»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.
00:04 bjz left 00:08 yeahnoob joined 00:10 virtualsue left 00:11 virtualsue joined 00:20 telex left 00:21 raiph left 00:22 telex joined 00:23 vendethiel left, nebuchadnezzar joined, vendethiel joined 00:24 raiph joined
dalek ast: dfd7d76 | (Solomon Foster)++ | S03-operators/bag.t:
New bag tests, fudged appropriately.
 00:24
00:29 firnsy joined, firnsy left, firnsy joined 00:33 dayangkun joined 00:34 dayangkun left 00:40 yeahnoob left 00:41 ggoebel111119 left 01:04 virtualsue left
TimToady now in Lexington, VA 01:08
01:09 yeahnoob joined
gtodd camelia: m: $_ = "python3"; tr/python3/ perl6/; .say; 01:09
camelia gtodd: rakudo-moar 502241: OUTPUT«(timeout)» 01:10
TimToady m: say "hi" 01:11
camelia rakudo-moar 502241: OUTPUT«hi␤»
TimToady m: $_ = "python3"; tr/python3/ perl6/; .say;
camelia rakudo-moar 502241: OUTPUT«(timeout)» 01:12
01:14 Mso150 joined 01:20 ggoebel111117 joined
BenGoldberg n: $_ = "python3"; tr/python3/ perl6/; .say; 01:43
camelia niecza : OUTPUT«Cannot open assembly './run/Niecza.exe': No such file or directory.␤» 01:44
BenGoldberg P: $_ = "python3"; tr/python3/ perl6/; .say;
camelia pugs: OUTPUT«/home/p6eval/.cabal/bin/pugs: error while loading shared libraries: libperl.so.5.14: cannot open shared object file: No such file or directory␤»
BenGoldberg j: $_ = "python3"; tr/python3/ perl6/; .say;
camelia rakudo-jvm 502241: OUTPUT«Can't call method "syswrite" on an undefined value at /home/p6eval/jvm-rakudo/eval-client.pl line 32.␤»
BenGoldberg p: $_ = "python3"; tr/python3/ perl6/; .say;
camelia rakudo-parrot 502241: OUTPUT« perl6 ␤»
TimToady hmm, why space? 01:45
01:46 leont left
BenGoldberg Well, the leading space in the output is due to the space here: / perl6/, however the trailing space is a bit odd. 01:46
01:47 kyun joined
TimToady p5 would do " perl66" 01:48
01:49 Mso150 left 02:09 noganex joined 02:12 noganex_ left
j4jackj TimToady: are there any Perl6 implementations written in Perl5? 02:21
02:23 chenryn joined 02:34 raiph left
BenGoldberg j4jackj, Perlito compiles perl6 to perl5. 02:45
j4jackj k
yeahnoob Why Perl 6 is not backend on LLVM? My very old question when I first heard about Perl 6. 02:46
BenGoldberg yeahnoob, If you want to write an LLVM backend for perl6, you are free to do so.
yeahnoob For a noob, i have no such powerful knowledge and skills for now...so 02:50
bonsaikitten so start learning 02:52
02:54 immortal left
tony-o p: $_ = "python3"; tr/python3/ perl6/; .print 03:01
camelia rakudo-parrot 502241: OUTPUT«(timeout)» 03:02
tony-o p: $_ = "python3"; tr/python3/ perl6/; .print;
camelia rakudo-parrot 502241: OUTPUT«(timeout)»
tony-o weird timeouts on eval bot
03:08 molaf joined
yeahnoob Yes Sir. Learning on my way... 03:11
p: $_ = "python3"; tr/python3/ perl6/; .print; 03:12
camelia rakudo-parrot 502241: OUTPUT«(timeout)»
yeahnoob m: $_ = "python3"; tr/python3/ perl6/; .print; 03:13
camelia rakudo-moar 502241: OUTPUT« perl6 »
yeahnoob Maybe they are tweaking the Bot-server. 03:14
03:14 avuserow left 03:16 erkan joined 03:19 molaf left
kyun Well done, the rakudo star update to 2014.09 03:19
03:42 chenryn left 04:12 chenryn joined 04:17 chenryn left 04:18 anaeem1_ joined 04:33 BenGoldberg left 04:40 SamuraiJack_ joined 05:16 mberends left 05:45 avuserow joined 05:50 kaleem joined, rindolf joined 05:55 jack_rabbit left 06:13 ecocode joined
sergot_ morning o/ 06:20
06:21 chenryn joined
kyun Good afternoon. 06:25
06:25 dayangkun joined 06:40 rindolf left 06:41 SamuraiJack_ left 06:44 rindolf joined 06:48 atroxaper joined
atroxaper Hello, #perl6 ! 06:49
06:58 FROGGS joined 07:01 virtualsue joined 07:04 zakharyas joined 07:12 erkan left 07:16 Mso150 joined 07:19 darutoko joined 07:24 dayangkun left 07:28 _slade_ left 07:49 odc joined 07:50 darutoko- joined 07:54 darutoko left 07:58 denis_boyun_ joined 08:04 fhelmberger joined 08:06 amkrankruleuen joined, fhelmberger left 08:11 virtualsue left 08:31 dakkar joined 08:48 kaare_ joined 08:58 mberends joined 09:03 erkan joined, erkan left, erkan joined 09:04 Mso150 left 09:05 ghostlines joined
lizmat good *, #perl6! 09:06
dalek ast: 2f5ef2a | (Elizabeth Mattijsen)++ | S02-types/WHICH.t:
Remove 4 classes to be removed
 09:07
09:15 virtualsue joined, denis_boyun_ left 09:16 ggherdov___ joined
yeahnoob m: my $t=localtime; say "Good ",$t; 09:17
camelia rakudo-moar 502241: OUTPUT«===SORRY!=== Error while compiling /tmp/XwdjQuW54H␤Undeclared routine:␤ localtime used at line 1␤␤»
09:17 denis_boyun joined
yeahnoob m: my $t=DateTime.new(now); say "Good ",$t; 09:17
camelia rakudo-moar 502241: OUTPUT«Good 2014-09-29T09:17:11Z␤»
09:17 pecastro joined
lizmat it appears $*PROGRAM_NAME is a P5ism 09:17
