?
masak ok, but not as far as I know :) 19:08
moritz (in the backlog) says you already have the ackles for it, though.
lizmat yes, server side
client side, I have some keychain issues :-(
don't want to try fixing that tonight
masak lizmat: irclog.perlgeek.de/perl6/2014-12-18#i_9825720
19:09 raiph joined
PerlJam masak: rakudo.org/downloads/nqp/nqp-2014.12.tar.gz # now you know :) 19:09
lizmat masak: yeah, I know... but when trying to do the scp, it fails
masak oki
PerlJam++ 19:10
m: sub f1 { say "first"; False }; sub f2 { say "sekkind"; True }; say f1() && f2()
camelia rakudo-moar 0a12e1: OUTPUT«first␤False␤»
masak m: sub f1 { say "first"; False }; sub f2 { say "sekkind"; True }; f1() && f2()
camelia rakudo-moar 0a12e1: OUTPUT«first␤»
masak m: sub f1 { say "first"; False }; sub f2 { say "sekkind"; True }; f1() ?& f2()
camelia rakudo-moar 0a12e1: OUTPUT«first␤sekkind␤»
masak thought so
m: sub f1 { say "first"; False }; sub f2 { say "sekkind"; True }; f1() & f2() 19:11
camelia rakudo-moar 0a12e1: OUTPUT«first␤sekkind␤»
lizmat is now testing the rakudo tar ball 19:14
masak S03:2530 says this about `and`: "In list context forces a false return to mean C<()>." 19:16
synopsebot Link: perlcabal.org/syn/S03.html#line_2530
masak what does that mean in Perl 6, "in list context"?
is it a fossil?
m: sub f1 { say "first"; fail }; sub f2 { say "sekkind" }; f1() andthen f2() 19:17
camelia rakudo-moar 0a12e1: OUTPUT«first␤»
lizmat masak: feels like a fossil to me 19:19
masak removes
lizmat at least the wording
masak eech, the phrase 'list context' occurs 26 times in S03. 19:20
[Coke] are there any kinds of contexts left in perl6 that aren't just notional? 19:23
lizmat 88 times in total in synopses
masak [Coke]: good question.
not in the literal Perl 5 sense, I think. 19:24
19:24 virtualsue left
masak wow, S03 postulates a statement_control:<ACCEPTS> macro -- I don't recall seeing that before. 19:25
[Coke] I hear "sink context" a lot... and that might actually be right, since I think we are checking to see if the return value is used or not. (or, is it different from void context? Iunno)
PerlJam 90 if you count the two occurences where it's "List context" rather than "list context"
masak ...but it's textual :/
PerlJam Just for grins: gist.github.com/perlpilot/7ed197a4167cc1e4b40f
lizmat
.oO( that gives quite some context )
19:26
masak .oO( Perl 6 is often in lol context, but sometimes the jokes fall in flat context ) 19:27
lizmat
.oO( I haz a sinking feeling )
masak amazing we don't have a "troll context" yet.
PerlJam I'm slightly amazed that we have 7 occurences of "flat context"
masak m: sub f1 { say "first"; False }; sub f2 { say "sekkind"; False }; f1() R&& f2() 19:28
camelia rakudo-moar 0a12e1: OUTPUT«sekkind␤first␤»
masak && loses its short-circuiting if you R it?
