Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2015-05-28

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:17 raiph joined #perl6
00:28 laouji joined #perl6
00:28 Ben_Goldberg joined #perl6
00:30 ShimmerFairy joined #perl6
00:35 virtualsue joined #perl6
00:35 [Coke] jnthn++ #congrats
00:42 virtualsue left #perl6
00:50 laouji joined #perl6
00:54 laouji_ joined #perl6
01:07 vendethiel joined #perl6
01:11 kurahaupo1 joined #perl6
01:13 firefish5000 joined #perl6
01:26 BenGoldberg_ joined #perl6
01:26 zcm joined #perl6
01:39 dayangkun joined #perl6
01:55 laouji_ joined #perl6
02:02 japhb .tell jnthn Congratulations!  Where are you getting married?  (Given your propensity to travel, I'm not ruling any countries out.  :-)
02:02 yoleaux japhb: I'll pass your message to jnthn.
02:09 dolmen joined #perl6
02:12 hoelzro jnthn: congratulations!
02:20 laouji joined #perl6
02:22 egrepnix joined #perl6
02:27 davido_ joined #perl6
02:27 BenGoldberg_ joined #perl6
02:29 colomon jnthn: congratulations indeed!
02:34 rmgk_ joined #perl6
02:43 colomon joined #perl6
02:46 noganex joined #perl6
02:49 laouji joined #perl6
02:54 Ben_Goldberg joined #perl6
02:54 aborazmeh joined #perl6
02:54 aborazmeh joined #perl6
03:18 fhelmberger joined #perl6
03:32 BenGoldberg_ joined #perl6
03:41 laouji joined #perl6
03:45 zcm joined #perl6
03:46 BenGoldberg joined #perl6
04:05 zcm left #perl6
04:11 laouji joined #perl6
04:20 skids m: multi sub postfix:<ꜝ> (Int $i where * == 0) { 1 }; multi sub postfix:<ꜝ> (Int $i where * > 0) is cached { "OHAI".say; $i * ($i - 1)ꜝ }; *ꜝ.say for 0..4 # LTA error
04:20 camelia rakudo-moar c2a57e: OUTPUT«1␤Cannot invoke object with invocation handler in this context␤  in whatevercode  at /tmp/JYRKWVt1gG:1␤␤»
04:21 skids m: proto sub postfix:<ꜝ> ($i) is cached { * }; multi sub postfix:<ꜝ> (Int $i where * == 0) { 1 }; multi sub postfix:<ꜝ> (Int $i where * > 0) { "OHAI".say; $i * ($i - 1)ꜝ }; *ꜝ.say for 0..4 # (this works fine)
04:21 camelia rakudo-moar c2a57e: OUTPUT«1␤OHAI␤1␤OHAI␤2␤OHAI␤6␤OHAI␤24␤»
04:22 skids I don't know if the first one is supposed to work or not.
04:41 dalek ecosystem: d84495f | (David Warring david.warring@gmail.com)++ | META.list:
04:41 dalek ecosystem: added Font::AFM (Perl 6 port); moved PDF::Grammar
04:41 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/d84495f07c
04:42 awwaiid joined #perl6
04:56 gtodd joined #perl6
05:04 jack_rabbit joined #perl6
05:04 b2gills skids: that's at the wrong precedence level though
05:04 b2gills m: proto sub postfix:<ꜝ> ($i) is cached { * }; multi sub postfix:<ꜝ> (Int $i where * == 0) { 1 }; multi sub postfix:<ꜝ> (Int $i where * > 0) { "OHAI".say; $i * ($i - 1)ꜝ }; say -1ꜝ
05:04 camelia rakudo-moar c2a57e: OUTPUT«OHAI␤-1␤»
05:05 zengargoyle joined #perl6
05:05 b2gills I don't know what the precedence should be, but I have come across an example that uses:
05:05 b2gills m: proto sub postfix:<ꜝ> ($i) is cached is tighter(&[*]) { * }; multi sub postfix:<ꜝ> (Int $i where * == 0) { 1 }; multi sub postfix:<ꜝ> (Int $i where * > 0) { "OHAI".say; $i * ($i - 1)ꜝ }; say -1ꜝ
05:05 camelia rakudo-moar c2a57e: OUTPUT«Cannot call postfix:<ꜝ>(-1); none of these signatures match:␤    (Int $i where { ... })␤    (Int $i where { ... })␤  in any call_with_capture at src/gen/m-Metamodel.nqp:3470␤  in any enter at src/gen/m-Metamodel.nqp:3542␤  in block <unit> at /t…»
05:16 inokenty joined #perl6
05:27 vendethiel joined #perl6
05:30 laouji_ joined #perl6
05:31 awwaiid joined #perl6
05:31 gfldex joined #perl6
05:35 virtualsue joined #perl6
05:47 cognominal joined #perl6
05:51 xinming joined #perl6
06:00 diana_olhovik_ joined #perl6
06:15 domidumont joined #perl6
06:15 kaare_ joined #perl6
06:19 domidumont joined #perl6
06:22 _mg_ joined #perl6
06:26 jack_rabbit joined #perl6
06:27 virtualsue joined #perl6
06:32 pdcawley joined #perl6
06:39 pdcawley joined #perl6
06:45 awwaiid joined #perl6
06:51 brrt joined #perl6
06:52 zakharyas joined #perl6
06:55 RabidGravy joined #perl6
07:04 ab5tract_ joined #perl6
07:11 tinyblak joined #perl6
07:13 FROGGS joined #perl6
07:13 domidumont joined #perl6
07:14 tinyblak_ joined #perl6
07:17 awwaiid joined #perl6
07:18 salv0 joined #perl6
07:19 tinyblak joined #perl6
07:34 AlexDaniel joined #perl6
07:37 g4 joined #perl6
07:37 [Sno] joined #perl6
07:40 masak morning, #perl6
07:41 nwc10 morning masak
07:47 brrt \o masak
07:50 vendethiel joined #perl6
07:52 lizmat morning masak nwc10 brrt
07:56 ab5tract_ m: (4..5).rand.ray
07:56 camelia rakudo-moar c2a57e: OUTPUT«Method 'ray' not found for invocant of class 'Num'␤  in block <unit> at /tmp/ZrkJta_Opv:1␤␤»
07:56 ab5tract_ m: (4..5).rand.say
07:56 camelia rakudo-moar c2a57e: OUTPUT«1.92761249929782␤»
07:56 ab5tract_ not what I was expecting, there
07:56 ab5tract_ and: good morning, #perl6
07:57 ab5tract_ derp
07:57 ab5tract_ that's what pick and roll are for :)
07:58 abraxxa joined #perl6
07:58 lizmat Files=1010, Tests=42510, 168 wallclock secs (10.08 usr  3.36 sys + 988.13 cusr 104.52 csys = 1106.09 CPU)
07:58 xinming joined #perl6
07:58 lizmat first time in a long time that I see a spectest < 170 secs
07:58 lizmat and at such a low CPU
07:59 ab5tract_ lizmat: sweet!
