»ö« | perl6.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' or rakudo: / pugs: / std: , or /msg p6eval perl6: ... | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by diakopter on 25 January 2010.
TimToady those may not be so simple in p6 00:00
P5 would consider UTF-16 and UTF-32 to be binary, unless they've changed recently
but they're really text
lue well, the spec lists binary as containing odd characters, "such as control codes or bytes with the high bit set" 00:01
TimToady yes, it's rather latin-1-centric
we will not be carrying over the exact behavior 00:02
00:02 athenot left
TimToady would be better to have some kind of file type guesser that has more clue 00:02
lue maybe something like the file command in *nix, or MIME type (I don't trust file extensions)
00:03 cotto_working left, athenot joined
markjreed How are you going to get MIME type from a file in the filesystem? 00:03
lue I do not know. Just throwing ideas.
the hard-coded way would be a database-of-sorts listing all known 'magic numbers' of binary files and such. 00:04
markjreed Something like the algorithm file(1) uses, but that depends on /etc/magic...
arnsholt For file type guessing, it might be possible to train some kind of statistical model? 00:06
00:07 petdance joined
arnsholt Er, encoding, rather than file type 00:07
00:07 petdance left, petdance joined
markjreed web browsers have encoding-guessing algorithms.. 00:09
could port Encode::Guess maybe. 00:10
ash_ does icu have something for this?
00:10 lest_away is now known as lestrrat
ash_ since parrot *can* have icu built into it, if its not there then you'd have to have some sort of fallback 00:10
TimToady replacing icu with a pure P6 library would be a good project at some point 00:11
could be based on the unicode tables like P5 is
markjreed according to fredeaker.blogspot.com/2007/01/char...tion.html, ICU does have an encoding detector.
lue well, let's see how dependant other vital functions (Xorg, KDE, and the like) of the typical system are on icu, and guage our decision on that. 00:14
s/guage/gauge/
I'll be back in a little while, then I'll continue musing with you 00:17
ash_ does parrot have this? 00:18
have you asked the parrot guys about an encoding detection pmc/something? 00:19
00:19 cotto_working joined 00:30 ash_ left 00:32 nihiliad left
lichtkind PerlJam: ping 00:36
00:39 justatheory left 00:48 Chillance left 00:57 snarkyboojum joined 01:04 justatheory joined 01:05 markjreed left 01:07 Ivan4 joined 01:08 Ivan4 left
lue wow. Do I have to be here to continue the conversation? :) 01:15
lichtkind lue: hai 01:16
lue hi
colomon I think "everyone" is getting ready for the big conference / hackathon this weekend, probably. 01:17
lue 7H3R35 G0!/\/G 2 B AY H4K470/\/? \/\/0\/\/! 01:19
lichtkind lue: please calm down
lue AI D0/\/7 7h!/\/k ai k4n! (I'll stop now) 01:20
colomon yes, there's going to be a hackathon.
lue I'll find people to bring testimonials. "Call 1-8TIMTOWTDI now, to donate to our cause!" 01:21
lue .oO(The Larry Wall 45th anual Perl Foundation Hackathon!) #oh, the fun I'll have before I start hacking >:) 01:23
s/anual/annual/
.oO(Wikipedia first ran on Perl code (before switching to the PHP MediaWiki))
01:26
colomon No applicable candidates found to dispatch to for '&infix:<=>' 01:27
that's a new one on me....
lue ACHTUNG!: infix:<=> could not be found. Please try the --things-do-equal-each-other option when you configure next time. 01:29
k23z__ will I get considerable speed boosts if I write some pieces that are now XS in nasm ? 01:34
I wrote a module for generating permutations using XS with the best known algorithm called Steinhaus-Johnson-Trotter and it's still 5 times slower than Algorithm::Permute and this disappoints me of my coding skills 01:35
I was hoping to beat it
now I am thinking about re-writing parts of it in nasm to make it faster
TimToady XS is kinda off topic here
not that we're strict about that... 01:36
arlinius you should write it in perl 6, then optimize rakudo/parrot to make it fast...
lue assembler is the fastest code you could possibly generate (assuming, of course, you're efficient about it)
diakopter .
01:38 yinyin joined
k23z__ ^ 01:39
|
he wrote a dot
diakopter so I did
k23z__ why/
01:39 wknight8111 joined
diakopter std: . 01:39
p6eval std 29935: OUTPUT«===SORRY!===␤Unsupported use of . to concatenate strings; in Perl 6 please use ~ at /tmp/0CIHJKxs7k line 1 (EOF):␤------> .⏏<EOL>␤FAILED 00:01 105m␤»
diakopter rakudo: say 900.00000000000000000 # masakbot 01:42
p6eval rakudo 4edd19: OUTPUT«PAST::Compiler can't compile node of type BigInt␤current instr.: 'perl6;PCT;HLLCompiler;panic' pc 137 (compilers/pct/src/PCT/HLLCompiler.pir:101)␤»
diakopter rakudo: say (90^^4).WHAT # masakbot 01:44
p6eval rakudo 4edd19: OUTPUT«Method 'WHAT' not found for invocant of class 'Undef'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
k23z__ can anyone help me understand what the guy did here ? cpansearch.perl.org/src/ROBIN/Algor...Permute.xs 01:45
what's a multicall ?
is this a concept found in other languages ?
diakopter it appears like he/she wrote an .xs file 01:46
k23z__ or did he invent it ?
01:46 xuzj joined, xuzj left
colomon Tene: errr.... you seem to have broken Rakudo somewhat? 01:48
lue
.oO(At least I didn't break it *phew*)
01:53 mssm left, cls_bsd joined 01:54 ShaneC left 01:55 lichtkind left
lue you guys know of the nazi armbands the soliders wore, right? 01:57
the NAZI soliders
diakopter squints sidelong 01:58
arlinius soliders? 01:59
diakopter rakudo: (sub {say (*)()})()
p6eval rakudo 4edd19: OUTPUT«!whatever_dispatch_helper␤»
diakopter # masakbot ^^ see above; I dunno...
arlinius hrm 02:00
lue Well, my friends and I came up with the Politically Correct and Copyright Infringment War Machines 02:01
diakopter lue plz stay closer to topic..
arlinius rakudo: my $x = *; say $x(0)
diakopter lue
p6eval rakudo 4edd19: OUTPUT«Method 'item' not found for invocant of class 'Integer'␤current instr.: '!whatever_dispatch_helper' pc 280 (src/glue/dispatch.pir:25)␤»
lue diakopter: oh fine, you'll never see the end :) . I just started talking because nothing was happening. 02:02
02:02 PZt joined
lue .rnd(please be advised the rnd bubbles can be ignored at your will (including this one)) 02:06
02:07 hicx174 joined
lue ghost hunters & 02:07
TimToady most topics can be related to Perl 6 if you try hard enough, but you *do* have to try hard enough :) 02:11
from time to time... 02:12
02:12 cotto_working left, nihiliad joined
k23z__ Rate tye SJT dchld A::P 02:16
tye 1.24/s -- -20% -52% -81%
SJT 1.55/s 25% -- -40% -76%
dchld 2.58/s 108% 66% -- -60%
A::P 6.45/s 421% 316% 150% --
this benchmark
does it say that SJT is slower by 76% than A::P ? 02:17
and that in turn A::P is faster by 316% than SJT ?
do I get this right ?
because it doesn't make much sense
diakopter tries not very hard
k23z__ those percents don't make much sense to me
do they make sense to you ?