lizmat thinks she's going to deprecate it 09:18
09:22 brrt joined
yeahnoob m: my $t=DateTime.new(now); $t=$t.in-timezone($*TZ); say "Good Time & Time Again, ",$t; 09:23
camelia rakudo-moar 502241: OUTPUT«Good Time & Time Again, 2014-09-29T05:23:21-0400␤»
09:26 yeahnoob left
FROGGS .u arrow 09:33
yoleaux 28 Sep 2014 21:19Z <lizmat> FROGGS: seems the 5daa5de broke the parrot build
U+02C2 MODIFIER LETTER LEFT ARROWHEAD [Sk] (˂)
U+02C3 MODIFIER LETTER RIGHT ARROWHEAD [Sk] (˃)
U+02C4 MODIFIER LETTER UP ARROWHEAD [Sk] (˄)
FROGGS eww
09:38 bjz joined
lizmat FROGGS: yeah, that was quite a big commit, so I have not been looking at a solution yet :-( 09:39
FROGGS no problem... I'll fix it
09:42 denis_boyun left
FROGGS lizmat: your last label Patch will break my panda :o( 09:45
looks like it need to find a third way :/
ahh, perhaps I already know how... 09:46
dalek ecs: 51bf410 | (Elizabeth Mattijsen)++ | S32-setting-library/IO.pod:
chmod/unlink() return files changed/removed
 09:49
lizmat FROGGS: the one that unbroke the JVM build ? 09:51
FROGGS lizmat: yeah 09:52
problem seems to be that we get an IO::Path outside of the setting 09:53
lizmat that shouldn't be a problem ?? 09:54
FROGGS well, you just can't p6box_s and IO::Path 09:57
lizmat hmmm/... so where does $?FILES become an IO::Path then ? 09:59
FROGGS m: say nqp::getlexdyn('$?FILES') 10:00
camelia rakudo-moar 502241: OUTPUT«/tmp/jE8B_GmN5f␤»
10:00 kyun left
FROGGS m: sub foo { say nqp::getlexdyn('$?FILES') }; foo 10:00
camelia rakudo-moar 502241: OUTPUT«/tmp/lMczF189L4␤»
FROGGS hmmm, maybe in a procompiled module or so?
because that is where it exploded... after precompiling Archive::Tar, when running its tests 10:01
10:07 virtualsue left
lizmat hmmmm... well, $?FILES should be precompiled and frozen 10:09
maybe the serializer / deserializer cannot handle IO::Path ?
where does that live anyway ? 10:10
FROGGS probably here: 10:11
rakudo/src/gen/ModuleLoader.nqp:264: my $?FILES := %chosen<pm>;
lizmat aha, I think I know where that get's set 10:12
could you try: 10:13
- $candi<provides>{$module_name}<pm><file>;
+ ~$candi<provides>{$module_name}<pm><file>;
in src/core/CompUnit.pm ?
line 100 10:14
10:19 virtualsue joined
FROGGS yeah, will try that in a bit 10:20
I'll fix parrot first 10:22
dalek ecs: 9131eb7 | (Elizabeth Mattijsen)++ | S32-setting-library/IO.pod:
Add mode parameter to mkdir()/.mkdir
 10:30
10:33 cosimo joined 10:39 denis_boyun joined 10:40 cosimo left 10:43 atroxaper left 10:44 atroxaper joined 10:47 tharvesh joined 10:51 tharvesh left 10:57 atroxaper left, atroxaper_ joined
timotimo o/ 10:58
lizmat timotimo o/ 11:01
timotimo meh 11:05
i wanted to play overgod (for old time's sake or something) and it segfaults at startup .. i can install the separate debuginfo package, but all it shows is "<optimized out>"
11:07 xinming joined
nine timotimo: that means that the value probably is in some CPU register 11:08
brrt timotimo o/
11:09 virtualsue left
timotimo nine: yeah, but that doesn't help me figure out how a null pointer gets its way into the src argument of some blitting function >_< 11:12
11:14 sqirrel__ joined 11:18 virtualsue joined 11:19 sqirrel__ left 11:23 chenryn left 11:32 atroxaper_ left, atroxaper joined 11:39 cosimo joined 11:43 leont joined 11:50 prevost joined, colomon left 11:53 colomon joined, chenryn joined
dalek kudo/nom: 17db5ce | (Tobias Leich)++ | src/Perl6/Grammar.nqp:
unbreak parrot build
 11:54
11:55 telex left 11:56 telex joined 11:59 rindolf left 12:10 rindolf joined
lizmat FROGGS: that line seems rather essential? 12:11
12:14 yeahnoob joined
timotimo is a \ at the end of a line always acceptable in the repl to give a new line of input? 12:15
as in: if i'm inside a string literal, will that work as expected? + all edge cases i may not have thought about
dalek rl6-roast-data: 076a6ed | coke++ | / (5 files):
today (automated commit)
 12:16
[Coke] ^^ yesterday 12:17
lizmat timotimo: I think that could only work if the REPL would always execute *all* lines entered from scratch 12:18
with perhaps $*OUT and $*ERR localized to dev/null in all but the last line 12:19
timotimo er, what? 12:20
oh, no, the repl has a little piece of code in it that'll ask for another line if the last one ended in a backslash 12:21
12:22 araujo joined, araujo left, araujo joined 12:28 rindolf left
dalek ast: bf164e1 | (Elizabeth Mattijsen)++ | S02-types/WHICH.t:
Argh, we need these classes (for now at least)

This reverts commit 2f5ef2aa78e28cfaac8edb1f5d61dfb99a12c7d7.		 12:34
timotimo m: say 1, 1, *+* ... *