08:00 dalek rakudo/nom: 014597e | lizmat++ | src/core/ (2 files):
08:00 dalek rakudo/nom: Put in goto NYI stub
08:00 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/014597e1ca
08:00 lizmat $ 6 'FOO: goto FOO'
08:00 lizmat Label.goto() not yet implemented. Sorry.
08:00 lizmat $ 6 'FOO: FOO.goto'
08:00 lizmat Label.goto() not yet implemented. Sorry.
08:07 lizmat timotimo: yeah, I saw that, and that's exactly what I got with the THROW MMD change
08:08 lizmat I also tried MMDing 'return', but that has a weird failure mode in building / starting the setting  :-(
08:08 lizmat pity, I think we can gain a lot there as well...
08:11 darutoko joined #perl6
08:19 dalek specs/newio: d2d0057 | lizmat++ | S32-setting-library/Containers.pod:
08:19 dalek specs/newio: Mention :into / :as for classify / categorize
08:19 dalek specs/newio: review: https://github.com/perl6/specs/commit/d2d0057cf1
08:21 stux|RC-only joined #perl6
08:30 dalek rakudo/nom: f5b360b | lizmat++ | docs/ChangeLog:
08:30 dalek rakudo/nom: Some more recent changes
08:30 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f5b360bd4e
08:32 dalek rakudo/nom: 802983e | lizmat++ | tools/contributors.pl6:
08:32 dalek rakudo/nom: Include any changes to "doc" in contributors
08:32 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/802983e773
08:40 telex joined #perl6
08:44 dakkar joined #perl6
08:49 laouji_ joined #perl6
08:51 espadrine joined #perl6
08:57 masak "This synopsis summarizes the non-existent Apocalypse 9, which discussed in detail the design of Perl 6 data structures. It was primarily a discussion of how the existing features of Perl 6 combine to make it easier for the PDL folks to write numeric Perl." -- S09
08:57 masak I've always been a bit curious about how this non-existent A09 would look :>
08:58 masak today is not the day, but maybe some day I'll attempt to write it, just to see what it looks like.
08:59 masak naturally, it would have to contain (a) TimToady's "voice" and jokes, (b) understandable mis-features that were since fixed, some of them historical, some of them not.
09:00 masak (c) later corrections to these misfeatures, explaining how things are now in S09
09:00 masak (d) discussions around the appropriate RFCs
09:01 masak (but probably not focused on that, since by the time of A06 and A12, Perl 6 had already established itself enough that "ok, here's what we'll do" was the main thrust of an apocalypse)
09:02 laouji joined #perl6
09:02 fhelmberger joined #perl6
09:02 fhelmberger joined #perl6
09:03 masak by the way, I think `use autoindex;` would be a misfeature, and would be much better served by a cute, targeted slang.
09:05 masak m: my num %hash; say %hash.^name
09:05 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/rbVL630mYq␤native value types for hashes not yet implemented. Sorry. ␤at /tmp/rbVL630mYq:1␤------> 3my num %hash7⏏5; say %hash.^name␤    expecting any of:␤        constraint␤»
09:05 masak m: my Num %hash; say %hash.^name
09:05 camelia rakudo-moar c2a57e: OUTPUT«Hash[Num]␤»
09:05 masak m: my Num %hash{<a b c d e f>}; say %hash.^name
09:05 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/bBm3QzYSDN␤Invalid hash shape; type expected␤at /tmp/bBm3QzYSDN:1␤------> 3my Num %hash{<a b c d e f>}7⏏5; say %hash.^name␤    expecting any of:␤        constraint␤»
09:05 andreoss joined #perl6
09:06 * masak is really looking forward to all the "shape" shenanigans of S09
09:11 andreoss joined #perl6
09:12 darutoko- joined #perl6
09:19 andreoss m: my @x; @x.push($_) for 1..1000000; @x.join.chars.say #
09:19 camelia rakudo-moar c2a57e: OUTPUT«5888896␤»
09:19 andreoss m: say chars [~] do $_ for 1..1000000
09:19 camelia rakudo-moar c2a57e: OUTPUT«Memory allocation failed; could not allocate 2833620 bytes␤»
09:31 moritz repeated string concatenation with immutable strings is a Really Bad Idea[tm]
09:33 masak one might hope that some day [~] will copmile down to the right thing, though...
09:33 masak it's a fairly natural pattern to apply to that problem.
09:34 inokenty joined #perl6
09:51 tinyblak joined #perl6
10:03 andreoss the second variant calls Str.new only as twice as much
10:05 andreoss so it's really ~ which is slowing things down. am i wrong?
10:05 andreoss my point was .join works much faster than ~
10:07 larion joined #perl6
10:09 * brrt is not really sure how ~ is implemented
10:10 brrt hmm
10:10 brrt can i do this:
10:10 moritz [~] probably naively iterates + concatenates, which is a O(n^2) algorithm
10:11 brrt m: sub foo($a,$b) { $a + $b*$a }; say [&foo] 1..6;
10:11 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/1_Ex3WpXB_␤Two terms in a row␤at /tmp/1_Ex3WpXB_:1␤------> 3ub foo($a,$b) { $a + $b*$a }; say [&foo]7⏏5 1..6;␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤    …»
10:11 andreoss m: sub foo($a,$b) { $a + $b*$a }; say [[&foo]] 1..6;  [06:11]
10:11 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5===␤Confused␤at /tmp/MDUcLLiN1v:1␤------> 3{ $a + $b*$a }; say [[&foo]] 1..6;  [06:7⏏11]␤    expecting any of:␤        colon pair␤Other potential difficulties:␤    Leading 0 does not indicate octal in Perl 6; please use…»
10:11 andreoss m: sub foo($a,$b) { $a + $b*$a }; say [[&foo]] 1..6;
10:11 camelia rakudo-moar c2a57e: OUTPUT«2520␤»
10:11 moritz m: sub foo($a,$b) { $a + $b*$a }; say reduce &foo 1..6;
10:11 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/DE8IzTFbZT␤Two terms in a row␤at /tmp/DE8IzTFbZT:1␤------> 3o($a,$b) { $a + $b*$a }; say reduce &foo7⏏5 1..6;␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤    …»
10:12 brrt why's that with double [[]]
10:12 moritz m: sub foo($a,$b) { $a + $b*$a }; say reduce &foo, 1..6;
10:12 camelia rakudo-moar c2a57e: OUTPUT«2520␤»
10:12 brrt ah, ok
10:12 andreoss brrt: the first pair makes it infix, the second pair makes it a reducer
10:12 moritz I'd say the other way round :-)
10:12 brrt oh, [foo] is reducing... clearly
10:12 brrt ah
10:13 brrt that explains why your code blows up, indeed
10:13 brrt actually
10:13 andreoss m: sub infix:«foo>($a,$b) { $a + $b*$a }; say [&foo] 1..6;
10:13 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/mTz04ZIbP5␤Unable to parse expression in shell-quote words; couldn't find final '»' ␤at /tmp/mTz04ZIbP5:1␤------> 3($a,$b) { $a + $b*$a }; say [&foo] 1..6;7⏏5<EOL>␤    expecting any of:␤       …»
10:13 g4 joined #perl6
10:13 g4 joined #perl6
10:13 andreoss m: sub infix:«foo»($a,$b) { $a + $b*$a }; say [&foo] 1..6;
10:13 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/_nan6g6aRl␤Two terms in a row␤at /tmp/_nan6g6aRl:1␤------> 3:«foo»($a,$b) { $a + $b*$a }; say [&foo]7⏏5 1..6;␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤  …»
10:13 brrt it does not really explain it, unless the reduce is recursive?