PerlJam apparently
masak is that spec, or just an accident? 19:29
TimToady: ^
lizmat away while stresstests run their course&
masak lizmat++ # avoiding stress
PerlJam lizmat: if you're going jogging, that would have been the perfect phrase to use ;) 19:30
19:31 spider-mario joined
masak I see S03 talking about ops losing their short-circuiting behavior when []'d. but nothing about if they're R'd. 19:31
PerlJam m: sub f1 { say "first"; False }; sub f2 { say "sekkind"; False }; f1() RR&& f2() 19:36
camelia rakudo-moar 0a12e1: OUTPUT«first␤sekkind␤»
bartolin masak: do you think it's worth to adjust two (currently not working) examples from the perl6 book? there were two RT tickets in october and I made a PR to adjust the examples: github.com/perl6/book/pull/82 19:39
masak bartolin: by all means, merge any improvements. 19:42
dalek ok: c27d240 | usev6++ | src/subs-n-sigs.pod:
Fix example in 4.8 Unpacking
19:43
19:43 dalek left, tinyblak left, atroxaper joined 19:44 tinyblak joined
bartolin thanks, masak++ 19:44
19:44 dalek joined, ChanServ sets mode: +v dalek
masak bartolin++ 19:44
m: say 1 [&atan] 1 19:45
camelia rakudo-moar 0a12e1: OUTPUT«Too many positionals passed; expected 1 argument but got 2␤ in sub atan at src/gen/m-CORE.setting:4451␤ in block <unit> at /tmp/G9FDyLqyu8:1␤␤»
masak m: say 1 [&atan2] 1
camelia rakudo-moar 0a12e1: OUTPUT«0.785398163397448␤»
bartolin also moritz++ ;-)
19:46 [Sno] left 19:47 FlipBill joined, [Sno] joined
bartolin I've got another question: is there a preferred way to write test for the REPL? I've seen quite a few tickets which are about a specific behaviour when using the REPL -- and most of them are lacking tests. 19:48
masak advent post draft provisional name: "Snow white and the seven conditionals" 19:49
19:49 atroxaper left
bartolin btw: I really enjoy reading the advent posts. @advent-bloggers»++ 19:55
masak if only there were more...
20:00 tinyblak_ joined 20:04 tinyblak left 20:10 rindolf left 20:15 beastd joined
dalek ast: d91bcc4 | usev6++ | S03-metaops/hyper.t:
Remove duplicate test for unary postfix hyper operator
20:15
20:18 woolfy left 20:21 Isp-sec joined, [Sno] left 20:22 zakharyas left 20:24 woolfy joined
lizmat PerlJam++ for uploading the tarball 20:27
mail with announcement sent
masak lizmat++ 20:28
20:28 darutoko left
moritz lizmat: do you need another ssh key added on the rakudo.org server? 20:29
lizmat moritz: tomorrow, after I upgraded to Yosemite
if that's ok with you :-
)
20:30 bcode left, tinyblak_ left 20:31 bcode joined
moritz lizmat: it is 20:32
lizmat WP entry updated
dalek kudo/nom: 30edf7f | lizmat++ | docs/announce/2014.12.md:
Fix oops so that we can copy next announce
20:34
lizmat everybody have the appropriate amount of fun!
rakudo.org/downloads/rakudo/rakudo-....12.tar.gz # download the tarball 20:35
bartolin lizmat++
20:37 jack_rabbit left 20:42 bcode left
lizmat hopes to see someone pick up a R* release soon 20:42
20:42 bcode joined
dalek Heuristic branch merge: pushed 27 commits to rakudo/newio by lizmat 20:43
20:55 mvuets left
masak advent blog post draft underway. 20:56
20:57 night0x joined
night0x hello 20:57
masak hi, night0x 20:58
night0x where should I start to learn perl6
masak "Begin at the beginning," the King said gravely, "and go on till you come to the end: then stop." 20:59
night0x okay
masak night0x: have you seen the book?
night0x no 21:00
masak github.com/perl6/book
night0x what's that?
masak it's an (unfinished) book about Perl 6.
there's also doc.perl6.org/
and perl6advent.wordpress.com/
raydiak perl6.org/documentation/ has a lot of good links too 21:01
masak and perlgeek.de/blog-en/perl-6/
those are all good starting points.
night0x Oh, thks 21:02
masak oh, and learnxinyminutes.com/docs/perl6/
last but not least, we're here, warm-blooded, and ready to help with any questions you might have.
masak winks: 😉 21:03
night0x it's so nice of you :)
masak don't mention it. 21:04
good luck learning about Perl 6, and stop by often! :D
vendethiel winks as well
Ha, damn, I forgot to push my revisions to the learnx. 21:05
masak masak => 😉 😉 <== vendethiel
night0x I found there are several VMs for perl6. How to choice?
any suggestions?
masak choose moarvm.
unless your boss is forcing you to use JVM.
vendethiel masak: why'd I get two arrows :P? 21:06
masak I ran out of the short type of arrow. 21:07
moritz what's a proper way to do a one-time, concurrency-friendly sleep?
lizmat sleep(1) ? 21:08
masak if `sleep` isn't concurrency-friendly, then something is wrong with it.
moritz isn't there a .delay or .interval or so?