camelia rakudo-moar 17db5c: OUTPUT«1 1 2 3 5 8 13 21 34 55 89 144 233 377 610 987 1597 2584 4181 6765 10946 17711 28657 46368 75025 121393 196418 317811 514229 832040 1346269 2178309 3524578 5702887 9227465 14930352 24157817 39088169 63245986 102334155 165580141 267914296 433494437 70140873…»
12:35 chenryn left 12:38 rindolf joined 12:43 chenryn joined
lizmat m: sub a(:$*SPEC) { say $SPEC }; a # hoped this would say IO::Spec::Unix 12:46
camelia rakudo-moar 17db5c: OUTPUT«===SORRY!=== Error while compiling /tmp/Rz1bA_y4vB␤Variable '$SPEC' is not declared␤at /tmp/Rz1bA_y4vB:1␤------> sub a(:$*SPEC) { say $SPEC⏏ }; a # hoped this would say IO::Spec:␤ expecting any of:␤ …»
lizmat hmmm... -e discrepancy :-)
would it make sense to have :$*SPEC be a shortcut for :SPEC($*SPEC) ? 12:47
12:49 chenryn left 12:54 rindolf left
FROGGS lizmat: sort of, it is meant to make $*STRICT available in EVALs but it does not work correctly with this line on moar either 13:05
13:05 anaeem1_ left
dalek kudo-star-daily: 2b8a039 | coke++ | log/MoarVM-version.log:
today (automated commit)
 13:05
FROGGS lizmat: yes, because "SPEC" is the name # :$*SPEC be a shortcut for :SPEC($*SPEC) 13:06
jnthn m: sub a(:$foo) { say $foo }; my $*foo = 42; a(:$*foo)
camelia rakudo-moar 17db5c: OUTPUT«42␤»
jnthn Already works, no? :)
FROGGS jnthn: bah, just because you did it right! :P
m: sub a(:$foo) { say $foo }; a(:$<foo>) 13:07
camelia rakudo-moar 17db5c: OUTPUT«Nil␤»
13:08 rindolf joined
jnthn m: 'abc' ~~ /$<foo>=[\w+]/; sub a(:$foo) { say $foo }; a(:$<foo>) 13:11
camelia rakudo-moar 17db5c: OUTPUT«｢abc｣␤␤»
jnthn ;)
13:11 kaare_ left
FROGGS jnthn: btw, we could take a look at your USAGE thingy again when you've time 13:12
jnthn FROGGS: Did you manage to reproduce it at all? 13:15
FROGGS jnthn: no, that's the point :/ 13:16
jnthn urgh
FROGGS I see that 'print' is pretty expensive on windows but that's kinda expected
jnthn Let me grab latest and produce another profile, then upload it, then at least we can check it's not just me misreading something...
timotimo yeah, it has to go through the spoon or what it's called 13:17
FROGGS (given the suckiness of cmd.exe)
timotimo and then through a parallel port cable
jnthn There Is No Spoon :P
timotimo what if LPT1 isn't even connected?
you can't detect that!
FROGGS well, *I* can, at least locally
jnthn: that's my profile: froggs.de/perl6/rcff-windows.html 13:18
jnthn Yeah, I saw that one
oh wait, I have your other one open
13:18 atroxaper left
FROGGS that one is the only profile on windows I've shared 13:19
jnthn yeah, it musta been a non-Windows one then
gtodd FROGGS: re: "Memory allocation failed; could not allocate 32 bytes" I though I got that error with panda only ... but it has disappeared again mysteriously maybe it's hardware or sunspots 13:20
jnthn Of course, it's taking ages to load, because first we need to make sure it doesn't talk about falun gong, hong kong democracy, and so forth... :/
timotimo ;( 13:21
jnthn 2 days until I have non-stupid Internet again :)
gtodd m: $_ = "python3"; tr/python3/ perl6/; .say;
camelia rakudo-moar 17db5c: OUTPUT« perl6 ␤»
FROGGS gtodd: ohh, that might be a more general issue (on your box) rather then a problem in v5's grammar
gtodd m: my $coding = "python3"; $coding.=trans(python3 => perl6); $coding.say ;
camelia rakudo-moar 17db5c: OUTPUT«===SORRY!=== Error while compiling /tmp/3zdIi4vUcI␤Undeclared routine:␤ perl6 used at line 1␤␤»
FROGGS gtodd: nice try :o) 13:22
gtodd can we use tr/// on lexicals ?
m: my $coding = "python3"; $coding.=trans("python3" =>" perl6"); $coding.say ;
camelia rakudo-moar 17db5c: OUTPUT« perl6 ␤»
FROGGS m: my $c = "python3"; $c ~~ tr/python3/ perl6/; $c.say;
camelia rakudo-moar 17db5c: OUTPUT« perl6 ␤»
timotimo m: say " perl6".chars; say "python3".chars 13:23
camelia rakudo-moar 17db5c: OUTPUT«6␤7␤»
FROGGS m: my $c = "python3"; say $c ~~ tr/python3/ perl6/;
camelia rakudo-moar 17db5c: OUTPUT«StrDistance.new(before => "python3", after => " perl6 ")␤»
gtodd ah ok
timotimo m: my $a = "abcdefghijklmnop"; $a ~~ tr/abcdefgh/1234/; say $a
camelia rakudo-moar 17db5c: OUTPUT«12341234ijklmnop␤»
timotimo that's why it puts a space there
grondilu m: my num $ = 1.Num; # this works
camelia ( no output )
grondilu m: my num @ = 1.Num; # this doesn't?
camelia rakudo-moar 17db5c: OUTPUT«Cannot modify an immutable Num␤ in block <unit> at /tmp/k1ZEvlSnCn:1␤␤»
gtodd m: my $coding = "python3"; $coding.= tr/python3/ perl6/ ; $coding.say ; 13:24
camelia rakudo-moar 17db5c: OUTPUT«===SORRY!=== Error while compiling /tmp/cEE7Fv0uNP␤Bogus statement␤at /tmp/cEE7Fv0uNP:1␤------> "python3"; $coding.= tr/python3/ perl6/ ⏏; $coding.say ;␤ expecting any of:␤ argument list␤ …»
gtodd oops
ok ~~
FROGGS .= is for methods 13:25
jnthn FROGGS: Did you do 10 iterations or 5?