10:14 moritz infix:«foo> doens't install a &foo
10:14 andreoss m: sub infix:«foo»($a,$b) { $a + $b*$a }; say [foo] 1..6;
10:14 camelia rakudo-moar c2a57e: OUTPUT«2520␤»
10:29 Ven joined #perl6
10:31 vendethiel joined #perl6
10:31 Ven_ joined #perl6
10:32 * brrt afk
10:33 oetiker joined #perl6
10:33 Ven_ o/, #perl6
10:34 Ven_ jnthn++: congrats as well
10:57 dustinm` joined #perl6
11:15 vendethiel joined #perl6
11:16 brrt joined #perl6
11:23 lizmat is there a reason why ops like infix:<===> do not have any Mu candidates ?
11:23 firefish5000 joined #perl6
11:24 lizmat m: Mu === Any  # expecting False
11:24 camelia rakudo-moar c2a57e: OUTPUT«Cannot call infix:<===>(Mu, Any); none of these signatures match:␤    (Any $?)␤    (Any $a, Any $b)␤    (Int:D \a, Int:D \b)␤    (int $a, int $b)␤    (Num:D \a, Num:D \b)␤    (Num $ where { ... }, Num $ where { ... })␤    (num $a, num $b --> …»
11:25 lizmat m: (100..200).classify({.comb})   # how I found it
11:25 camelia rakudo-moar c2a57e: ( no output )
11:25 lizmat m: (100..200).classify({.comb}).perl   # how I found it
11:25 camelia rakudo-moar c2a57e: OUTPUT«Cannot call infix:<===>(Any, Mu); none of these signatures match:␤    (Any $?)␤    (Any $a, Any $b)␤    (Int:D \a, Int:D \b)␤    (int $a, int $b)␤    (Num:D \a, Num:D \b)␤    (Num $ where { ... }, Num $ where { ... })␤    (num $a, num $b --> …»
11:25 [ptc] joined #perl6
11:25 burnersk joined #perl6
11:26 pauluu joined #perl6
11:28 moritz lizmat: === autothreads (by TimToady++'s explicit wish), so it can't accept Mu as its argument
11:28 moritz m: (100..200).classify({.comb}).eager
11:28 camelia rakudo-moar c2a57e: ( no output )
11:29 moritz m: (100..200).classify({.comb}).perl
11:29 camelia rakudo-moar c2a57e: OUTPUT«Cannot call infix:<===>(Any, Mu); none of these signatures match:␤    (Any $?)␤    (Any $a, Any $b)␤    (Int:D \a, Int:D \b)␤    (int $a, int $b)␤    (Num:D \a, Num:D \b)␤    (Num $ where { ... }, Num $ where { ... })␤    (num $a, num $b --> …»
11:29 _mg_ joined #perl6
11:29 moritz I'm not too happy with this.
11:30 lizmat so: 'TValue === Mu' is by definition wrong then
11:31 moritz right
11:50 laouji joined #perl6
11:50 noface joined #perl6
11:51 noface Testing
11:51 masak noface: hi! \o
11:52 noface Oh my, are you really bored? I just needed a channel to test my theme on, I really detest perl. I like C, Lisp and Haskell.
11:53 DrForr Sorry you accidentally joined us then.
11:54 laouji_ joined #perl6
11:54 * masak hugs noface
11:54 masak noface: not bored, far from it ;)
11:55 masak doing $dayjob stuff, wishing I had more time for #perl6 today
11:56 noface Why would anyone like Perl? Python for scripting.
11:57 DrForr Fun language, quirky syntax, plenty of tools to play wth.
11:57 DrForr *with
11:57 masak maybe you've been unfortunate enough to see a lot of shoddy Perl. if you've seen enough beautiful Perl, your perspective changes.
11:58 masak of course, Perl 6 makes it easier to get things right in this regard. it's been rebased on more modern industry standards.
11:59 noface It's practically a flavour of the month language
11:59 DrForr Flavor of the month that's been around for 25+ years, I'd call that staying power.
12:00 noface Flavour of the month in terms of usage. Hipsters wanting to use new libraries and so on, so forth
12:02 DrForr Shrug, you mean like new frameworks like Dancer and Mojolicious? You're right, that's *so* last month.
12:03 rindolf joined #perl6
12:03 masak wow, I had honestly not heard Perl called that before.
12:04 noface I have no idea, as I don't use Perl. Like I've said before, I use python for scripting, and C/Haskell/Lisp for everything else. GLSL for shaders, though I have contributed to HLSL projects since I dualboot windows.
12:04 masak maybe you meant to troll #ruby?
12:04 noface Ruby and Perl are constantly at war with each other, I know that muhc
12:04 noface much*
12:05 masak your knowledge does you credit.
12:05 noface Still, I'm not too fond of high level languages like Perl. For that, I'd much rather use python, as it's far more pleasant to use.
12:05 profan who joins a language channel just to say they don't like the language though?
12:05 noface I was merely testing my theme!
12:05 masak profan: we don't get a lot of them, for sure. but they do exist.
12:05 DrForr profan: Boring trolls.
12:05 noface Hardly a "troll"
12:06 profan noface: the "i detest perl" part though wasn't really necessary was it though? :P
12:06 profan but i digress
12:06 masak noface: I use both Perl and Python. I find Perl a lot more pleasant to use, for several reasons. people's tastes diverge.
12:07 DrForr You're right. Minor nuisance would be a better description. Maybe go back to #python and practice somewhere first?
12:07 noface Well, Perl is objectively less intuitive, which is slightly off putting when all you want to do is write a high level script
12:07 masak he's not a very good troll, I agree.
12:07 masak needs practice.
12:07 masak but I admit I laughed at the "testing my theme" excuse :D
12:08 noface Back to python? Never been there, never will be. Barely use it anyway.
12:08 DrForr Good to know that intuition is so objective, I'm sure philosophers worldwide will cheer.
12:08 noface It's not an excuse. I was testing some themes and couldn't find a channel to test them on (would rather not annoy my home channels!)
12:09 masak well, you annoyed a very agreeable channel instead.
12:09 profan Maybe he redefined what objective means? He does like C after all, them there "macros" and all ;)
12:09 DrForr I fail to see how joining a random channel to chatter at it is considered a "theme".
12:09 masak it is if the theme is trolling...
12:09 noface I said "test" and Masak immediately greeted me
12:09 profan I assume he means his client's theme
12:09 masak noface: how dare he!
12:09 noface Profan is right
12:10 nwc10 test?
12:10 DrForr I assume I don't care. *plonk*
12:10 masak nwc10: hi! \o
12:10 profan nwc10: test!
12:10 masak profan: hi! \o
12:10 profan masak: hi :D
12:10 masak :D
12:10 noface Very well. What are you guys working on?