vendethiel masak: do you have a good idea for a possible p6advent I might do? 21:09
dalek href="https://perl6.org:">perl6.org: 0c1a66b | raydiak++ | source/documentation/index.html:
Add 2014 advent calendar link to /documentation
lizmat moritz: what's wrong with sleep(1) ? 21:10
moritz lizmat: does sleep block the current thread?
masak if it doesn't, then something is wrong with it.
lizmat internallu, it calls nqp::sleep 21:11
21:11 kaare_ left
lizmat I assume that knows what to do on Moar to be conc friendly 21:11
21:13 erdic left
jnthn I'm pretty sure nqp::sleep delegates fairly directly to the system call. 21:13
21:14 night0x left
jnthn It will block a thread. That's not always good. The easy way to say "do something a bit in the future" is like Promise.in(10).then({ do_stuff }) 21:14
21:14 hekmek left
jnthn But if blocking the thread is fine, then use sleep. :) 21:15
21:17 shgdfgdfhg joined
shgdfgdfhg atta oing 21:17
hay
lizmat is off to have an appropriate amount of fun
shgdfgdfhg how do i do that
how
masak hovers over the "kick" button 21:18
shgdfgdfhg nm moar'
no shit
jnthn lizmat++ # release
21:18 shgdfgdfhg left
PerlJam lizmat++ every once in a while you should have a wildly inappropriate amount of fun too :) 21:20
21:21 erdic joined
woolfy Fun means this time, one or more episodes of the tv-series Bones, with fresh tea, and maybe some 16-year old whisky, and some nougat. Yoohoo. 21:22
lizmat+= for good ideas...
:-)
jnthn woolfy: += needs a term to the right... :P 21:23
.oO( lizmat+=Inf )
PerlJam heh
bartolin . o O ( lizmat+=2014.12 ) 21:24
woolfy darn, should have been lizmat++
but eternal lizmat without eternal woolfy is no fun...
PerlJam woolfy: So ... when are you going to do a release? :) 21:25
21:31 ptc_p61 joined 21:32 ptc_p6 left
dalek : c632207 | moritz++ | misc/perl6advent-2014/schedule:
Advent: Claim day 23 (Sleepsort)
21:35
timotimo hm, we didn't have sleep sort in the advent calendar yet? 21:36
moritz grep -i sleep misc/perl6advent-20*/schedule # didn't turn up anything
21:37 atroxaper joined
timotimo OK :) 21:38
moritz also did a quick search on the website, nothing
masak Scheduled for 21:39
12/19/2014 0:01
advent post scheduled for tomorrow. 21:40
moritz masak: please add it to the schedule in mu
masak sure.
I'm not saving your bacon again on Saturday, #perl6! you will have to find someone else.
moritz masak++ 21:41
dalek : b113141 | masak++ | misc/perl6advent-2014/schedule:
[schedule] add myself for tomorrow
masak the post is in Wordpress, in the usual place. feel free to review and give early feedback. 21:42
moritz ok, I have the code that I want to write about for day 23
21:43 atroxaper left 21:44 gfldex joined 21:48 night0x joined 21:50 night0x left 21:52 night0x joined
masak LOL! my www.amazon.com/Let-Over-Lambda-Doug...1435712757 has arrived! 21:52
vendethiel niice :) 21:53
masak macros, here I come! 21:54
'night, #perl6 21:55
21:57 [Sno] joined 22:01 spider-mario left 22:02 ptc_p61 left, spider-mario joined 22:09 pmurias joined
pmurias hi 22:09
22:19 ab5tract joined
timotimo moritz: are we planning to get https working for p6c.org? 22:21
gtodd LOL 22:27
22:42 ab5tract left 22:49 colomon left, Isp-sec left 22:51 colomon joined 22:55 night0x left 22:57 Rounin left 22:58 treehug8_ joined 23:01 aborazmeh joined 23:02 gfldex_ joined, pmurias_ joined, jepeway joined 23:03 gfldex left, pmurias left, japhb left, treehug88 left, erdic left, erdic_ joined 23:04 erdic_ is now known as erdic 23:07 treehug8_ left 23:09 japhb joined 23:16 aborazmeh left 23:23 hekmek joined 23:28 atroxaper joined 23:35 atroxaper left 23:37 spider-mario left, beastd left 23:49 virtualsue joined 23:51 jepeway left 23:53 virtualsue left 23:54 telex left 23:56 telex joined