FROGGS jnthn: 5
jnthn FROGGS: Even if your one, 10.46% is spent in gen-usage
grondilu thinks there is "trans" for a method version of "tr"
jnthn (in
FROGGS: Maybe not in it directly, but in things it does 13:26
timotimo gtodd: did you see what i ran up there?
FROGGS jnthn: where do I have to look?
gtodd yeah ...
timotimo :)
grondilu m: say "foo".trans: /foo/ => "bar"
camelia rakudo-moar 17db5c: OUTPUT«bar␤»
jnthn FROGGS: On routines tab on the one you just linked me, search for gen-usage
FROGGS ohh
FROGGS facepalm
jnthn m) 13:27
FROGGS well, perhaps it makes sense to read it that way :o)
gtodd timotimo: so far I'm think p6's tr/// is smarter than perl5 tr// but you have to use ~~
timotimo if you tr without ~~, it'll just apply to $_
jnthn FROGGS: Inclusive means "time spent between entering gen-usage and leaving gen-usage, including the things it calls"
FROGGS jnthn: yeah, I understand now 13:28
gtodd I'm also thinking p6 is like crack ....
jnthn I note that strip_path_prefi, right below it, is 10.29%
FROGGS jnthn: I'll have a look at that this evening
timotimo but crack is bad
jnthn And the call graph shows that gen-usage calls that
gtodd ok "crack" is the wrong word .... "kale" once you start using it it's hard to not have it with every meal
jnthn FROGGS: OK, cool. So do you need me to upload my one? :) 13:29
timotimo i don't know what kale is :S
FROGGS jnthn: no, thanks :o)
jnthn timotimo: Maybe it's a word for beer in some language... :)
timotimo hah :D
gtodd it's like arugula salad .. hard to eat iceberg lettuce after that :-)
timotimo a friend let me try a "creamy root beer" thingie ... 13:30
like, imported from the americas
it tasted like mouthwater
he left it here, because he didn't want it
i don't want to finish it, either
gtodd heheh
timotimo "blue sky natural soda" 13:31
"made with real sugar"
gtodd ?
heh
timotimo "creamy root beer"
gtodd searched for p6's tr/// docs .... 13:32
really I did
timotimo for many things there's not really docs ... but the specs are always there :\
gtodd perl6advent.wordpress.com/2010/12/2...nd-beyond/ did not have the bit about tr/// hmm could add it there as a comment :-D 13:33
jnthn timotimo: That sounds...no like beer :P 13:34
*not
timotimo yeah, i don't consider this beer at all
gtodd would be nice to have a bot we could tell to set aside the above $X lines somewhere as possible rough docs .... or maybe use yoleaux .tell docmunger to write a paragraph using the above 10 lines" ... 13:36
timotimo teh what now? 13:37
gtodd haha
FROGGS a bot taking the irc logs and stick it in the docs?
O.o
timotimo oh? 13:38
well, that won't really give a prose-like kind of thing
gtodd heheh not official docs ... I often find stuff in the irc logs that is helpful ...
timotimo it should instead generate a markov chain from the lines
and generate a book and sell it on amazon based on that
gtodd hahaha
$kaching!
moritz hey, I have the data already 13:39
FROGGS let's sell moritz for $$$!! /o/
:P 13:40
timotimo don't sell the cow if you can still milk it!
FROGGS ohh, very true
gtodd heheh ... #perl has GumbyBrain ... who I'm sure is very wealthy :-)
I was just cutting and pasting stuff into a text file when an example explanation was helpful and memorable and using POD markup to make it "grepable" 13:42
it worked great until I lost that disk :-) 13:43
FROGGS :o) 13:45
gtodd and there's something beautiful about using bots in clouds for niceness instead of evil
bots have a bad rep in the "mainsteam media" ... :) 13:46
moritz hugme: hug mainstream 13:47
hugme hugs mainstream
timotimo hugme: hug hug 13:48
hugme hugs hug
timotimo hugme: hug hugs
hugme hugs hugs
PerlJam "mainstream media" is stupid too. 13:50
I heard the term "hackathon" in a TV show last night, but it was being used as something that "evil hackers" do 13:51
timotimo wat ... srsly?
there was this time when fox news (or something similar) reported about github
FROGGS I guess these hackers wired their own telephone line to their line-out or something... 13:52
lizmat those commie programmers!