12:11 masak there's a compiler called Rakudo. a lot of work goes on that.
12:11 masak but also the underlying VMs (JVM and MoarVM) and specification/documentation work.
12:11 masak this year is a big year for Perl 6.
12:11 DrForr Tracing a segfault in a perl6 readline module while setting up a Redis cluster.
12:12 tinyblak joined #perl6
12:12 * profan is writing a toy language and compiler to learn P6
12:12 masak last time I had time for Perl 6 I was tinkering with scripts and old modules of mine.
12:12 DrForr profan: Check out perl6-ANTLR4 on github for ideas :)
12:12 andreoss profan: don't forget to put it in perl6-examples
12:13 DrForr Oh, yeah, I probably should add that to the module list.
12:13 profan DrForr: nice, thanks
12:13 tinyblak_ joined #perl6
12:13 profan it's pretty hard to find whats out there for perl6 already sometimes
12:13 masak yeah, it's a tough problem to crack :/
12:14 DrForr I'm almost done writing the AST, which should make an ANTLR => Perl6 grammar translator straightforward, with the exception of a few edge cases.
12:15 profan It's pretty interesting to see how different the attitude is in certain communities surrounding programming languages
12:15 noface profan: You mocked C a few minutes ago, but knowing C is very useful, especially when you want to branch out to GPU programming, like writing shaders for instance, since HLSL and GLSL both look a lot like C.
12:15 profan noface: There are valid reasons to mock every language, I like C myself, but that doesn't mean I can't criticize it, quite the opposite
12:16 masak I believe C macros can be used for good. MoarVM's use of them is a nice example of this.
12:17 profan If one is comfortable in their choices of tools, you should be well versed in both the pros and cons!
12:17 masak but I also believe there are macro systems which help you a lot more and which are safer.
12:17 noface And if you know C to a reasonable degree, you most likely know some assembly as well (it actually makes pointers considerably easier to understand and use since you know WHY they were created in the first place)
12:17 arnsholt Yeah. You can do good things with C macros, it's just that they go out of their way to make it hard to do so
12:17 DrForr One of my backlog projects is to snarf enough of the existing C parser to write a quick C=>P6 nativecall converter.
12:18 DrForr Sort of like the old h2xs but with less cruft :)
12:18 arnsholt Yeah, it's on my list of awesome things too
12:19 arnsholt Just be warned: There's a whole army of dragons lurking in there
12:19 DrForr I assume the pun was intentional.
12:20 profan the best puns are unintentional
12:20 arnsholt Not sure which pun you mean, so not intended, no =)
12:20 profan arnsholt: compilers... dragons? :P
12:20 DrForr That one, yes.
12:21 DrForr noface: I keep meaning to check out GPU stuff, got any references?
12:21 arnsholt Oh right, that one
12:21 larion joined #perl6
12:23 noface Depends, what would you like to do?
12:24 noface I am in the process of fixing MPDN's (media player dot net) chroma scaler, since it upscales each chroma channel individually before merging them, instead of merging them AND THEN upscaling
12:24 DrForr Offloading matrix calculations, nothing actually screen-visible.
12:24 noface Which incurs a large performance penalty (80-100%)
12:24 noface You can just use OpenCL
12:25 DrForr Nod, checking the usual references.
12:25 tinyblak joined #perl6
12:26 dalek rakudo/nom: 478994f | lizmat++ | src/core/Hash.pm:
12:26 dalek rakudo/nom: Fix .perl on uncontained typed hashes
12:26 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/478994fba4
12:27 DrForr Roight.
12:27 DrForr WW
12:27 masak :)
12:28 lizmat afk&
12:29 DrForr The FFT compute kernel on Wikipedia looks reasonable.
12:29 Alina-malina joined #perl6
12:31 hoelzro morning #perl6
12:31 DrForr Morning yourself ;)
12:46 Ven joined #perl6
12:47 no_face left #perl6
12:53 muraiki joined #perl6
13:02 pdcawley joined #perl6
13:03 pdcawley joined #perl6
13:03 jdv79 moritz: can i get an acct on the server?
13:04 moritz jdv79: sure; what username do you want?
13:05 jdv79 jdv is fine
13:06 inokenty joined #perl6
13:09 jdv79 moritz: seems ok, thanks
13:11 skids joined #perl6
13:12 RabidGravy joined #perl6
13:17 DrForr Cool, I just blew up MoarVM core :)
13:17 DrForr Unhandled exception: Label with no handler passed to newlexotic at src/gen/m-CORE.setting:17011  (/home/jeffreygoff/.rakudobrew/moa​r-nom/install/share/perl6/runtime/​CORE.setting.moarvm:MATCH_SAVE:0)
13:18 lizmat wow, wonder if any of my recent Label related changes is to blame
13:19 DrForr There's NativeCall involved here as well, just so you're aware.
13:19 Util m: say [ 10 <<**<< (1,2,3,4) ].perl;
13:19 camelia rakudo-moar c2a57e: OUTPUT«[10, 100, 1000, 10000]␤»
13:19 Util say [ 10 <<**<< (1..4) ].perl;
13:19 Util m: say [ 10 <<**<< (1..4) ].perl;
13:19 camelia rakudo-moar c2a57e: OUTPUT«Cannot find method 'Range'␤  in block <unit> at /tmp/nk6UBshobl:1␤␤»
13:19 DrForr And also not easily repeatable. Should I perlbug it though?
13:19 Util ??? Bug?
13:20 masak Util: looks like.
13:21 lizmat repeatable WAT's should be perlbugged  :-)
13:22 masak Util: please submit.
13:24 Util lizmat: Noted. masak: will do.
13:24 Util lizmat: I confess that when I post 2 (working,then non-working) `m:` on this channel, and ask `bug?`
13:24 Util I am also secretly asking "is this a *known* bug, that I would already know if I read the channel as much as I should" :^)
13:24 laouji_ joined #perl6
13:25 Util thanks all!
13:25 lizmat Util: fwiw, it didn't look familiar to me
13:25 inokenty joined #perl6
13:28 raiph joined #perl6
13:28 Util Is there a bug script now, or do we submit directly to RT?
13:28 lizmat rakudobug@perl.org
13:29 Util lizmat: right, thanks!
13:29 lizmat yw  :-)
13:30 tinyblak joined #perl6
13:31 smls joined #perl6
13:41 lizmat afk again
14:10 smls Is pipe() documented anywhere?
14:11 smls S32 only says "pipe   Gone, see Pipe.pair"
14:11 smls yet that does not reflect reality :P
14:12 Ven joined #perl6
14:12 smls Ok, this seems to work locally:   .say for pipe("ls -l").lines.map({ "<$_>" })
14:13 * Ven waves
14:13 masak \o
14:13 smls Who needs documentation when your programming languages has simple english routine names and built-in introspection capabilities :P
14:16 * cognominal particulates
14:20 * masak is more about information, probabilities, observables, and how they relate to each other
14:21 masak man, that's probably the innest in-joke I've ever done. I'd be (happily) surprised if someone gets the reference. :)
14:22 DrForr Grammar matches with optional keys now dump rather oddly.