02:18 rv2733 left
k23z__ do you guys also make benchmarks in Perl6 ? or is it not yet time to make them ? 02:18
diakopter not really
k23z__ 316 - (76/100) * (316) = 75.4 02:20
wasn't I supposed to get 100 ?
I mean it A::P is 316% of SJT 02:21
and SJT is 76 slower
bah nevermind too tired
colomon rakudo: say 316 - (76/100) * (316) 02:22
p6eval rakudo 4edd19: OUTPUT«75.84␤»
TimToady this sort of benchmark discussion is usually more appreciated on, say, perlmonks 02:23
02:23 eternaleye left
k23z__ ok here we discuss Parrot and Rakudo 02:25
diakopter and Perl 6 things
k23z__ ok
I'll go to bed
gnight
diakopter 'nite
02:25 k23z__ left 02:32 meppl left 02:35 eternaleye joined, hercynium joined
TimToady we also, of course, discuss anything *I'm* interested in. :) 02:38
since I'm related to Perl 6 somehow...
02:57 rv2733 joined
colomon TimToady: For series, "When no limit is given, the function need not be monotonic:"" 03:00
I've already coded up a bunch of non-monotonic series with limits for my tests. 03:01
03:01 snarkyboojum left
colomon Is that line a fossil? 03:02
TimToady the monotonic really only applies to intuitied functoins
io
if you have your own generator, matching of limits must be exact
or write your own terminator closure, now that we can have those
so it's kinda fossily 03:03
colomon I don't know of any actually useful examples, but the non-exact limit code is smart enough to handle your own generator too. 03:04
Like { (1..2).pick } ... 1.5 03:05
03:05 wknight8111 left
colomon Unless there's some trick I'm missing, NOT supporting that case with the user-generator will actually be more work than just supporting it. 03:06
TimToady okay, most interesting non-monotonics will be to * anyway
colomon agreed. 03:07
TimToady++ # I'm still really digging the current series spec, even if I'm not quite sure how to efficiently lazily support the arity > 1 cases. :) 03:08
colomon to bed.
TimToady night 03:10
03:36 policomb joined
policomb hi all 03:36
I've a little question
how is it possible to change/set an ethernet device mac address in perl (without using ifconfig) ... ? 03:37
03:38 petdance left
spinclad policomb: in perl 6? 03:41
with the right libraries, and right authorization... 03:42
policomb perl 6 or perl 5, it's the same for me
i would like to know how to do that
i'm root
spinclad in perl 6: you'd have a lot to invent/implement yourself. in perl 5: dunno, search cpan... 03:44
03:44 rv2733 left
spinclad or use pack and ioctl, with no portability i expect 03:46
the replacements for pack in perl 6 are not yet implemented, ioctl likewise 03:48
(classes managing packed buffers, for pack) 03:50
TimToady my impression is that parrot is still very weak on binary data
need compact arrays of ints of various sizes to do it right
spinclad aye 03:53
03:54 petdance joined 03:55 petdance left 03:58 am0c joined 04:00 policomb left 04:12 nihiliad left, athenot left 04:21 athenot joined
lue hello again. Can't wait for the Jerry Lewis Hackathon! 04:26
s/Jerry Lewis/Larry Wall/
04:31 ned left
lue (methinks the hackathon will be a pain for me on the Rakudo side if I kept on pasting patches. Shouldn't be too much a bother though) 04:34
04:38 Trashlord joined
lue --afk & 04:41
04:42 LionMadeOfLions joined 04:49 athenot left 04:54 ned joined 05:10 yinyin left, yinyin joined 05:17 meppl joined 05:46 drasnian joined, drasnian left 05:51 gfx joined 06:04 yinyin left, meppl left 06:05 yinyin joined 06:09 snarkyboojum joined 06:19 snarkyboojum left 06:20 justatheory left 06:46 rgrau left 06:47 yinyin left 06:50 kfo left 06:51 kfo joined 07:02 kaare joined 07:03 kaare is now known as Guest61020 07:07 Su-Shee joined
Su-Shee good morning 07:08
07:08 uniejo joined 07:28 xabbu42 joined 07:32 gfx left 08:02 iblechbot joined 08:16 snarkyboojum joined 08:19 Toony joined 08:23 masak joined
masak good morning, #perl6. 08:23
08:29 flip214 joined
flip214 Hi 08:29
mathw hi 08:32
masak hi!
Su-Shee sees your hi and raises to "hihi!" ;) 08:33
masak wonders if that falls under yesterday's ablaut discussion 08:34
sjohnson hi 08:36
masak [haI] 08:37
sjohnson heh
masak let's do something cool today on #perl6. 08:38
sjohnson that's a good idea 08:40
masak diakopter: I think you've long since filled your quota of 'this number has so many digist/decimals that it causes Rakudo to whimper' rakudobugs :)
rakudo: say (90^^4).WHAT # diakopterbug 08:41
p6eval rakudo 4edd19: OUTPUT«Method 'WHAT' not found for invocant of class 'Undef'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
masak submits rakudobug
pugs: say (90^^4).WHAT
p6eval pugs: OUTPUT«Bool␤»
08:46 flip214 left
masak rakudo: say (0 ^^ 0).WHAT # actually, this is wrong too 08:47
p6eval rakudo 4edd19: OUTPUT«Int()␤»
masak should be Bool().
mathw why? 08:48
masak mathw: S03:1304. 08:49
08:49 mathw sets mode: +o masak
masak "Returns the true argument if there is one (and only one). Returns C<Bool::False> if all arguments are false or if more than one argument is true." 08:49
mathw aaah 08:50
sorry, I was confusing it with infix:<**>
masak I can see how that would be confusing :) 08:52
08:52 am0c left
masak also from S03: 'Not all types can support the concept of infinity. Therefore any value of any type may be compared with C<+Inf> or C<-Inf> values, in which case the infinite value stands for "larger/smaller than any possible value of the type."' 08:54
I'm confused by the 'Therefore...' phrasing.
mathw I'm not sure it's 'Therefore' 08:55
masak oh, good.
mathw 'Therefore' implies that it's obvious
masak yes.
mathw which this isn't particularly, it's just rather handy
masak :)
I'd prefer 'However, "
mathw but it just needs a slight rephrase
hmm no, I don't think that'd work
masak "Nevertheless, " :) 08:56
mathw No, you need to change more than that
The second sentence has other problems
eternaleye s/Therefore/So that comparing finite values to infinite ones functions as most people would expect/? 08:57
masak :)
eternaleye Plus a comma
masak I feel that Damian's reply to the 'long literals' thread essentially settled the matter. 08:59
meeting & 09:01
09:04 k23z__ joined
k23z__ can Perl5 code be run in Perl6 ? 09:07
09:07 am0c joined
k23z__ can modules written in Perl5 be called from Perl6? 09:07
if not, is this in any future plans of development ?
09:07 araujo joined
mathw k23z__: Currently, no and no. jnthn++ has done some work on use of Perl 5 modules. I'm not sure what state that's currently in. It's certainly something highly desirable. 09:16
There is a subset of Perl 5 code which is valid Perl 6, but it's fairly small due to the syntax changes around arrays and hashes and loops and other common things.