13:53 kaleem left, raiph joined, rindolf left
dalek kudo/nom: 1f4ca04 | (Tobias Leich)++ | src/core/CompUnit (2 files):
stringify CU candidates that can become $?FILES

When we let slip an IO::Path into $?FILES the Label.new code will explode. Beware.		 13:55
FROGGS uh, lizmat++
Woodi hi today :) 13:57
raiph gtodd: vaguely similar idea: questhub.io/realm/perl/quest/53516...ef19000022
lizmat any Cygwin people here? would C:\\foo be a path that you could get on Cygwin ? 13:58
FROGGS hi Woodi
lizmat I *thought* it hid all of that in a unix-like access to filesystems ? 13:59
colomon lizmat: I think it's spelled something like /cygwin/C/foo there
FROGGS lizmat: I deleted my cygwin like four hours ago :o(
13:59 atroxaper joined, rindolf joined
colomon sorry, /cygdrive/C 13:59
colomon isn't sure how old the cygwin on his Windows Vista box is, though... 14:00
lizmat so a test like "my $path = IO::Path::Cygwin.new('C:foo\\\\bar\\');" is bogus
colomon well, I suppose IO::Path::Cygwin might be smart enough to translate 14:01
but
ls C:
lizmat it isn't
hence my question... :-)
colomon ls: cannot access C:: No such file or directory
lizmat colomon++ thanks for the reality check 14:02
colomon lizmat: aren't you defining it right now? :)
Woodi I'm thinking and thinking and I think that "mainstream media" don't know what 'bots' are... droids, drones, viruses @ Co, yes... but if "gamers media" counts as mainstream then yes :) or maybe I lack some references... or we talking about is shadow-o-bot-net ? :)
lizmat colomon: well, I'm trying to get the code to be like the spec
and suddenly tests start to fail :-)
colomon lizmat: and you're rewriting the spec, no? :)
lizmat well, I would say generalizing 14:03
on the outside not a lot would change
API's mostly stay the same
"foo".IO.e still works :-)
gtodd raiph: nice :-) 14:04
lizmat even though underneath it creates an IO::Path rather than an IO::Handle
"bar".IO.lines still works, and even closes the handle :-)
14:09 prevost left 14:20 mberends left 14:21 anaeem1 joined 14:27 Guest33157 left 14:30 kaare_ joined, zoosha_ left
Woodi lizmat: I think that IO::Path things are word-y, ppls probably will just want to have: open("/some/string"), etc., something that just work... 14:30
lizmat Woodi: don't despair, open() will continue to work 14:31
14:32 kaleem joined
Woodi lizmat: it is not despair :) just lazines, maybe not good one but kind of natural, IMO 14:33
14:35 H2O1 joined
lizmat we'll cater for the lazy ones as well :-) 14:35
14:36 H2O1 left
Woodi btw. I managed to upload redis corrections to github, hope they are "corrections" ;) github.com/slunski/perl6-redis.git 14:36
14:36 bowtie joined, bowtie is now known as Guest71734
Woodi lizmat: probably I just lazy and don't know current API in spec :) I will check it before talkoing more... 14:36
hoelzro o/ perl6 14:37
lizmat hoelzro o\ 14:38
14:40 rindolf left 14:42 rindolf joined 14:47 kaleem left
dalek kudo/nom: a6af28f | (Elizabeth Mattijsen)++ | src/core/IO/Spec/ (3 files):
Use "dirname" instead of "directory"
 14:49
kudo/nom: 4191d07 | (Elizabeth Mattijsen)++ | src/core/IO/ (3 files):
Implement IO::Spec.dirname (instead of .directory)

  .directory will still work, but is deprecated
FROGGS lizmat++ 14:50
14:50 Hor|zon joined
dalek ast: a0bf974 | (Elizabeth Mattijsen)++ | S32-io/io-spec- (3 files):
Bring IO::Spec tests up to spec wrt 'dirname'
 14:51
ecs: c4a4fc1 | (Elizabeth Mattijsen)++ | S32-setting-library/IO.pod:
Some more s/directory/dirname/
 15:02
ecs: a8dd445 | (Elizabeth Mattijsen)++ | S32-setting-library/IO.pod:
Add paragraph about IO::Path subclasses
15:05 zoosha joined 15:10 Hor|zon left 15:16 Guest71734 left, bowtie_ joined, jerrycheung joined 15:17 jerrycheung left 15:18 Hor|zon joined 15:24 Hor|zon left 15:25 Hor|zon joined 15:27 brrt left 15:30 denis_boyun left
gtodd is a bit slow ... 15:31
dalek kudo/nom: 658409a | (Tobias Leich)++ | src/core/Main.pm:
optimize $?USAGE

We only generate the USAGE message in case we need it, and then the new code only takes .13s where the old code took .75s (on windows)		 15:32
15:33 bowtie_ left 15:34 bowtie joined 15:35 bowtie is now known as Guest25531, zakharyas left
gtodd when we do this --> my $code = "perl5"; say $code ~~ tr/perl5/perl6/; (i.e. the ~~ with tr/// instead of the method .trans) is that using the "infix ~~ " ... i.e. is it the smart match in another guise ?? 15:35
since smartmatch ~~ aliases stuff to $_ and checks ACCEPT(S_) etc etc
15:35 raiph left 15:36 ecocode left 15:39 FROGGS left 15:40 ghostlines left 15:43 Guest25531 left 15:45 bowtie_ joined
gtodd errm ACCEPTS($_) 15:48
15:50 Hor|zon left 15:53 jerrycheung joined 15:54 FROGGS joined, jerrycheung left, bowtie_ left 15:55 bowtie_ joined 15:57 yeahnoob left 16:00 psch joined
psch gtodd: saw your question in the clog. in nom $ ~~ tr/// currently doesn't call .ACCEPTS, because StrDistance doesn't implement an ACCEPTS method yet. 16:02
m: StrDistance.^methods.say
camelia rakudo-moar 4191d0: OUTPUT«Bool Numeric Int <anon> <anon>␤»
psch i have opened a PR, but iirc it hasn't been discussed yet, probably because i didn't ask anyone to look at it :) 16:03
gtodd: basically, .ACCEPTS is for creating the boolean value that &infix:<~~> returns, afaiu. $ ~~ s/// and $ ~~ tr/// currently are both special-cased. 16:04
gtodd: github.com/rakudo/rakudo/blob/nom/....nqp#L4925 in case you haven't found it yet. 16:06
16:06 rindolf left 16:07 bowtie_ left 16:10 bowtie_ joined 16:14 bowtie_ left
psch m: $_ = "abc"; say so s:g/x/u/; 16:15
16:16 anaeem1 left, Hor|zon joined, anaeem1_ joined
gtodd psch: thanks 16:16
camelia rakudo-moar 4191d0: OUTPUT«(timeout)»
16:17 anaeem1_ left 16:19 bowtie_ joined 16:20 atroxaper left 16:21 atroxaper joined
gtodd camelia: say "hi" unless 0; # :-D 16:21
... camelia is cocooning again 16:22
RT #122810 16:25
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=122810
16:26 atroxaper left
gtodd that reminds me ... I opened that PR ... then wrote a test (probably wrong or unecessary) and committed it to roast somewhere 16:26
and then closed the PR
16:26 bowtie_ left
gtodd the test then disappeared from current roast or was moved/edited - it was written in a way that encouraged editing :) 16:27
I really only wrote the test because jnthn fixed the issue but said tests are preferred for closing PRs :-| 16:28
16:32 Mso150 joined, bowtie joined 16:33 bowtie is now known as Guest75651
gtodd (and to learn a bit about the bug->PR->test workflow) ... anyway, I left the bug closed since TTTT I wasn't sure where or if it belonged - it seemed more MoarVMish maybe I should stick a link to it in Moar's github issues 16:33
[Coke] there isn't a hard policy, but rakudo-moar bugs probably still belong in RT, but you -could- bug them in MoarVM/MoarVM on github... 16:35
I imagine the latter as "bugs in the VM". ... of course, there is no standalone VM stuff, it's all through perl.