14:22 Woodi masak: I instantly thinked about La Femme Nikita :)
14:22 Woodi hi today :)
14:23 DrForr actions => , content => action => (Any ), ... # for a quick sample.
14:25 DrForr Actually a lot has gone away.
14:27 RabidGravy making a Solaris VM just for fun
14:29 smls Another thing I like about Perl 6: It has syntactic sugar which lets me often avoid writing things twice...
14:29 smls m: my %color = (<cherry strawberry> X=> "red"), (<banana lemon> X=> "yellow"); say %color.perl
14:29 camelia rakudo-moar c2a57e: OUTPUT«{:banana("yellow"), :cherry("red"), :lemon("yellow"), :strawberry("red")}<>␤»
14:30 Ven smls: that's why cobol doesn't have docs (it actually does)
14:30 DrForr eval: grammar Foo { token TOP { <Foo>? <Bar> } token Foo { 'foo' } token Bar { 'bar' } } class Action { method TOP($/) { make { foo => $/<Foo>.ast, bar => $/<Bar>.ast } } method Foo { make ~$/ } method Bar { ~$/ } } my $g=Grammar.new;my $a=Action.new;my $x=$g.parse('bar',:actions($a));say $x;
14:30 DrForr m: grammar Foo { token TOP { <Foo>? <Bar> } token Foo { 'foo' } token Bar { 'bar' } } class Action { method TOP($/) { make { foo => $/<Foo>.ast, bar => $/<Bar>.ast } } method Foo { make ~$/ } method Bar { ~$/ } } my $g=Grammar.new;my $a=Action.new;my $x=$g.parse('bar',:actions($a));say $x;
14:30 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/NO5vuc_z3I␤Strange text after block (missing semicolon or comma?)␤at /tmp/NO5vuc_z3I:1␤------> 3grammar Foo { token TOP { <Foo>? <Bar> }7⏏5 token Foo { 'foo' } token Bar { 'bar' }␤    expecting an…»
14:31 DrForr Hit another whitespace "improvement", have I.
14:32 DrForr m: grammar Foo { token TOP { 'a' } token Foo { 'a' } }
14:32 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/2NDELqetlt␤Strange text after block (missing semicolon or comma?)␤at /tmp/2NDELqetlt:1␤------> 3grammar Foo { token TOP { 'a' }7⏏5 token Foo { 'a' } }␤    expecting any of:␤        infix␤     …»
14:32 DrForr yep.
14:33 DrForr I swear, if I keep hitting these "improvements"...
14:34 DrForr m: grammar Foo { token TOP { 'a' }; token Foo { 'a' } }
14:34 camelia rakudo-moar c2a57e: ( no output )
14:34 PerlJam DrForr: ... you'll get used to them?
14:35 DrForr m: grammar Foo { token TOP { <Foo>? <Bar> }; token Foo { 'foo' }; token Bar { 'bar' } } class Action { method TOP($/) { make { foo => $/<Foo>.ast, bar => $/<Bar>.ast } } method Foo { make ~$/ } method Bar { ~$/ } } my $g=Grammar.new;my $a=Action.new;my $x=$g.parse('bar',:actions($a));say $x;
14:35 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/LfL6VKlwu4␤Strange text after block (missing semicolon or comma?)␤at /tmp/LfL6VKlwu4:1␤------> 3ken Foo { 'foo' }; token Bar { 'bar' } }7⏏5 class Action { method TOP($/) { make { ␤    expecting an…»
14:35 DrForr m: grammar Foo { token TOP { <Foo>? <Bar> }; token Foo { 'foo' }; token Bar { 'bar' } }; class Action { method TOP($/) { make { foo => $/<Foo>.ast, bar => $/<Bar>.ast } }; method Foo { make ~$/ }; method Bar { ~$/ } }; my $g=Grammar.new;my $a=Action.new;my $x=$g.parse('bar',:actions($a));say $x;
14:35 camelia rakudo-moar c2a57e: OUTPUT«P6opaque: no such attribute '$!shared'␤  in block <unit> at /tmp/2xxvhA0I36:1␤␤»
14:35 koo6 joined #perl6
14:36 DrForr It's just the cumulative effect.
14:36 PerlJam per unit time probably
14:37 DrForr In any case, what I was hoping to demonstrate without too much code is that the output from 'say' on a match was readable and I seem to remember actually nested, but that's gone completely away.
14:41 dayangkun joined #perl6
14:41 DrForr https://gist.github.com/dr​forr/70b62c98f623b0495176
14:46 DrForr I would swear that Match or datastructure output was *much* more informative than this dump that I'm now getting.
14:47 PerlJam DrForr: are you outputting your Match object for debugging purposes?
14:47 DrForr Yep.
14:47 PerlJam If so, consider using perl6-debug-m for executing your program.
14:49 DrForr <facepalm/>
14:50 masak Woodi: nope, not La Femme Nikita. :)
14:54 DrForr Okay, I'm guessing that the rationale was that ... displaying nesting and quoting on a nested data structure would encourage people to use 'say' for simple serializing rather than resorting to a proper module?
14:55 skids say is always supposed to be for humans.
14:57 DrForr And humans shouldn't be confused with nesting?
14:58 [Coke] say is definitely no longer meant to be a canonical rep of the data
14:58 [Coke] it's using .gist, which by design doesn't show you everything.
14:59 DrForr Hrm, I was getting used to not having to use a separate module. What's the new recommended method for being able to view nested data structures? Running under -debug-m?
15:00 robinsmidsrod joined #perl6
15:00 PerlJam DrForr: perhaps you want  $match.perl  ?
15:02 skids m: grammar Foo { token TOP { <Foo>? <Bar> }; token Foo { foo }; token Bar { bar } }; class Action { method TOP($/) { make { foo => $/<Foo>.ast, bar => $/<Bar>.ast } }; method Foo ($/) { make ~$/ }; method Bar ($/) { make ~$/ } }; Foo.parse("bar",:actions(Action)).print;
15:02 camelia rakudo-moar c2a57e: OUTPUT«bar»
15:02 skids m: grammar Foo { token TOP { <Foo>? <Bar> }; token Foo { foo }; token Bar { bar } }; class Action { method TOP($/) { make { foo => $/<Foo>.ast, bar => $/<Bar>.ast } }; method Foo ($/) { make ~$/ }; method Bar ($/) { make ~$/ } }; Foo.parse("bar",:actions(Action)).perl.print;
15:02 camelia rakudo-moar c2a57e: OUTPUT«Match.new(ast => {:bar("bar"), :foo(Any)}, list => (), hash => EnumMap.new(:Bar(Match.new(ast => "bar", list => (), hash => EnumMap.new(), orig => "bar", to => 3, from => 0))), orig => "bar", to => 3, from => 0)»
15:03 DrForr Ah, yes. And good to see that no longer dumps the entire string each time. That was what drove me away from it in the first place.
15:08 hoelzro TimToady: I'm giving a talk about Perl 6 tonight, and I use your likeness in one of the slides.  Is that ok with you? If not, I'll happily remove it!