09:17 dakkar joined
hejki also there is a bigger subset of ill-written perl5 code which is valid in perl6 :) 09:36
e.g. my @ar; @ar[0] = "foo"; :)
09:37 payload left
mathw But that's wrong Perl 5 09:39
So it's not ill-written, it's just a syntax error 09:40
hejki err..
it does work in perl5 tho
but it's not correct thus making it ill-written imo
mathw It works?? 09:41
Oh wait
array slice
huf it's probaby a warning or something
mathw it probably only works by sheer fluke
hejki er.. all ill-written perl5 begins with: don't { use warnings; use strict; } :) 09:42
09:48 mssm joined 09:50 jnthn_and_mberen joined
jnthn_and_mberen oh hai 09:51
masak: ping
moritz_: ping
phenny: tell pmichaud please see gist.github.com/321595 if you get chance. :-) 09:53
phenny jnthn_and_mberen: I'll pass that on when pmichaud is around.
jnthn_and_mberen phenny: tell masak take a look at gist.github.com/321596 09:55
phenny jnthn_and_mberen: I'll pass that on when masak is around.
jnthn_and_mberen phenny: tell moritz_ take a look at gist.github.com/321596 09:56
phenny jnthn_and_mberen: I'll pass that on when moritz_ is around.
jnthn_and_mberen We'll be back. o/ o/ 09:59
09:59 jnthn_and_mberen left 10:05 k23z__ left 10:16 |newbie| joined
|newbie| hi 10:16
where can I find gimme5?
10:26 orafu left, orafu joined
masak |newbie|: svn.pugscode.org/pugs/src/perl6/ 10:29
phenny masak: 09:55Z <jnthn_and_mberen> tell masak take a look at gist.github.com/321596
masak takes a look
(so, jnthn and mberends have finally merged into a fearsome mechbot?) 10:30
|newbie| hi masak 10:31
masak hi jaffa4
|newbie| jaffa4? 10:32
masak sorry, am I confusing you with someone else?
10:32 am0c left
|newbie| no 10:33
masak |newbie|: irclog.perlgeek.de/perl6/2010-03-02#i_2055372
10:33 |newbie| is now known as jaffa4
hejki a bit offtopic, but still "related": www.patentstorm.us/patents/7028023.html :) 10:34
jaffa4 Could someone tell me how to use gimmi5? 10:35
masak hejki: that's just so depressing. 10:36
hejki almost as depressing as the 'peanut butter sandwich'-patent :P
sjohnson almost as depressing as having this happen to you: images.4chan.org/b/src/1267698521810.png 10:39
masak sees '4chan', chooses not to be curious
sjohnson i drew it myself
so it's safe 10:40
:)
masak what does this Japanese twitterer say about Perl 6? twitter.com/mr_konn/status/9961701457 10:41
something about politics.
jaffa4: I suggest you look at the Makefile in that directory. it uses gimme5. 10:43
jaffa4 never mind, I have figured it out.
masak goodie. 10:44
10:54 stepnem left
jaffa4 masak, how can I downloade the whole svn.pugscode.org/pugs/src? 10:57
10:58 xinming left
masak svn checkout svn.pugscode.org/pugs 10:58
svn checkout svn.pugscode.org/pugs/src, even
hejki s/(heck|ut)//g; 10:59
10:59 xinming joined
masak both 'checkout' and 'co' should work, according to 'svn help'. 11:00
hejki sure, i just see no reason to waster precious bytes by typing the word as a whole :) 11:01
11:02 stepnem joined
masak hejki: obbviously you're not using Ctrl+R in bash nearly enough :P 11:02
hejki what? :P 11:06
11:06 lestrrat is now known as lest_away
hejki i'm more of a history | grep-kinda guy :) 11:07
masak Ctrl+R in bash simply rocks. 11:09
hejki it has it's use cases yes
my favorite bash hotkey is definetly M-_ tho :)
masak that one's nice as well. 11:11
masak tends to use Esc . for that, though
jaffa4 masak, is there an example how to use viv? 11:12
11:13 daemon left
masak jaffa4: there's a USEAGE heredoc in the source. 11:13
hejki masak: well you prolly aren't on a similar keyboard layout either :)
jaffa4 yes, I know about that. 11:14
I am interested in something else.
how to catch actions in viv? 11:15
masak hejki: I run a US layout most of the time. 11:16
jaffa4: I don't understand the question.
jaffa4 is there a way to hook actions in viv?
or walkthrough the datastructure of the syntax tree using a perl program? 11:17
masak the latter is definitely possible. 11:18
just pipe it into a perl program.
jaffa4 without processing 11:19
the output
masak er.
That's like asking "Is there a way to not be so thirsty? Without drinking." 11:20
11:21 Gothmog_ left 11:23 _Jedai_ left
jaffa4 look 11:25
here it is 11:26
$r = STD->parse($PROG, actions => 'Actions')->{'_ast'};
the whole ast will be in $r.
no need to parse the output!
masak ok, so there you go then.
jaffa4 line 91 in viv.pl
11:28 lest_away is now known as lestrrat
masak jaffa4: good luck, whatever it is you're attempting. it sounds interesting. 11:28
lunch &
hejki masak: ok. i'm on FI layout all the time :) 11:30
11:30 am0c joined 11:38 Gothmog_ joined, stepnem left 11:40 f00li5h[HireMe] left, f00li5h[HireMe] joined 11:41 stepnem joined 11:42 k23z__ joined, k23z__ left, k23z__ joined
k23z__ does p6 have a way of looping through an array taking each consecutive pair ? 11:43
like
suppose the array were @a
and I want $f->($a[$i],$a[$i+1]) for $i (0..@a-2) 11:44
does p6 provide some sugar for that ?
hejki for @ar -> $x, $y { ... } ?
k23z__ ah yes forgot
thanks
hejki wait
k23z__ I used it a couple of months ago I think
yes ?
hejki it doesn't work exactly like you want it to
rakudo: for (1..10) -> $a, $b { say "a: $a b: $b"; } 11:45
p6eval rakudo 4edd19: OUTPUT«a: 1 b: 2␤a: 3 b: 4␤a: 5 b: 6␤a: 7 b: 8␤a: 9 b: 10␤»
k23z__ hmm ..
hejki i thought you wanted: 1,2 2,3 3,4 4,5 5,6
k23z__ yes
hejki you could use Z or X with 2 arrays
k23z__ I could shift one of the arrays 1 position forward 11:46
how do I shift an array with 1 position forward ?
hejki rakudo: for (1..10) X (2..11) -> $a, $b { say "a: $a b: $b"; } # might fail :)
k23z__ I mean having (1,2,3) -> (bla,1,2,3)
p6eval rakudo 4edd19: OUTPUT«Missing block at line 11, near "X (2..11) "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
k23z__ hejki: yeah like that
but ...
it's cartesian product
it's more than I need
hejki yes 11:47
rakudo: for 1..10 Z (2..11) -> $a, $b { say "a: $a b: $b"; }
p6eval rakudo 4edd19: OUTPUT«a: 1 b: 2␤a: 2 b: 3␤a: 3 b: 4␤a: 4 b: 5␤a: 5 b: 6␤a: 6 b: 7␤a: 7 b: 8␤a: 8 b: 9␤a: 9 b: 10␤a: 10 b: 11␤»
hejki voilá
k23z__ yeah that looks great 11:49
hejki: now we just need to generate (2..11) from (1..10)
betterworld alpha: my @a = 1..10; for @a Z (@a >>+>> 1) -> $a, $b { say "a: $a b: $b"; }
p6eval alpha 30e0ed: OUTPUT«a: 1 b: 2␤a: 2 b: 3␤a: 3 b: 4␤a: 4 b: 5␤a: 5 b: 6␤a: 6 b: 7␤a: 7 b: 8␤a: 8 b: 9␤a: 9 b: 10␤a: 10 b: 11␤»
k23z__ betterworld: excellent
thank you 11:50
I didn't know p6 had these operators but it's very nice it does
11:50 snarkyboojum left
hejki hmm.. 11:50
yeh
11:52 jaffa4 left
betterworld or maybe @a Z @a[1..*] so it works with other arrays too 11:55
hejki even better :) 11:57
and looks better as well :>
12:08 korpenkraxar joined 12:11 lichtkind joined
lichtkind moritz_: got my msg? 12:12
masak: moin 12:13
12:13 rv2733 joined
masak lichtkind: hej, hej 12:15
lichtkind :)
masak: currently working on www.perlfoundation.org/perl6/index.cgi?timeline
masak nice. 12:16
lichtkind: guess you saw my comments on that page from yesterday?