16:39 virtualsue left 16:40 Guest75651 left, bowtie_ joined 16:45 firefish5000 joined 16:47 bowtie_ left, anaeem1 joined 16:53 bowtie_ joined 16:54 jack_rabbit joined 16:55 Hor|zon left 16:56 dakkar left 16:58 Hor|zon joined 17:07 gfldex joined
pmichaud good morning, #perl6 17:15
FROGGS hi pmichaud!
tony-o FROGGS: did removing the .dylib|.so in DBIish Pg work in linux? 17:16
FROGGS tony-o: yes
vendethiel o/, #perl6 17:18
colomon \o
lizmat pmichaud o/ 17:21
17:22 atroxaper joined
timotimo o/ pmichaud 17:25
17:27 atroxaper left 17:28 spider-mario joined 17:33 leont left 17:36 bowtie_ left 17:38 erkan left 17:39 pecastro left, bowtie joined, bowtie is now known as Guest2662, erkan joined 17:42 mberends joined 17:50 Guest2662 left, bowtie_ joined 17:52 araujo left 17:57 anaeem1 left 18:00 virtualsue joined 18:01 Ven joined 18:02 anaeem1_ joined 18:07 anaeem1_ left 18:08 Hor|zon left
lizmat I have a giant commit that is getting larger and larger, but which isn't spectest clean 18:17
but I'd like to get it out there to more quickly find out if I'm breaking other stuff as well 18:18
should I commit to nom or not? 18:19
putting in a branch is not really a fix either...
opinions anyone? 18:20
FROGGS branch 18:25
[Coke] pmichaud: hio. 18:26
PerlJam lizmat: if you want people to try it but are concerned about breaking stuff, a branch sounds like the right thing. (You just have to "advertise" the branch a little bit) 18:29
18:34 renormalist left
dalek nda/CPAN: 407089a | (Tobias Leich)++ | ext/HTTP__UserAgent:
updated HTTP::UserAgent
 18:35
18:36 Ven left
FROGGS lizmat: to let travis build it you would need to call it smoke-me-foobar 18:36
lizmat well, I know it builds :-) 18:37
well, on moar :-)
FROGGS on your box ... :o)
lizmat true
18:37 H2O1 joined
FROGGS how big is the spectest impact? would you be able to fix All The Things in a reasonable timeframe? 18:37
lizmat matter of days max 18:38
FROGGS that really sounds like a branch would be the best
[Coke] please no one ever intentionally break spectest for days. 18:39
lizmat okokokok :-)
[Coke] I will allow "inadvertently for one commit" but not much more. :P
colomon has been trying to be good with his latest roast commits…
18:40 H2O1 left, Ven joined, psch left
dalek nda/CPAN: 548a938 | (Tobias Leich)++ | lib/Panda/Fetcher.pm:
use UA.get instead of getstore() for binary data

getstore() is meant to work for text, so is not ideal for fetching .gz files		 18:47
18:48 anaeem1 joined, telex left
PerlJam why does getstore care about the disposition of the data it's downloading and writing to a file? 18:48
FROGGS that works again now: panda --cpan --github search native # gist.github.com/FROGGS/4e97e0d8db94df7b3713 18:49
18:50 renormalist joined
FROGGS PerlJam: I dunno 18:50
18:50 telex joined
FROGGS I'm going to work on `panda --cpan install NativeCall` now, which requires a working gzslurp.... we'll see 18:52
18:53 anaeem1 left
vendethiel FROGGS++ :-) 18:59
FROGGS :o)
vendethiel made a list of everything he still wants to talk about the learnxiny, and now thinks about doing a perl6 talk about those instead 19:00
FROGGS yeah, sounds good
19:02 brrt joined 19:03 Mso150_g joined, Mso150 left 19:06 darutoko- left, ivanshmakov joined 19:09 denis_boyun_ joined 19:11 bartolin joined 19:20 beastd joined
gtodd FROGGS: does panda self update a la cpanm ? 19:25
I use rebootstrap.pl 19:26
but if there is a better way ...
FROGGS you can install panda via panda, yes, that should work 19:27
gtodd :)
FROGGS yeah, panda works quite well fortunately :o) 19:28
gtodd I don't see the --cpan --github switches so I should update :) 19:30
dalek kudo/nom: fb6a7ba | usev6++ | src/core/ (3 files):
Extend DISTRO and KERNEL for OpenBSD and NetBSD
kudo/nom: 264eafb | (Tobias Leich)++ | src/core/ (3 files):
Merge pull request #314 from usev6/nom

Extend DISTRO and KERNEL for OpenBSD and NetBSD
gtodd vendethiel: for no good reason I'm trying to make the learnxiny code samples into a perl6 script I can exec to produce errm the html page ... :-P not quite literate programming yet but I'm getting there ! 19:32
vendethiel gtodd: how, good luck :) 19:34
[Coke] is perl6-bench reporting on rakudo-jvm? 19:40
nine I guess we care about compatibility to Windows, don't we?