15:09 dalek ecosystem: 34a5fa2 | (Sterling Hanenkamp)++ | META.list:
15:09 dalek ecosystem: Adding HTTP::Headers to the ecosystem
15:09 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/34a5fa2bcf
15:23 vendethiel joined #perl6
15:26 circ-user-IRjqo joined #perl6
15:27 * grondilu_ bought a rasb
15:27 * grondilu_ bought a raspberry pi 2
15:27 grondilu_ I can insrt
15:27 grondilu_ o
15:27 grondilu_ ahh new keyboard, sorry
15:28 grondilu_ so, I bought a raspberry pi.  I can install rakudo on ths
15:28 grondilu_ so, I bought a raspberry pi.  I can install rakudo on this, right?
15:29 * grondilu_ also bought a chromebook
15:29 [Tux] building your questions to a climax?
15:30 grondilu_ no, just not quite used to the keyboard on the chromebook
15:30 captain-adequate joined #perl6
15:31 ssqq_ joined #perl6
15:35 rarara joined #perl6
15:40 Ulti grondilu_ the downside is floating point is not straight forward on the pi if you want it to be fast, you need to get a special JVM for it no idea how hard it would be to hack that into moarvm
15:53 skids zostay++
15:54 timotimo hmm
15:54 timotimo ah, it's the raspberry pi *2*
15:55 timotimo that one's a whole lot more powerful than the number 1
15:55 timotimo i'm not sure anybody has tried building rakudo on it yet
15:56 raiph joined #perl6
15:58 ugexe i did
15:58 lizmat sometimes I wish we had slurpies like: sub a(*?@a) { ... }
15:58 lizmat which would match on at least 1 parameter passed, and not none
15:58 ugexe i built rakudo on rpi2 on linux and netbsd. jvm takes too much memory though
15:59 lizmat m: await  # #125257
15:59 camelia rakudo-moar c2a57e: ( no output )
16:00 diana_olhovik_ joined #perl6
16:00 dalek rakudo/nom: d841d4e | lizmat++ | src/core/asyncops.pm:
16:00 dalek rakudo/nom: Make bare await die (fixes #125257)
16:00 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/d841d4e14f
16:01 lizmat $ 6 'await'
16:01 lizmat Must specify a Promise or Channel to await on
16:01 grondilu_ timotimo: I'd gladly report success on the Raspb 2 but I need to buy a micro SD and it's late.  So later in the week.
16:01 lizmat if we would have something like a slurpy of 1..Inf parameters, the above would become a compile time error
16:03 timotimo right
16:03 raiph joined #perl6
16:03 dalek roast: 5df1265 | lizmat++ | S17-promise/start.t:
16:03 dalek roast: Add test for bare await
16:03 dalek roast: review: https://github.com/perl6/roast/commit/5df1265317
16:03 timotimo wouldn't that be +@a rather than *?@a?
16:04 lizmat timotimo: perhaps
16:04 lizmat but that would be ambiguous with @a.elems ?
16:04 lizmat perhaps *+@a
16:05 timotimo well, it's only in signatures, but you're right
16:05 lizmat std: sub a(+@a) { }
16:05 camelia std 28329a7: OUTPUT«5===SORRY!5===␤Malformed parameter at /tmp/_t7EQs9zOW line 1:␤------> 3sub a(7⏏5+@a) { }␤    expecting any of:␤     name␤   new name to be defined␤ parameter␤      routine_def␤    signature␤Parse failed␤FAILED 00:01 136m␤»
16:06 lizmat std: sub a(*+@a) { }
16:06 camelia std 28329a7: OUTPUT«5===SORRY!5===␤Malformed parameter at /tmp/DQhvOLtVNm line 1:␤------> 3sub a(7⏏5*+@a) { }␤    expecting any of:␤    name␤   new name to be defined␤ parameter␤      routine_def␤    signature␤Parse failed␤FAILED 00:00 136m␤»
16:06 lizmat m: sub a(+@a) { }
16:06 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/55RU8tWEgp␤Malformed parameter␤at /tmp/55RU8tWEgp:1␤------> 3sub a(7⏏5+@a) { }␤    expecting any of:␤        formal parameter␤»
16:09 vendethiel joined #perl6
16:12 telex joined #perl6
16:13 hoelzro if one wanted to add better .gist/.perl support for Regex objects, what would be the best way to go about that? I was thinking of storing the raw source in the Regex object, but it occurred to me that might bloat things up a bit
16:16 zostay skids: thx, perl6 made that fun to write
16:18 gfldex joined #perl6
16:19 muraiki joined #perl6
16:20 lizmat m: use trace; rx/foo/; no trace; say "bar"   # perhaps some of the trace code can be repurposed ?
16:20 camelia rakudo-moar c2a57e: OUTPUT«2 (/tmp/YhE_BBcwCY:1)␤rx/foo/␤bar␤»
16:21 lizmat hoelzro ^^
16:21 hoelzro lizmat: nice tip, thanks!
16:21 laouji joined #perl6
16:24 * hoelzro .oO( TIL about 'use trace' )
16:25 hoelzro ah, it seems use trace just prints out the source to stderr
16:25 lizmat yeah, it's underdocumented
16:25 lizmat yes, that's what it does atm
16:25 domidumont joined #perl6
16:26 lizmat I was thinking about making it write to a file instead of STDERR
16:26 Juerd stderr can be very useful; please make it configurable
16:26 tony-o_ does it just print what is executed or all source?
16:27 larion joined #perl6
16:27 lizmat it prints the source of the statement, as seen at compile time
16:27 lizmat it basically just adds the ops to do that at compile time
16:27 lizmat Juerd: yeah, but I lacked the nqp / QAST foo to make that happen
16:28 masak +1 to making regexes .gist/.perl to their source text
16:29 smls how about normal Perl 6 routines? shouldn't they be deparsed as well in .perl
16:29 PerlJam surely not .gist?  That seems slightly counter to gist's purpose
16:30 [Coke] I would imagine a human readable version of a regex would have to include its source. no?
16:31 [Coke] even if it was truncated due to size limits.
16:31 [Coke] m: my $a = rx/foo/; say $a
16:31 camelia rakudo-moar c2a57e: OUTPUT«␤»
16:32 lizmat if we're  doing this, should this text make it to precomped versions ?
16:32 lizmat fwiw, use trace will not put the statements in the code when precomping
16:34 tony-o_ anyone here very familiar with apache log format?
16:38 moritz tony-o_: you mean the "combined" access log format?
16:38 moritz or error log? other formats?
16:38 vendethiel joined #perl6
16:40 tony-o_ the common log formats - i was trying to determine if %!401{Content-type}i would be skip the log or if it was replaced with '-' on the line
16:40 tony-o_ right now i'm just replacing '-' until i can get a clearer understanding of the format
16:43 tony-o_ found it in some older docs, finally, i've been looking for about 3 hours.  way more than i spent even on this module hah
16:43 brrt joined #perl6
16:46 hoelzro lizmat: I would expect that it would make it into precomp
16:46 hoelzro maybe I should just TIAS
16:46 hoelzro see how much things bloat up
16:47 lizmat TIAS ?