lichtkind: "lasted until 2006 August 23" -- maybe make this a separate timeline item instead? makes it easier to see where it fits into the history. 12:18
lichtkind yes and yes 12:19
i thought about that
masak lichtkind: "2010 April Rakudo star release" doesn't fit into the list of "what happened so far".
lichtkind: also, you misspelled "happenEd".
lichtkind soon it will
masak lichtkind: you're missing the point.
it hasn't happened so far. 12:20
thus it doesn't belong on the list.
it belongs on an other list, "things which haven't happened so far".
12:20 daemon joined
masak along with a lot of other important things :) 12:20
hejki imagine slacking like that in a corporate job :) 12:21
k23z__ masak: is there a plan for SDL or OpenGL bindings along the way ?
12:21 ruoso joined
masak k23z__: don't think so. please go right ahead and start a project. 12:22
12:31 athenot joined
lichtkind masak: bah www.perlfoundation.org/perl6/index.cgi?timeline 12:32
masak lichtkind: looks better and better :) 12:33
lichtkind masak: i added the hague grant too
masak 'and other stuff' -- questionable phrasing.
it's too informal and not at all informative.
'for, among other things, Perl 6' would work better, IMHO. 12:34
12:42 petdance joined 12:53 athenot left 12:56 nacho joined 13:00 petdance left 13:04 k23z__ left
takadonet morning 13:04
masak takadonet: \o 13:09
takadonet masak: how are you doing?
masak takadonet: pretty well. you?
takadonet masak: busy but good 13:10
13:10 payload joined
masak takadonet: yes, me too. presentation in two days. :) 13:12
13:12 nacho left
takadonet where and what is your presentation? 13:12
masak www.opensourcedays.org/2010/node/267 13:13
takadonet nice
masak thanks -- really looking forward to the conference, the hackathon, and everyone's talks. 13:14
takadonet will your presentation be online after the conference? 13:15
masak online, cc-licenced, possibly even with a video (eventually).
m-locks aye we'd like to see ;) 13:16
masak that makes me happy, too :)
13:22 athenot joined
Tene colomon: explain what I broke? 13:23
colomon Tene: 20-30 tests on my machine.
including the make test try.t
reverting 4edd19dbaaee2dd2c61cbc1044ef39651d45f3e4 made everything work again 13:25
13:37 riffraff joined
Tene colomon: I see you didn't push a revert commit, yes? 13:54
colomon nope.
just reverted locally
(well, I think I pushed one in the series branch, but that's a side thing) 13:55
I was hoping it was something easy for you to fix. :)
Tene okay, just checking. The problem is that I'm not returning the value of the try block in cases where the block doesn't fail. 13:56
Looks like it's a bug with the PAST compiler, not setting the return value of a Stmts node properly when there's a handler on it. 14:06
PerlJam good morning #perl6 14:07
Tene oh, and looks like it misbehaves badly in the case where it's handled 14:08
takadonet morning PerlJam 14:09
Tene Let's try this...
oh, no, that won't work. 14:10
14:13 Salada345 left
Tene colomon: feel free to push a revert commit. I'll try to get back to this later today. 14:14
eh, nm, I'll revert it myself. 14:15
pushed 14:16
dalek kudo: 4055fde | tene++ | src/Perl6/Actions.pm:
Revert "Re-work try {} to ignore control exceptions"
m-locks a new one? 14:17
Tene what is a new what?
14:19 nacho joined 14:26 nacho left 14:28 stepnem left
m-locks a new review :) 14:34
14:40 nacho joined
masak the use.perl.org home page looks depressingly desolate. 14:43
the 12th German Perl Workshop in June is announced there, but not this weekend's Hackathon, or NPW in Iceland, or YAPC::EU in Pisa this summer... 14:44
PerlJam masak: submit those stories then! 14:46
masak what, and lose the opportunity to complain? :P 14:49
you're right, of course.
I guess this situation occurs, though, because no-one really cares about use.perl any more, and we should all see about giving it a humane final rest. 14:50
it used to be that Parrot and Rakudo releases were announced there too, but not anymore.
Su-Shee I've never used it and rarely even looked at it. 14:52
14:53 Jedai joined
masak Su-Shee: ...except when you assiduously read pmichaud's, jnthn's and my blog posts, right? :) 14:57
std: .say if .so 14:59
p6eval std 29935: OUTPUT«ok 00:01 106m␤»
Su-Shee masak: I don't read them over use.perl.org.
masak Su-Shee: oh, good. because frankly, the CSS is terrible over there.
15:00 SmokeMachine joined
Su-Shee "what css?" ;) 15:00
masak 'xactly.
Su-Shee probably, one day a perl (6?) forum is needed.
masak maybe a combined forum-and-blog-engine? 15:01
Su-Shee no. let blogs be blogs and forums be forums.
masak fair enuf. 15:02
Su-Shee though most contemporary forums have a feed.
(and even fvwm (!) has a forum. ;)
mathw fvwm still exists?? 15:04
masak Perl 6 isn't dead?? 15:05
mathw I guess I'm not really all that surprised
I'm just bored 15:06
15:06 Guest61020 left
Tene man, I've never been able to get into forums. 15:06
masak neither 15:07
Tene IRC and mailing lists are great, but I feel awkward trying to squeeze anything meaningful and interactive into a web browser.
I don't really blog, either, though... so I'm a bit unusual.
masak some people are crazy about them, it seems. different tools for different people, I guess.
Tene nodnod
mathw I find forums work for some things, and not for others 15:08
I use them, but only for certain things
A lot of it's down to the people rather than the subject matter
frettled and the problem that Yet Another Forum means yet another sign-up procedure, another set of credentials, another set of preferences, and another web page to load :) 15:09
15:09 uniejo left
masak *sigh* 15:11
hopefully Web 3.0 will have some concept of a web identity, so one doesn't have to create new accounts all the time.
mathw OpenID, then 15:12
masak yeah.
15:14 nihiliad joined
mathw You're all intelligent people, so maybe some of you know about good resources for multithreaded system design and programming techniques in C++? 15:16
15:17 colomon_ joined, colomon left, colomon_ is now known as colomon 15:23 masak left
vorner mathw: Actually, I say that C++ is extremely unfriendly to multithreaded programming. I use haskell if possible for such things. 15:25
15:28 IllvilJa left 15:32 masak joined 15:34 masak left
Su-Shee mathw: of course it does. and it has features very few window managers have. still a bitch to config, though. 15:35
Tene which is this, Su-Shee? 15:36
Su-Shee I HATE web identities. none of them work really properly or fast and hello? I don't want to spread my identity all over everywhere.
15:37 nihiliad left
Su-Shee Tene: native svg support for elements of the wm which actually can be scaled for example. 15:37
15:38 Psyche^ joined
mathw vorner: yes, I know it is but I don't have a choice of language here 15:39
15:39 Patterner left, Psyche^ is now known as Patterner, nihiliad joined
vorner mathw: I did some small things using pthreads only, usually by trying to separate the things as much as possible 15:40
15:41 korpenkraxar left 15:42 bronco_creek joined
mathw Yeah, we're a bit beyond small things with this 15:47
I've never been able to find anything seriously good on large-scale architecture of multithreaded systems in C++
maybe nobody knows how to do it well
m-locks some stack reconstruction approaches have been researched 16:03
bronco_creek Hi. This is my first time here. I'm a has-been/wanna-be perl programmer. I just got around to installing parrot & rakudo and starting to play around. I may be seeing a bug. What's my next step? 16:04
m-locks you try telling whats wrong 16:05
Tene bronco_creek: it's pretty likely that you've found a bug. You'll want to produce as minimal a test case as you can, and then submit it to the bug tracker.
rakudobug?
Oh, right, no purl in here.
bronco_creek Where should I look to see if it is already reported?
Tene [email@hidden.address]
Hmm.
bronco_creek Thanks. 16:06
16:08 masak joined
pugssvn r29936 | lwall++ | [file-tests.t] fix parsefail 16:18
16:20 kfo left, k23z__ joined, k23z__ left, k23z__ joined
masak bronco_creek: we have a rakudo evalbot on the channel. reproducing the bug with this often produces some useful discussion which you can use in the bug report. 16:22
rakudo: say (1 ^^ 2).WHAT
p6eval rakudo 4055fd: OUTPUT«Method 'WHAT' not found for invocant of class 'Undef'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
masak like that. :)
16:23 kfo joined
bronco_creek What I'm seeing involves instantiating an object. Can the evalbot handle multiple lines of code? 16:24
m-locks nope 16:25
TimToady yes, it can 16:26
m-locks whoa
masak it can handle multiple statements.
just separate them by semicolons.
TimToady you can feed it ␤ characters that translate into newlines
masak that, too.
but golfing to a minimal program size helps both bots and people :)
m-locks but it is all single line input
bronco_creek Thanks, I'll give it a try. 16:27
16:27 uniejo joined
TimToady to clarify, the evalbot's irc interface is single line, but it will do translation both coming and going of ␤ characters 16:28
so the program it is running has a multi-line interface
m-locks ;) 16:29
TimToady rakudo: { say 'I am a bare block'}␤{ say 'and so am I' }
p6eval rakudo 4055fd: OUTPUT«I am a bare block␤and so am I␤»
bronco_creek So, how do I type the "HL" characters? 16:30
TimToady it's NL
depends on your windowing system, likely
but you can always copy/paste from evalbot output
bronco_creek I'm using the IRC web client on a Windows system. 16:31
masak TimToady: did you see the confusion in the backlog about the 'Therefore' on S03:1340? any opinions?
bronco_creek: copy/paste?
bronco_creek OK.
masak that's what I do. I'm on Emacs, so copying is a no-op. 16:32
bronco_creek rakudo: class Test1 { has $.a; has @.b; has %.c; } my $tester = Test1.new(a => 'test'); say $tester.a;
p6eval rakudo 4055fd: OUTPUT«Confused at line 11, near "class Test"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
TimToady "Most types don't have Inf, therefore we will supply one"
masak bronco_creek: need a semicolon after the block.
bronco_creek OK 16:33
masak TimToady: if it really said that, I'd be less confused.
TimToady or that ␤ character
but semicolon is easier to type :)
bronco_creek rakudo: class Test1 { has $.a; has @.b; has %.c; }; my $tester = Test1.new(a => 'test'); say $tester.a;
p6eval rakudo 4055fd: OUTPUT«test␤» 16:34
masak but it says 'Therefore any value of any type may be compared...' -- which is not a consequence of not all types having Inf.
16:34 abra joined
TimToady it is a pragmatic connector, much like "hopefully" is not saying that the thing you're talking about is happening with hopefulness 16:35
masak ok. it's not vital that it be changed. I just found it unclear when I saw it.
TimToady Hopefully the technical point is clear. :) 16:36
bronco_creek Hmm. Must be something wrong with my installation. On my system, any class with has %.x ; causes errors.
japhb masak: Someone was asking about SDL and OpenGL bindings earlier? In what context? Rakudo?
masak japhb: Perl 6 in general, I think. but probably implicitly Rakudo, yes. 16:37
TimToady: :) 16:38
japhb OpenGL already works in Rakudo (or rather, it did, back in the old branch, I haven't tested it with new master). SDL bindings for Parrot are under development again (now by kthakore++), so that should be coming soonish too. 16:39
16:39 uniejo left
japhb Come to think of it, I guess now is the perfect time to rebuild and try it. ;-) 16:40
m-locks do eeeet
masak japhb: cool! that's good news.
a quick Google on Language Log shows that someone has proposed 'hopeably' as the statement-modifying adverb :) 16:41
16:42 vorner left 16:43 cotto_work joined
bronco_creek Thanks all. Your reputation for helping newbies is well deserved. Hope I can contribute something more useful in the future. 16:46
Now, how do I quit the IRC?
Tene bronco_creek: /quit
16:46 bronco_creek left, betterworld left 16:47 SmokeMachine left, jnthn_mberends_b joined
jnthn_mberends_b oh hai 16:47
the mech bot returns
masak: ping
(btw, our wifi is HTTP only at the moment, so can't connect properly to IRC :-/) 16:48
masak jnthn_mberends_mechbot: pong 16:49
jnthn_mberends_b masak: hi
masak: did you read our stuffs?
masak \o/ hi!
I did indeed. 16:50
jnthn_mberends_b ok
masak you stopped right before the good parts! :)
(i.e. the ones I'm interested in)
jnthn_mberends_b thoughts on the talk?
masak: add them to the agenda then! :-)
masak oh, is it a talk?
I think I'm missing a bit of context here.
jnthn_mberends_b masak: no, did yuou see the gist about us trying to do something as a group in pm's talk slot? 16:51
masak no, but I've been toying with the same idea myself.
wondering how to pitch it. :)
jnthn_mberends_b gist.github.com/321595
masak reads
Plan A sounds like what I thought of doing. Plan B sounds like fun too. I'm up for either. :) 16:52
Plan C sounds a bit risky :P
jnthn_mberends_b Well, we have no input towards Plan A so far. 16:53
And we're kinda tight on time...
masak you don't think we can base it on earlier similar pmichaud talks?
jnthn_mberends_b such as?
masak searches
jnthn_mberends_b wonders whether there's been more recent Rakudo update talks than YAPC::Europe. 16:54
16:54 pmurias joined
masak p'haps not. 16:54
Tene let's how you don't end up implementing Plan Omega: Kill All Humans
masak right. the YAPC::EU slides are here. www.pmichaud.com/2009/pres/yapceu-r...start.html 16:55
going through them now to see how much is current and salvageable. 16:56
jnthn_mberends_b Welll, they pre-date nqp-rx and ng.
masak sure. I'm just thinking whether that impacts the message greatly. 16:57
ooh, pmichaud++ used my show-of-hands gimmick! :)
I had forgotten about that.
jnthn_mberends_b www.pmichaud.com/2009/pres/ seems to not have more recent than the YAPC::EU one on Rakudo. 16:58
masak the cards example is teh awesome.
if we do any of the Plans, we must do that.
16:59 abra left
TimToady in the phone call yesterday pmichaud++ suggested a panel discussion 17:00
masak I can see that working, too.
jnthn_mberends_b ooh, interesting
masak hm, www.pmichaud.com/2009/pres/yapceu-r...ide85.html is the first really out-of-date slide.
jnthn_mberends_b As in questions from the floor? Or?
TimToady he didn't give any details 17:01
just noted that four of the leading lights would be there :)
and four is a good number for a panel
masak seems there's a real danger of questions drying up at minute 14... or a meatspace flamewar breaking out... :) 17:02
where do I sign up? :)
TimToady hopefully among the panel :)
masak I'd pay to see that!
the real problem we're facing is that there's not really any time to rehearse anything.
from that viewpoint, panel discussion sounds like a good option. 17:03
jnthn_mberends_b yes, it's lwo on prep
masak: were you at the nordic perl workshop last year, when the go open panel sessions happened? 17:04
masak aye.
jnthn_mberends_b (Bruce Perens, Larry, Jon Hall)
masak sat between half of the jnthn-mberends mech bot.
jnthn_mberends_b iirc that worked as first there was somebody asking the panel questions to get it seded
japhb just tried OpenGL examples under current Rakudo
jnthn_mberends_b And then it was handed over to the audience
TimToady Maddog drank ten beers during that 17:05
japhb Looks like 'use OpenGL:from<parrot>;' no workie.
jnthn_mberends_b *seeded
japhb Is that syntax changed, or just nyi?
jnthn_mberends_b TimToady: ooh.... :-D
Tene japhb: it *used* to work.
TimToady had to run off to pee in the middle...
Tene I expect it didn't get put back in.
I can work on that today.
17:05 uniejo joined
masak jnthn_mberends_b: I think I'm onto which half of you is at the keyboard :P 17:05
17:05 SmokeMachine joined
jnthn_mberends_b how'd you guess? :-P 17:06
japhb Tene, OK, thanks
masak reads the last line of www.pmichaud.com/2009/pres/yapceu-r...ide93.html and snickers 17:07
jnthn_mberends_b lol
japhb "Magic unicorns may appear." ... watches white 4-hooved animal with conspicuous spiral horn trot by ... "Right. Magic unicorns *will* appear." 17:08
jnthn_mberends_b In light of lack of prep time etc, the panel session seems like a good idea.
We're up for it.
TimToady and you all have different cool accents 17:09
masak vee're yp for it.
jnthn_mberends_b OK, we haz a plan. 17:10
\o/
And pmichaud++ for suggesting it.
And TimToady++ for mentioning.
masak masak: cool Nordic accent. mberends: cool Dutch accent. moritz: cool German accent. jnthn: cool Lolcat accent.
TimToady shouldn't that be \o/\o/?
jnthn_mberends_b the mechbot junction collapsed 17:11
Tene people actually prepare and rehearse talks? Huh, I've been doing it wrong all this time.
masak Tene: they've threatened to record me on film! that sure got me working. 17:12
jnthn_mberends_b on your accent?
masak jeees.
jnthn_mberends_b lollol
masak I'm also thinking how I might integrate the "Lifts produced by Schindler" pun into my talk... and then send the video to Tene... :) 17:13
Tene :P
masak: I'm sure you'll get plenty of other puns in. 17:14
masak some of which are even known to work on real humans!
I pass a bit more of the Turing test each day. 17:15
I mean... of course I'm human. *cough*
Tene I've accidentally failed the turing test on IRC several times before. 17:16
TimToady so how come lollol didn't collapse too?
jnthn_mberends_b it's a one junction
masak Do you think that I've accidentally failed the turing test on IRC several times before is the reason you're here, Tene?
17:17 SmokeMachine left
masak jnthn_mberends_b: 1o11o1 <--- one junction 17:17
TimToady 101 17:18
masak The Matrix was clearly before its time.
TimToady where are those backwards kana in Unicode, I wonder... 17:19
Tene TimToady: they were reversed because they were *outside* the matrix looking *in*.
the real world is little-endian. 17:20
TimToady oh, I thought maybe they were hebrew kana or some such...
jnthn_mberends_b OK, if anyone has question suggestions for the panel discussion, prefix them as panelbot: and then we can grep 'em out of the infrared clogs
colomon proposes the Higgins Test, where we see if a person can successfully fool another person into thinking he is Eliza.
masak whoa. Hebrew kana...
colomon++ # :D
TimToady colomon: don't you mean the 'iggins Test, guv'vor? 17:21
colomon TimToady++ 17:22
jnthn_mberends_b panelbot: Is the Rakudo grammar expected to converge with STD? # for example 17:26
We could also have questions taken from IRC too. ;-)
masak that'd be cool.
jnthn_mberends_b Just log someone in on the web irc client as panelbot ;-) 17:27
TimToady if you can can have it up on the screen, pm and I can arrange to be on irc then 17:28
we mentioned that in the phonecall too 17:29
Tene panelbot: Which line of STD.pm is your favorite?
jnthn_mberends_b can arrange that :-)
17:34 mberends_jnthn_b joined
TimToady oh noes 17:35
mberends_jnthn_b gah, ran out of web minutes
17:35 justatheory joined 17:36 jnthn_mberends_b left
TimToady you're in a maze of twisty little irc nicks, all different... 17:36
maybe you should just use each web connection for one IP packet, then run TCP on top of that... 17:37
mberends_jnthn_b We has limited minutes, so we'll log off now...we'll be back around 11 CET. 17:38
TimToady 11 p.m.?
mberends_jnthn_b TimToady: yes
:-)
o/ o/ 17:39
TimToady \o xx *
17:39 mberends_jnthn_b left
lichtkind :) 17:39
TimToady: www.perlfoundation.org/perl6/index.cgi?timeline ist so allright?
17:43 masak left 17:47 stepnem joined 17:52 jaldhar left
diakopter I keep missing masak 17:58
colomon btw (as I see you were involved) I'm pretty sure 1...^* is getting parsed as 1....(^*) 17:59
rakudo: say (^*).batch(10)
p6eval rakudo 4055fd: OUTPUT«Method 'batch' not found for invocant of class 'Range'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
colomon rakudo: say (^*).iterator.batch(10) 18:00
p6eval rakudo 4055fd: OUTPUT«␤»
colomon :(
18:00 dakkar left
colomon rakudo: say (0..^*).iterator.batch(10) 18:02
p6eval rakudo 4055fd: OUTPUT«␤»
colomon rakudo: say (0..10).iterator.batch(10)
p6eval rakudo 4055fd: OUTPUT«0123456789␤»
colomon interesting...
diakopter colomon: you were talking to me? 18:03
colomon diakopter: more or less. :)
diakopter oh :)
colomon you reported that bug to massk, no?
diakopter yeah
'cept, as usual, I wasn't certain it was a bug 18:04
colomon well, for sure it's a feature which isn't implemented yet.
but it may be showing a bug in the existing features, too. ;) 18:05
18:05 athenot left
colomon What I don't know is if it is being treated as 1...0, 1, 2, 3, etc., or 1...(scalar range 0..^*) 18:06
diakopter ah
oh
rakudo: say (0..{^*}).iterator.batch(10) 18:07
p6eval rakudo 4055fd: OUTPUT«0123456789␤»
colomon for sure my examples above indicate some kind of range bug has snuck into the code.
diakopter rakudo: say (0..^{*}).iterator.batch(10)
p6eval rakudo 4055fd: OUTPUT«0123456789␤»
lichtkind www.perlfoundation.org/perl6/index.cgi?ofun 18:09
rewrote that a bit
any complains?
18:10 Toony left 18:14 ShaneC joined
arnsholt lichtkind: "Pugs was allegedly developed this way and grow thatswhy in many directions." looks like a mix of an old and a new sentence 18:30
lichtkind arnsholt: it is, your free to optimze :)
im no native speaker 18:31
18:33 _newbie_ joined
arnsholt I'll have a look at the edit log and try to figure out what you intended to say =) 18:33
18:35 payload left 18:37 jonasbn joined
jonasbn good evening all 18:37
colomon o/ 18:41
18:42 IllvilJa joined 18:46 quietfanatic left 18:52 justatheory left 18:53 justatheory joined, cosimo left 18:54 cosimo joined 18:55 prism joined, daemon left, prism is now known as daemon 19:04 jaldhar joined 19:05 jaldhar left, jaldhar joined 19:07 athenot_ joined 19:08 athenot joined, athenot left, athenot_ is now known as athenot 19:13 payload joined, jmarti joined, _jaldhar joined 19:16 jaldhar left 19:20 flavioglock joined, Chillance joined
flavioglock hi 19:21
colomon \o 19:23
19:25 sbp left 19:29 takadonet left, rv2733 left 19:32 flavioglock left 19:34 fglock joined
diakopter fglock: how goes it 19:36
or should I say, how "go"es it
fglock diakopter: all is fine :) 19:37
19:37 jaldhar_ joined, _jaldhar left
fglock Perlito now has an interpreter written in plain mp6, so eval() is now possible 19:38
diakopter oo
fglock doesn't work in Go yet, needs some debugging
today I've added t/20-eval.t - it passes all tests in the Perl5 backend; I'm now debugging the Javascript backend 19:40
diakopter does it work in JS?
oh
19:40 ashleydev joined
fglock js is missing "exists" 19:40
(looking up in google for hints how to implement it) 19:41
diakopter I can help
wait, which "exists" 19:42
hash key existence?
fglock yes 19:43
diakopter jsobj.hasOwnProperty(key) tells you if the object has it defined on itself
fglock thanks :) - testing...
diakopter typeof(jsobj[key])!='undefined' tells you if the object has it defined anywhere on its prototype chain
19:45 ashleydev left, ashleydev joined
fglock hasOwnProperty works, testing eval again... 19:55
19:59 ive joined 20:01 nsh_ is now known as nsh, nsh left, nsh joined, nsh is now known as nsh-, nsh- is now known as nsh 20:05 riffraff left 20:10 payload left 20:22 buubot left 20:23 nsh is now known as nsh_, nsh_ is now known as nsh 20:38 payload joined 20:40 k23z__ left
pmichaud good afternoon, #perl6 20:42
phenny pmichaud: 09:53Z <jnthn_and_mberen> tell pmichaud please see gist.github.com/321595 if you get chance. :-)
pmichaud phenny: tell jnthn I will put together an outline of what I intended for the talk; may either be just an outline or may be slides. 20:43
phenny pmichaud: I'll pass that on when jnthn is around.
pmichaud phenny: tell jnthn I should have the outline a bit later tonight (about 6 hrs from now, or 03h00 UTC) 20:44
phenny pmichaud: I'll pass that on when jnthn is around.
frettled pmichaud: by golly, you're right, it's after noon :) (hi) 20:48
pmichaud: how are you doing, recovered from snowy hell yet?
pmichaud snowy hell? 20:49
frettled Didn't you have some rather bothersome snowy days a while ago?
pmichaud oh, a couple of weeks ago, yes.
that snow's all gone :) 20:50
frettled \o/
20:51 ashleydev left 20:59 meppl joined
colomon we've still got snow here. :) 21:06
m-locks :) 21:08
frettled some streets are bare, here 21:09
salt helps a lot :)
21:09 frettled sets mode: +o colomon 21:13 jaldhar_ left, nacho left 21:17 SmokeMachine joined 21:23 payload left 21:26 buubot joined, k23z__ joined, k23z__ left, k23z__ joined 21:27 buu joined 21:28 justatheory left 21:30 justatheory joined
colomon frettled: our streets are bare, it's actually pretty warm today. 21:30
21:31 ruoso left
colomon but there was a lot of snow of the ground, so where it hasn't been cleared at all, it's still pretty deep. 21:31
frettled it was pretty warm here, too, sun shining for three or four days straight, maybe two or three above freezing 21:33
I'm glad I don't live in inland Canada :) 21:35
21:35 _newbie_ left
colomon for sure I'd prefer Newfoundland to most of it... :) 21:35
and my winter trips to Toronto always seem to involve blizzards. 21:36
21:36 ive left 21:38 uniejo left 21:45 athenot left
pmurias fglock: hi! 21:45
21:51 am0c left 21:53 fglock left
dalek kudo: 37e5746 | pmichaud++ | docs/release_guide.pod:
Update release_guide.pod with recent release information.
21:58
22:02 jonasbn left 22:12 k23z__ left 22:13 SmokeMachine left 22:14 k23z__ joined, k23z__ left, k23z__ joined 22:15 k23z__ left, masak joined
masak <diakopter> I keep missing masak 22:16
22:16 k23z__ joined, k23z__ left, k23z__ joined
masak diakopter: awww! I missed you too! :) 22:16
22:16 frettled sets mode: +o masak
masak nom nom nom @ 22:16
22:16 k23z__ left
masak in www.perlfoundation.org/perl6/index.cgi?ofun 22:17
'-O is the command line option for compiler'
should be '-O is A command line option for compilerS' 22:18
(hm, perhaps better written '-O is *a* command line option for compiler*s*')
lichtkind: ^
lichtkind ive seen it :) 22:19
masak s/we human/us humans/
s/our doing/our actions/
frettled but that's no fun ;)
masak frettled: no-one said grammatical correctness was going to be fun :) 22:20
lichtkind frettled: whats no fun?
22:20 dennisroberts joined
lichtkind ah 22:20
masak frettled: it's *about* fun, not fun per se. :P
lichtkind change it later
im officially i my bath
masak I can see how that might complicate wiki editing. 22:21
22:21 dennisroberts left 22:22 jmarti left
lichtkind masak: so feel free to edit yourself :) 22:24
frettled masak: writing about fun should be fun :) that's why they invented fun-roll-loops 22:25
masak lichtkind: thanks for the offer... I think I'll leave it in your competent hands. still working on my slides over here.
masak is less stressed-up about the slides than usual
frettled: I remember the Talk page about Discordianism on Wikipedia at some point. 22:26
frettled masak: that must have been fun :)
masak frettled: it was all about convincing stubborn people not to write the article in a Discordian tone of voice.
masak heads over there to see if that discussion is still there
frettled too 22:27
"This article needs additional citations for verification." - bwahahahaha
22:29 Su-Shee left
diakopter masak: caught! 22:29
masak diakopter: hey there. :) 22:30
diakopter fuzz-stress time
masak charges up his bionic arms 22:31
I might not be here for long. 22:32
forewarned is bionic-armed.
(need to regenerate before tomorrow's big journey)
frettled: it must have been this discussion page I was thinking about: en.wikipedia.org/wiki/Talk:Fnord 22:34
std: [][] 22:35
p6eval std 29936: OUTPUT«ok 00:01 106m␤»
masak rakudo: [][]
p6eval rakudo 37e574: OUTPUT«Confused at line 11, near "[][]"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
masak is it just something unimplemented, or worthy of a bug report?
alpha: [][]
p6eval alpha 30e0ed: ( no output )
diakopter berg 22:36
masak it regressed, so I'm reporting it.
masak submits rakudobug
diakopter: berg? as in salty mountain?
diakopter rakudoberg
masak oh, it's a distortion of 'bug'. I see.
well, there's been a berg of bugs lately, that's for sure. 22:37
alpha: my $a = 0; $a++++; say $a 22:43
p6eval alpha 30e0ed: OUTPUT«1␤»
masak rakudo: my $a = 0; $a++++
p6eval rakudo 37e574: OUTPUT«Cannot assign to readonly value␤current instr.: '&die' pc 16799 (src/builtins/Junction.pir:347)␤»
diakopter that seems more proper
masak does it? got spec to back that up?
I always thought postfix:<++> produced lvalues. 22:44
diakopter seeing as how $a++ should return a copy of the value
oh
masak or is that only prefix:<++>?
what does Perl 5 do?
"Can't modify postincrement (++) in postincrement (++)"
diakopter Can't modify postincrement (++) in postincrement (++) at -e line 1, near "++++" 22:45
masak excellent error message. :) funny, and to the point.
diakopter yes, that
22:45 wknight8111 joined
masak (incidentally funny, that is. I'm sure it's a specific case of a more general error message) 22:45
ok, so I won't report Rakudo for that, then.
jnthn o/
phenny jnthn: 20:43Z <pmichaud> tell jnthn I will put together an outline of what I intended for the talk; may either be just an outline or may be slides.
jnthn: 20:44Z <pmichaud> tell jnthn I should have the outline a bit later tonight (about 6 hrs from now, or 03h00 UTC)
masak rakudo: my $a = 0; ++$a++; say $a 22:46
p6eval rakudo 37e574: OUTPUT«2␤»
masak \o/
diakopter rakudo: say 1%%1 22:47
p6eval rakudo 37e574: OUTPUT«Method 'postcircumfix:<[ ]>' not found for invocant of class ''␤current instr.: '!postcircumfix:<[ ]>' pc 10568 (src/builtins/Role.pir:90)␤»
masak ooh!
masak submits rakudobug
diakopter wins
jnthn phenny: tell pmichaud I think we'll go with pannel session rather than a replacement talk.
phenny jnthn: I'll pass that on when pmichaud is around.
masak diakopter: not so. it's 1-1 so far.
diakopter oh
diakopter catches up
std: say 1%^^1 22:48
p6eval std 29936: OUTPUT«===SORRY!===␤Preceding context expects a term, but found infix ^^ instead at /tmp/Yminm8Zt2F line 1:␤------> say 1%^^⏏1␤FAILED 00:01 105m␤»
diakopter rakudo: say 1%^^1
p6eval rakudo 37e574: OUTPUT«1␤»
masak submits [WEIRD] rakudobug 22:49
diakopter heh
that counts as 0.5
masak TimToady still hasn't acknowledged the 'my @stuff[ %100_000 ]' STD.pm-bug I found a few days back.
diakopter I didn't see it 22:50
masak std: my @stuff[ %100_000 ]
diakopter maybe you were netsplit
p6eval std 29936: OUTPUT«===SORRY!===␤Whitespace is required between alphanumeric tokens at /tmp/mTBTXUFHnX line 1:␤------> my @stuff[ %100⏏_000 ]␤FAILED 00:01 105m␤»
diakopter isn't that just a LTA error?
22:51 k23z__ joined
diakopter rakudo: say @*_ ~ 'hi' 22:52
masak it should be kosher, but due to how STD.pm doesn't special-case % there, it isn't.
no, it shouldn't be an error.
by my reading of the spec.
'%' here means 'wrap the edges'/
s[/][.]
according to S09.
p6eval rakudo 37e574: OUTPUT«Contextual @*_ not found␤current instr.: 'perl6;Perl6Exception;throw' pc 14482 (src/builtins/Seq.pir:52)␤»
diakopter rakudo: say @*_
p6eval rakudo 37e574: ( no output )
diakopter rakudo: say @*_; say 'hi' 22:53
p6eval rakudo 37e574: OUTPUT«hi␤»
diakopter sometimes say fails silently
masak yes. I've had that too.
time for me to sleep. 22:54
diakopter: I'll collect any further bugs you find when I wake up :) 22:55
diakopter rakudo: @_.=[3]
p6eval rakudo 37e574: OUTPUT«Method '!postcircumfix:<[ ]>' not found for invocant of class 'ResizablePMCArray'␤current instr.: '!dispatch_.=' pc 440 (src/glue/dispatch.pir:118)␤»
masak that one, for example :)
o/
22:55 masak left
diakopter anytime I see ResizablePMCArray in an error msg, I consider it a bug 22:55
l8r
rakudo: say @_.perl 22:57
p6eval rakudo 37e574: OUTPUT«Method 'perl' not found for invocant of class 'ResizablePMCArray'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
jnthn wonders why it's init'd to @_.
22:58 pmurias left 22:59 k23z__ left 23:04 lestrrat is now known as lest_away 23:12 rv2733 joined 23:16 iblechbot left
lichtkind i saved www.perlfoundation.org/perl6/index....iecode_org 23:18
jnthn: o/ 23:20
23:20 meppl left 23:21 quietfanatic joined 23:26 nihiliad left 23:32 lest_away is now known as lestrrat
pmichaud jnthn: ping 23:39
phenny pmichaud: 22:47Z <jnthn> tell pmichaud I think we'll go with pannel session rather than a replacement talk.
pmichaud phenny: tell jnthn panel session sounds awesome. Between today and tomorrow I'll put together a list of bullet points of things I wanted to get across/accomplish with my talk; I suspect you all can work those in (or use them as starting items for the talk) on Saturday 23:41
phenny pmichaud: I'll pass that on when jnthn is around.
colomon pmichaud: Did you see the gather / take bug the other day? (I'm somewhat out of it at the moment, but...) 23:43
rakudo: ((1...*) Z (3...*)).batch(20).perl.say 23:45
p6eval rakudo 37e574: OUTPUT«(3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22)␤»
colomon That's a little vague, but I'm 99% sure the issue is that you can only have one instantiation of a given gather block running at a time. 23:47
pmichaud colomon: okay, I can look into a bit. 23:48
*into it
the code is supposed to be set up to handle multiple given/take blocks 23:49
colomon Thanks.
original (more complicated) report was:
lastofthecarelessmen.blogspot.com/2...acked.html
Tene pmichaud: the problem with gather/take is the sub cloning, unless that's been fixed. 23:50
pmichaud that makes sense 23:51
I did some work on the sub cloning, but my "fix" caused other things to break so I abandoned it 23:52
I'll look into it again.
23:53 eternaleye left
colomon just to make sure I was clear -- I'm not seeing a problem using different gathers at the same time. It's when the same gather code is called twice and you then try to run both iterators at the same time. 23:53
which makes it sound like cloning could indeed be the issue.
23:55 snarkyboojum joined, rv2733 left
lue hello! (looking at the backlogs, it seems we could define "day" and "night" by the most prominent people around at certain times) 23:58
(ie. not day, but masak. Not the nightshift, but the jnthn-shift)