[Coke] (any slowdowns lately?)
nine: yes.
nine darn
[Coke] (assuming you mean "run on windows") 19:41
nine So what do I do if I need unistd.h in p5helper.c to get the environ variable to pass to PERL_SYS_INIT3?
moritz nine: do you need to do that? or is it enough to call PERL_SYS_INIT without the env? 19:42
also, perlembed has this example: 19:43
int main(int argc, char **argv, char **env)
{
PERL_SYS_INIT3(&argc,&argv,&env);
nine moritz: currently I pass NULL for argv and env, but carlin++ found that this breaks on openBSD
Better question: is there a way to get at argc, argv and env? 19:44
moritz nine: yes, but PERL_SYS_INIT (no trailing 3) doesn't need an env
also, that perlembed example suggest that you might get get env from the signature of main
FROGGS nine: look at how it is solved in MoarVM
moritz no idea how portable that is
19:45 Psyche^ joined
nine oh, I didn't know there was a PERL_SYS_INIT without the 3. perlembed doesn't mention it, but perlapi does and it has exactly the same description as the other version 19:46
moritz I left a comment on the commit in carlin's pull request 19:48
19:49 Psyche^_ left
nine Seems like %ENV works just fine in P5 with PERL_SYS_INIT 19:50
19:50 bowtie_ left
moritz p5 probably knows how to access the ENV :-) 19:50
nine Also I guess it would be more safe to make argc and argv global variables. Who knows what PERL_SYS_INIT does with those pointers... 19:51
[Coke] has an Albany.pm meeting tonight. 19:53
19:53 bowtie joined, bowtie is now known as Guest23430
[Coke] github.com/japhb/perl6-bench is just source, yes? do we have pretty pictures hosted? 19:54
japhb: is "./bench quickstart" expected to work with an $http_proxy set? 19:55
19:55 avuserow left
dalek nda/CPAN: 30e2a79 | (Tobias Leich)++ | ext/Archive__Tar/lib/Archive/Tar (2 files):
unbreak Arhive::Tar
 19:58
nda/CPAN: 0f57540 | (Tobias Leich)++ | lib/Panda/Fetcher.pm:
remove debug output
FROGGS it almost seems as if one could install stuff from CPAN now O.o 20:00
20:01 araujo joined, araujo left, araujo joined
nine Interesting: the only platform where PERL_SYS_INIT3 actually does anything with env is OS2. All other platforms map directly to PERL_SYS_INIT (which uses argc, argv for checking for taint mode) 20:01
20:01 avuserow joined 20:02 kjs_ joined
[Coke] the jvm run is usually done hours ago. it's on S12. 20:04
20:04 Hor|zon joined
dalek line-Perl5: fd041ae | Carlin++ | p5helper.c:
Fix segv calling Perl_sys_init3 on some platforms

eg. OpenBSD		 20:05
line-Perl5: 18877ee | Carlin++ | p5helper.c:
do not pass env parameter

per suggestion by moritz++
line-Perl5: 37ce3ad | niner++ | p5helper.c:
Merge pull request #7 from carbin/stop-segv-on-obsd

Fix segv calling Perl_sys_init3 on some platforms
nine carlin: from reading the P5 source code, I'd say it's not dependend on platform but if perl is compiled with its own malloc or not. But many thanks for fixing my bug :) 20:09
20:09 Hor|zon left 20:10 rmgk_ joined, rmgk is now known as Guest26726, Guest26726 left, rmgk_ is now known as rmgk 20:12 gfldex left
carlin nine: ah, that would explain it 20:15
20:16 ggoebel111117 left 20:21 Mso150_g left 20:22 Hor|zon joined 20:23 immortal joined 20:24 [Sno] joined 20:26 erkan left 20:27 Hor|zon left 20:29 kaare_ left 20:31 FROGGS left 20:35 erkan joined 20:37 immortal left 20:38 wk joined
wk Hi! 20:40
What happened with :replace argument to the .pick?
.pick docs does not mention it (doc.perl6.org/routine/pick) 20:41
yoleaux wk: You must provide at least two options.
colomon wk: I believe that's now the .roll method
m: say (1..10).pick(20) 20:42
20:42 zeropool joined
wk and i tried (1..6).pick(10, :replace); 20:42
camelia rakudo-moar 658409: OUTPUT«8 1 4 6 5 10 2 3 7 9␤»
colomon m: say (1..10).roll(20)
camelia rakudo-moar 658409: OUTPUT«7 2 5 5 6 5 2 5 10 8 6 9 9 7 5 7 4 2 3 9␤»
wk got just 6
colomon: ok, i'll look for it
colomon right, :replace is gone.
20:43 prevost joined
wk colomon: did not found any indication for that, just that .grab does not have it anymore 20:43
colomon wk: where are you looking? 20:44
wk colomon: googled for "perl6 pick replace", got only results telling it should work 20:46
20:46 molaf joined
bartolin speaking of .pick: The following happened on a netbsd box: 20:47
perl6-p -e 'say (1..2).roll(20)'
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
(could be 2 2 2 ... as well)
with (1..3) it looks better: 20:48
2 2 1 3 1 3 1 1 3 3 2 1 3 1 1 3 2 3 2 3
carlin bartolin: the BSDs have a broken rand() implementation
MoarVM uses arc4random() there by default
lizmat bartolin: probably a rand() issue
carlin I guess parrot doesn't
lizmat m: $ perl6 -Mstrict -e1
===SORRY!===
Could not find strict in any of:
colomon wk: huh, when I do that google search, the first result is a spectest which clearly indicates :replace is an error, and the third result is the actual commit to the spec that changed pick :replace to roll.
the second result is an old blog post which is no longer corrent. :( 20:49
bartolin Ah! Thanks.
camelia rakudo-moar 264eaf: OUTPUT«(timeout)»
colomon not sure if I meant current or correct there….