16:48 [Coke] try it and see
16:49 PerlJam S99:TIAS
16:49 PerlJam (seeing if the bot is alive)
16:50 synbot6 joined #perl6
16:51 PerlJam S99:TIAS
16:51 synbot6 Link: http://design.perl6.org/S99.html#TIAS
16:52 lizmat .oO( I tried and I didn't see )
16:56 TimToady hoelzro: I think I'm officially in the category of "famous enough that they cannot prevent people from using their photo" :)
16:56 TimToady but thanks for asking :)
16:56 hoelzro =)
16:59 PerlJam If I do a google image search for "larry wall" there's a few pictures of RMS thrown in the mix.  Pick one of those as your "larry"  ;)
17:00 tony-o_ hail larry
17:03 Jonis joined #perl6
17:03 hoelzro PerlJam: I noticed that when looking for pictures
17:03 hoelzro TimToady: I should probably mention that the picture in question has your face on Ned Stark's body, with the caption "Christmas is Coming"
17:04 tony-o_ m: $*PID.say;
17:04 camelia rakudo-moar c2a57e: OUTPUT«21125␤»
17:05 hoelzro (in the interest of full disclosure)
17:05 TimToady you could use Boromir's body instead, with all the arrows...
17:05 brrt can you check in $*STDIN is a tty?
17:05 hoelzro hahaha
17:06 brrt m: say $*STDIN :t;
17:06 camelia rakudo-moar c2a57e: OUTPUT«Unexpected named parameter 't' passed␤  in block <unit> at /tmp/_RPPKxlUWd:1␤␤»
17:06 lizmat brrt: that *is* specced, but not implemented
17:06 brrt oh really
17:06 brrt hmmm
17:06 brrt how would you do it
17:07 brrt m: say $*IN.IO :t;
17:07 camelia rakudo-moar c2a57e: OUTPUT«IO::Special.new(what => "<STDIN>")␤»
17:07 * brrt supsect camelia special-cases that
17:09 lizmat m: say $*IN.t
17:09 camelia rakudo-moar c2a57e: OUTPUT«Cannot find method 'isatty': no method cache and no .^find_method␤  in block <unit> at /tmp/S8ETcGRlxl:1␤␤»
17:13 lizmat brrt: it expects to be able to call "isatty" on the $!PIO of the IO::Handle
17:13 brrt aye
17:16 smls I explored the inconsistent itemization of values going through .make some more: https://gist.github.com/smls/8a16296b48b4e4f6f011
17:16 smls Surely that's a bug, right?
17:16 smls (look at the output at the bottom)
17:18 smls Wasn't it decided sometime last year that 'make' should never force itemization?
17:18 smls ^^ TimToady, lizmat & co
17:18 TimToady I believe that's correct
17:18 brrt (surely we can implement isatty?)
17:19 brrt (method isatty is not found on instance of class BOOTIO .. either)
17:19 smls TimToady: your reply is ambiguous :P
17:19 TimToady I meant I agree with you
17:20 smls ok
17:23 brrt oh, hmm
17:23 brrt windows *does* have isatty, but calls it _isatty instead
17:23 lizmat brrt: please do  :-)
17:24 smls brrt: see rt #123347
17:24 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=123347
17:24 smls isatty seems to be availabe on parrot, but not on moar and jvm
17:25 brrt yes, lets check that out, shall i
17:25 lizmat please  :-)
17:27 pauluu left #perl6
17:32 virtualsue joined #perl6
17:32 colomon please please please make isatty work!
17:34 RabidGravy it would be easy if the fd was available from an IO::Handle
17:35 lizmat fwiw, I seem to recall it being less than straightforward because of libuv
17:35 lizmat but I could very well be misremembering
17:39 smls TimToady: I'm looking at the Rakudo source code, and method make does   $!made = made;  while sub make does  nqp::bindattr(..., '$!made', made);
17:39 smls Changing the first to  $!made := made;  might fix it, right?
17:40 dalek perl6-roast-data: f0863ce | coke++ | / (8 files):
17:40 dalek perl6-roast-data: today (automated commit)
17:40 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/f0863cec45
17:46 spider-mario joined #perl6
17:54 lizmat smls: think so
17:54 lizmat commute to FRA&
17:57 vendethiel FRA?
17:58 masak Frankfurt.
17:58 vendethiel thanks masak :)
17:58 masak Frankfurt airport, to be exact.
17:58 * masak has traveled a bit -- one gets to Frankfurt sooner or later :)
18:00 * nwc10 has travelled quite a bit within Frankfurt Airport
18:00 ShimmerFairy joined #perl6
18:01 nwc10 by foot, bus and taxiing aircraft
18:02 nwc10 it's faster to get from the airport to central Franfurt than from some of the gates to one of the runways.
18:02 nwc10 this is LTA
18:02 RabidGravy I'd just like to say building rakudo on FreeBSD 9.2 on a KVM is quite awfully slow
18:03 TimToady don't ever try to reclaim your VAT at Frankfurt, they can run you from one place to another for hours until you have to catch your flight
18:05 masak yeah, FRA is huge. but it's also nice and new and airy.
18:06 masak I've traveled a fair bit, but not enough to be jaded and tired of trav'lin'.
18:06 japhb nwc10: My impression of FRA was roughly a shopping mall embedded in a labyrinth masquerading as an airport.
18:06 RabidGravy I think all airports should be modeled on LCY
18:07 DrForr Naah, give me AMS any day.
18:07 DrForr I may be a touch biased :)
18:07 * masak likes CPH
18:08 japhb LCY: Fifth busiest airport *serving London*.  Wow, that's a lot of airports ...
18:09 [Sno] joined #perl6
18:09 RabidGravy heathrow, gatwick,luton, stansted,city
18:09 DrForr I'm heading CLJ-OTP-CDG-SLC;SLC-STL;STL-ATL-MUC-OPT-CLJ for YAPC. Someone's going to have to get me home in a *wheelbarrow*.
18:10 RabidGravy I remember when stansted was just a bunch of left over WWII huts
18:11 DrForr Stansted was where I got to spend some time in the back room, and no longer has to wonder what happens when you're in that little roped-off area.
18:11 yqt joined #perl6
18:13 DrForr *that* was a conversation killer :)
18:14 masak :)
18:18 DrForr It's a litle long to explain over IRC, if you're intereted ask me @YAPC :)
18:21 laouji joined #perl6
18:23 masak I'll try to remember that :)
18:23 synbot6 joined #perl6
18:25 pierrot joined #perl6
18:25 DrForr Wish I could be there for the hackathon, but family matters intrude.
18:33 smls lizmat:  $!made := made;  does indeed fix it, and the 'make spectest' still passes with it. I've submitted the patch to rt.
18:36 smls Not that a formal patch was necessarily warranted, seeing as it only adds a single character which you could have easily added yourself, but I never made a Rakudo patch before so I though why not... :)
18:38 synbot6 joined #perl6
18:39 PerlJam S99::TTIAR
18:39 PerlJam S99:TTIAR
18:39 synbot6 Link: http://design.perl6.org/S99.html#TTIAR
18:47 [Coke] The HTML of that seems odd. quotes around words signifies to me that we're using an incorrect word deliberately, when we really just wanted to show some kind of emphasis.