lizmat M: class A { has Str $a; $a = "" } # segfaults for me
m: class A { has Str $a; $a = "" } 20:50
camelia rakudo-moar 264eaf: OUTPUT«(timeout)»
20:53 immortal joined
wk colomon: thank you for clearing things! have you thoughts, why this functionality was deprecated(moved to roll)? 20:53
i got the example from from the same 2009 advent calender 20:54
20:55 erkan left
bartolin carlin: Do you also have an idea why "1**Inf" gives "NaN" on said netbsd box? (happens with Moar and Parrot) Could that also be a OS specific thing? 20:56
m: say 1**Inf 20:57
camelia rakudo-moar 264eaf: OUTPUT«(timeout)»
carlin bartolin: no idea, sorry
colomon wk: discussion is on this day, I think: irclog.perlgeek.de/perl6/2010-09-15 20:58
bartolin carlin: Thanks anyway! ;-)
wk colomon: thank you
colomon wk: maybe? That's around the time the change was made, anyway. 20:59
21:00 atroxaper joined
carlin 1**Inf = 1 on Linux, FreeBSD and OpenBSD. NetBSD is the odd-one-out there 21:01
21:03 erkan joined, erkan left, erkan joined 21:04 spider-mario left, atroxaper left 21:05 Hor|zon joined
hoelzro 2014.09 docker image is up on registry.hub.docker.com 21:05
21:05 immortal left
wk colomon: seems right discussion 21:05
bartolin carlin: in 2013-06 someone reported a failing test for 1**Inf == 1 on OpenBSD (RT #85750). But with a current version (5.5) the test runs fine. 21:08
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...l?id=85750
21:09 ab5tract joined 21:10 ggoebel111117 joined 21:11 Hor|zon left
bartolin hoelzro++ (I played a tiny bit with the 2014.08 image) 21:12
colomon wk: yeah, here's where the discussion starts
irclog.perlgeek.de/perl6/2010-09-15#i_2829483
hoelzro awesome =)
21:13 kjs_ left 21:15 Ven left
colomon wk: you can really see the language design process at work. :) 21:15
21:22 molaf left 21:26 beastd left 21:28 lichtkind joined 21:30 bartolin left 21:34 BenGoldberg joined 21:35 lichtkind left
wk colomon: took about hour to find right term and minutes to implement it ;9 21:36
21:37 SamuraiJack_ joined
colomon wk: :) 21:43
21:46 SamuraiJack_ left
TimToady now in Reading, PA 21:50
21:52 brrt left
ab5tract TimToady: what brings you to Pennsylvania? 21:53
wk i am a bit lost in docs.perl6.org, where to look for switch (given/when) syntax, for example? 21:56
search does not suggest anything, and browsing did not helped too. 21:58
21:58 leont joined
ab5tract wk: i had luck with google and/or duckduckgo. that syntax has been pretty solid for a while. 22:00
22:02 SamuraiJack_ joined
ab5tract you get to do anything that the smart match operator provides for you, basically 22:04
wk ab5tract: thank you 22:05
actually, i had 2 goal:
1. better understand docs.perl6.org
2. to have sure, how to use succeed/proceed with when 22:06
japhb botsnack 22:14
Ah, saw a channel highlight, thought someone had tried to reach me
japhb scrolls back ...
22:15 SamuraiJack_ left
TimToady ab5tract: taking my son's car to him, after he switched jobs from CA to NJ 22:16
japhb [Coke]: Yes, nqp-jvm and rakudo-jvm can be timed by perl6-bench (this is determined by entries in components.json and compilers.pl) 22:21
[Coke]: However, I myself haven't been timing them recently because the brutal startup time makes *-jvm testing REALLY SLOW
[Coke]: I'm not sure what you mean about the project being "just source" unless you mean "Is there an official awfy clone using perl6-bench?" in which case I'm not aware of any, though timotimo seems to be keeping his up. 22:22
timotimo oh? 22:23
what am i doing?
ab5tract TimToady: Road trip! 22:24
TimToady yup, we've been taking all the scenic routes 22:25
japhb timotimo: Keeping your perl6-bench runs posted 22:26
carlin nine: on OpenBSD a sub with an empty return; hangs. p5_av_top_index should be -1 but is 4294967295, something unsigned that should be signed? 22:27
japhb [Coke]: It looks like two https URLs slipped into components.json; I'll have to fix that. http_proxy shouldn'e be involved at all normally. 22:28
22:29 anaeem1_ joined, anaeem1_ left
ab5tract TimToady: I grew up in southeast PA, which is why I got curious. 22:30
TimToady on my 60th birthday, I hit my 50th state 22:31
obviously I'm missing ten years somewhere...
timotimo japhb: aye, i'm posting them semi-regularly 22:32
i should probably build a small perl6 script that keeps my benchmark files synced and generates a pretty index.html from it
ab5tract which state was your 50th, TimToady? 22:37
TimToady Alabama, on the Natchez Trace
22:39 j4jackj joined, virtualsue left
TimToady hadn't been in OK, LA, MS, or AL before this trip 22:39
or AR
22:39 j4jackj is now known as Guest61593 22:40 zeropool left, bjz left 22:41 Guest61593 is now known as j4jackj
carlin .tell nine on OpenBSD a sub with an empty return; hangs. p5_av_top_index should return -1 with an empty return; but is returning 4294967295. Something somewhere unsigned that should be signed? 22:42
yoleaux carlin: I'll pass your message to nine.
woolfy TimToady: happy birthday, bit belated, but still, hope it was nice! 22:43
[Coke] japhb: I must use an http proxy. 22:44
(at work)
whether it's https or not 22:45
22:49 atroxaper joined
ab5tract TimToady: nice work! That's a lot of travel 22:53
dalek ecs: bbfa780 | (Elizabeth Mattijsen)++ | S32-setting-library/IO.pod:
IO::Path.absolute takes an optional base parameter
22:53 atroxaper left
japhb timotimo: Yes, up to date benchmark index sounds like a very good idea indeed. 22:54
[Coke]: But you can use git+ssh, yes? 22:55
dalek rl6-bench: 6c0b86e | (Geoffrey Broadwell)++ | components.json:
Fix a couple repo URLs that got set to https accidentally
 22:56
japhb timotimo: Is there a p6weekly planned for this week? 22:57
22:58 immortal joined
japhb Yes, belated Happy Birthday, TimToady! :-) 22:58
23:00 erkan left
timotimo oh 23:04
yes, i wanted to do it. then i took a really long nap and went out with friends and forgot all about it >_< 23:05
i'll drive home soon, which will be the point when i'll be able to put a bit of work into that again
23:07 Hor|zon joined
japhb Ah, OK. Excellent. :-) 23:09
23:12 Hor|zon left
lizmat goes to sleep& 23:14
[Coke] f 23:16
ww 23:17
23:18 rjbs- is now known as rjbs
timotimo i haven't even started the research phase for the weekly >_< 23:19
i just wrote a bit of preface text ...
anyway, AFK
23:28 denis_boyun_ left
tony-o got the orm resizing columns automagically 23:28
oops
23:32 firnsy left 23:39 firnsy joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!