18:48 [Sno] joined #perl6
18:48 smls yeah, that probably wants <i>...</i> instaed of "..."
18:49 * jercos places a "T_PAAMAYIM_NEKUDOTAYIM" on Text::SmartLinks
18:50 TimToady The T_PAAMAYIM_NEKUDOTAYIM sits on the Text::SmarLinks, but appears to do nothing.
18:51 jercos back in the pokéball then
18:52 TimToady The Text::SmartLinks suddenly explodes.
18:52 jercos I just don't know what went wrong.
18:53 DrForr you die. --more--
18:53 jercos DYWYPI?
18:53 AlexDaniel joined #perl6
18:54 brrt joined #perl6
19:01 _mg_ joined #perl6
19:07 masak S99:floofudoofugronkle
19:07 synbot6 Link: http://design.perl6.org/S9​9.html#floofudoofugronkle
19:09 masak "warnocked - This lemma will never be explained." -- a case of trying to make a joke but ending up frustrating whoever was looking for information, methinks.
19:10 masak and I agree about the quotes. most of those quotes around links should just go away. no need even for <i>...</i>
19:14 masak wow, the `ufo` project is so old it has a non-Markdown README!
19:14 masak despite all the fun I had writing that README file, my favorite sentence nowadays is one that was added later: "My friend Ingy, who has lots of prior experience with alien technology [...]" :P
19:17 moritz :-)
19:22 laouji joined #perl6
19:32 ab5tract_ joined #perl6
19:32 raiph joined #perl6
19:50 DrForr Ingy *is* alien technology. Wasn't sure you knew that.
19:51 masak Ingy is a hoopy frood.
19:54 zakharyas joined #perl6
19:57 diana_olhovik_ joined #perl6
20:01 virtualsue joined #perl6
20:01 koo6 joined #perl6
20:03 moritz .oO( Sounds like bunny technology )
20:08 jdv79 the same ingy that created yaml?
20:10 masak the very same.
20:12 pdcawley joined #perl6
20:30 dalek roast: 1c19fa4 | jdv79++ | S14-roles/parameterized-basic.t:
20:30 dalek roast: Add tests for RT #109880.
20:30 dalek roast: review: https://github.com/perl6/roast/commit/1c19fa41a7
20:30 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=109880
20:34 tony-o_ m: require Test; Test.perl.say;
20:34 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Hh512cn5_n␤Undeclared name:␤    Test used at line 1␤␤»
20:34 tony-o_ m: require Test; Test::.perl.say;
20:34 camelia rakudo-moar c2a57e: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Xb1gGt_QqX␤Variable '&Test' is not declared␤at /tmp/Xb1gGt_QqX:1␤------> 3require Test; 7⏏5Test::.perl.say;␤»
20:34 tony-o_ m: require Test; ::('Test').perl.say;
20:34 camelia rakudo-moar c2a57e: OUTPUT«Test␤»
20:35 retupmoca masak: I'm building a new markdown module: https://github.com/retupmoca/p6-markdown - can I steal the Text::Markdown name from you, or would you prefer that both modules are in the ecosystem?
20:35 masak retupmoca: if you exceed what my Text::Markdown already does, then feel free to steal.
20:35 [Coke] Or you could test that having two modules with the same name works as expected. :)
20:36 masak or that.
20:36 masak 'night, #perl6
20:37 retupmoca is that actually supposed to be supported now?
20:37 * retupmoca still needs to get up to speed on all the CompUnitRepo changes
20:38 retupmoca if same-name-different-auth is supposed to work, or can easily be made to work, I'll just do that
20:48 Peter_R joined #perl6
20:56 dolmen joined #perl6
20:57 tony-o_ timotimo: HTTP::Server::Logger now works with anything abiding by HTTP::Server
21:00 timotimo neat :)
21:00 timotimo do we have anything to track exceptions and such, btw?
21:01 timotimo is there any way to get something like a "debug overlay" into a H:S:* powered program?
21:08 tony-o_ timotimo: i'm building that into the perl6-hiker app i'm working on
21:08 timotimo ooooh, you're already up to that? :D
21:08 tony-o_ mvc framework
21:08 tony-o_ otherwise, right now it just catches and .says
21:11 tony-o_ https://github.com/tony-o/pe​rl6-hiker/tree/master/usage
21:11 tony-o_ that's kind of the syntax i'm going for
21:11 tony-o_ similar to mojo but more concise
21:11 timotimo why is it "$app.bind; $app.listen" instead of "$trail.go-hiking"? :)
21:12 tony-o_ haha, because it's still really rough atm :-)
21:12 espadrine joined #perl6
21:41 liztormato joined #perl6
21:42 * liztormato waves from a labyrinth / shopping mall / airport / train station / hotel lobby
21:46 timotimo o/ liztormato
21:46 raiph joined #perl6
21:47 liztormato timotimo: /o
21:51 liztormato left #perl6
21:54 telex joined #perl6
21:58 tony-o_ m: role A { proto method x {...}; }; class B does A { has $.x = 5; }; B.new.x.say; #is there a better way to write this role?
21:58 camelia rakudo-moar c2a57e: OUTPUT«5␤»
22:03 vendethiel joined #perl6
22:09 lizmat joined #perl6
22:15 tony-o_ another question, does .WHO require an instance?
22:16 lizmat tony-o_: a type object is an instance of the MOP class, afaik  :-)
22:17 lizmat m: say Int.WHO
22:17 camelia rakudo-moar c2a57e: OUTPUT«␤»
22:17 lizmat m: say Int.WHO.WHAT
22:17 camelia rakudo-moar c2a57e: OUTPUT«(Stash)␤»
22:17 lizmat m: say Int.WHO.methods
22:17 camelia rakudo-moar c2a57e: OUTPUT«Method 'methods' not found for invocant of class 'Stash'␤  in block <unit> at /tmp/lNM5MymLvi:1␤␤»
22:17 lizmat hmmm....
22:17 * lizmat mixed hop WHO and HOW
22:17 lizmat *up
22:18 dalek rakudo/nom: 0a0051c | lizmat++ | src/core/Match.pm:
22:18 dalek rakudo/nom: Make sure Match.make doesn't containerize, smls++
22:18 dalek rakudo/nom:
22:18 dalek rakudo/nom: Fixes #125269
22:18 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/0a0051c697
22:18 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=125269
22:19 timotimo excellent
22:19 timotimo smls++ lizmat++
22:20 * lizmat has an early flight in the morning, so tries to get some sleep
22:22 tony-o_ night
22:23 laouji joined #perl6
22:27 timotimo good night lizmat :)
22:55 vendethiel joined #perl6
22:59 jepeway joined #perl6
23:11 AlexDaniel how can I write a comment on rt.perl.org without having an account? It said that all I have to do is just write an email with the bug id in the title, and it said that just replying to this message should work. And that's what I did. But my comment did not appear
23:17 amurf joined #perl6
23:20 tinyblak joined #perl6
23:29 laouji joined #perl6
23:34 vendethiel joined #perl6
23:44 tinyblak_ joined #perl6
23:57 itz2 joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo