»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:00 miatomi left 00:01 Cabanossi left 00:02 Cabanossi joined 00:03 mcmillhj left, eroux left, mr-foobar left 00:04 TEttinger joined 00:06 perlawhirl left 00:08 bjz left 00:13 TheovdH left, dct joined 00:14 bdmatatu joined 00:26 mcmillhj joined 00:30 Cabanossi left, mcmillhj left 00:32 dct left, Cabanossi joined 00:48 mcmillhj joined 00:49 bdmatatu left 00:53 mcmillhj left 01:09 aborazmeh joined, aborazmeh left, aborazmeh joined 01:13 mcmillhj joined 01:18 mcmillhj left 01:21 lukaramu_ left 01:40 mcmillhj joined 01:46 mcmillhj left
samcv ugh how do i use captures again with .subst i never get it right. so confusing 01:52
m: 'word'.subst(/(\w+)/, $0).say 01:53
camelia Use of Nil in string context
in block <unit> at <tmp> line 1
02:02 mcmillhj joined 02:07 mcmillhj left 02:17 dj_goku left 02:28 dj_goku joined, dj_goku left, dj_goku joined 02:30 Cabanossi left, b2gills left 02:32 Cabanossi joined, b2gills joined
llfourn m: 'word'.subst(/(\w+)/, $[0]).say 02:33
camelia 0
llfourn m: 'word'.subst(/(\w+)/, $/[0]).say
camelia Use of Nil in string context
in block <unit> at <tmp> line 1
llfourn m: 'word'.subst(/(\w+)/, $/).say 02:34
camelia Use of Nil in string context
in block <unit> at <tmp> line 1
llfourn hmmm
m: "word".subst(/(\w+)/, { $0 } ).say # seems to work 02:37
camelia word
02:38 cibs left 02:40 mcmillhj joined, cibs joined 02:45 mcmillhj left
Geth whateverable: f03f0a2780 | (Aleks-Daniel Jakimenko-Aleksejev)++ | build.p6
Start from 2015.07 instead of 2015.10

This change has been there for a long time, but it was never committed.		 02:55
whateverable: 9dc5935cba | (Aleks-Daniel Jakimenko-Aleksejev)++ | build.p6
Commit build.p6 as is

For a long time this script was suffering due to various instabilities in rakudo (see issue #24). As of today, some of the bugs were fixed, but at the same time this script itself was wiggled into some stable state. I am afraid to touch it.
Therefore, committing what we have without any clean up.
In fact, there are some issues I can see right now, but the script is proven to work in practice… 🙈
02:57 aborazmeh left 03:01 Cabanossi left 03:02 Cabanossi joined 03:03 khw left 03:06 mcmillhj joined 03:10 xtreak joined, mcmillhj left 03:15 BenGoldberg left
samcv thanks ilbelkyr 03:22
err illfourn
TEttinger www.xkcd.com/1813/ 03:23
03:25 vike joined 03:28 hobbs left 03:29 noganex joined, mcmillhj joined
samcv lol 03:29
03:30 Cabanossi left 03:31 hobbs joined
samcv i have read that new laptops aren't really much faster than older ones, i have sandybridge. do you guys think this is true. i need more speed but am thinking about maybe using another laptop i have and swapping in a quad core, if i can't get much faster speed by getting a newer 2 core laptop 03:31
03:31 noganex_ left 03:32 Cabanossi joined 03:33 sufrosti1o left, sufrostico left, sufrosti2o left 03:34 mcmillhj left
TEttinger woah, sandy bridge is rather old 03:37
how's the GPU, if any?
samcv gpu is fine for not gaming 03:38
i don't play games so that doesn't matter much
but i have 16GB ram and SSD but i need more cpu speed 03:39
have a 2.4GHz i5-2540M
03:39 KDr2_c joined
TEttinger ram performance has improved a lot if you have ddr2 03:39
samcv er 2.6ghz. that's prolly stock
2.6-3.3ghz 03:40
03:40 mcmillhj joined
samcv i think i have ddr3 03:40
yeah i have ddr3
TEttinger there's ddr4 now and GPUs often have GDDR5
samcv well i need cpu power though :)
TEttinger yes 03:41
I'd check notebookcheck
samcv i have a t540p and x220 (only laptop i actually use)
so for 160 bucks i can buy a i7-4800MQ with 4 cores and 2.7Ghz to 3.7ghz. forget how much ram is in the other laptop. probably 8 03:44
and stick it in the t540
TEttinger yours isn't bad, www.cpubenchmark.net/cpu_lookup.php...amp;id=812
motherboard?
samcv what?
TEttinger new ones probably use a newer socket
can't just update processor usually 03:45
03:45 mcmillhj left
MasterDuke ryzen is looking pretty nice... 03:46
so many cores
samcv these are all with 4 cores with the right sockets 46TDP or under ark.intel.com/Search/Advanced?s=t&...xTDPMax=47 03:47
got an ebay search to search for all 8 lol
checking cpubenchmark.net
TEttinger hm, your i5-2540 isn't on that list, samcv 03:48
samcv that's another laptop
TEttinger ohhh
samcv point is the one i use atm has that i5 cpu
other one is heavy but if it were super fast would probaby use it 03:49
03:51 mcmillhj joined
samcv and those quad core cpu's aren't on that list 03:51
none of them
TEttinger mmmm 03:53
www.microsoftstore.com/store/msusa....333475500
samcv maybe i need to search for it
ok here www.cpubenchmark.net/cpu_lookup.php?cpu=4800mq 03:54
8,515 vs 3,756
TEttinger quite a bit better, mostly the twice as many cores thing
samcv yeah
hopefully the TDP is not too much. but it's the chipset with nvidia gpu so has more cooling so i'm hoping i'll be fine 03:55
TEttinger are you in the US or somewhere in europe or somewhere else?
samcv disable the nvidia to save heat since they're 15TDP different i think
US
TEttinger ok, I also know of some sites for buying laptops in europe, since people sometimes ask for help on that 03:56
the price range is ridiculous if you are buying something at a bad time for it
03:56 mcmillhj left
samcv because of currency? 03:56
TEttinger no, more sales 03:57
u-ou perl6 rules
i mean, hi
samcv hi
TEttinger if something's being discounted heavily due to a new processor version coming out, the last version is way better on price
samcv ah
TEttinger hey u-ou
u-ou I'm a beginner 03:58
but I know perl5
03:58 mcmillhj joined
samcv this is a good intro: learnxinyminutes.com/docs/perl6/ and then there's also docs.perl6.org if you know what yo uwant to look up 03:59
04:00 Cabanossi left
u-ou I started to read that one but didn't like it 04:00
I'm watching this youtube thing and reading the docs 04:01
04:01 Cabanossi joined 04:04 cibs left 04:06 cibs joined 04:08 astj left 04:27 astj joined 04:28 cibs left 04:30 cibs joined, jraglin joined 04:31 Cabanossi left 04:32 Cabanossi joined
u-ou how do you test if a value is of a given type? 04:39
AlexDaniel ~~
m: say 42 ~~ Int 04:40
camelia True
u-ou ty
04:41 astj left
u-ou is !~ the inverse of that? 04:42
AlexDaniel !~~
u-ou ty
geekosaur note there's corner cases here, because of subtyping (e.g. Bool values match Int because enums are subtypes of Int) 04:43
04:43 Cabanossi left, astj joined 04:45 Cabanossi joined 04:46 kyan joined
u-ou > say True ~~ Int 04:47
AlexDaniel m: say True ~~ Int
camelia True
u-ou but I got that as False :o 04:48
it looks like I'm using a really old version.
04:49 mcmillhj left, wamba joined
samcv perl6 --version 04:51
what is the output u-ou
u-ou 2015.11 lol
geekosaur that one might still have Bool as a boot-enum
samcv could be newer. how did you install it
u-ou apt
on ubuntu 04:52
geekosaur (because Bool had to exist before enums did, in the setting)
samcv buggable, deb
did i do that right
geekosaur huggable, deb
huggable geekosaur, CentOS, Debian, Fedora and Ubuntu Rakudo packages: github.com/nxadm/rakudo-pkg/releases
samcv there we go
u-ou I will just use docker run rakduo-star :p 04:53
samcv what os do you have?
u-ou ubuntu
samcv i would just install the deb file from there after uninstalling with apt-get
u-ou alright 04:54
samcv sudo deb -i blah.deb # i think that's how
err
dpkg i mean
dpkg -i will work right 04:55
just make sure to remove the old one first
u-ou cool 04:57
thanks
AlexDaniel commit: 2015.11 say True ~~ Int 05:04
committable6 AlexDaniel, ¦2015.11: «False»
AlexDaniel heh
I wonder when this happened exactly 05:05
bisect: old=2015.11 say True ~~ Int
bisectable6 AlexDaniel, Bisecting by output (old=2015.11 new=43e0902) because on both starting points the exit code is 0
AlexDaniel, bisect log: gist.github.com/c01b7442ea01f5918b...1758b32cc8
AlexDaniel, (2015-11-23) github.com/rakudo/rakudo/commit/a0...d77f6a71ec
05:07 mcmillhj joined 05:10 bwisti joined 05:13 mcmillhj left 05:21 Tonik joined 05:29 mcmillhj joined 05:32 xtreak left, bwisti left 05:35 mcmillhj left, xtreak joined 05:46 curan joined 05:49 mcmillhj joined 05:50 xtreak left 05:52 xtreak joined 05:54 mcmillhj left 05:56 Todd_ joined, parv joined
Todd_ File question: If I open a file handle, then read the entire file, do I have to close it and reopen it to overwrite the file? In other words, how do i set the file pointer back thh\o the biginning? 05:57
06:00 mcmillhj joined
samcv Todd_, docs.perl6.org/routine/seek 06:01
look and you shall Seek
Todd_ Thank you Master! 06:02
06:04 Xliff_ joined, mcmillhj left 06:05 smash joined 06:08 literal_ joined, kshannon_ joined
Todd_ mumble, mumble: Undeclared routine: seek used at line 43 06:09
samcv it's a method
Todd_ 43: seek( $LogFileHandle, 0 );
samcv so get an IO object and save it to a variable and then perform things on that handle
06:09 reveres joined
samcv $var.seek(blah blah) 06:09
Todd_ what do you mean? I already have the handle and have been reading from the file as a test. Now I want to overwrite the file, also as a test 06:10
06:11 Tonik left
Todd_ How does "$var.seek(blah blah)" set my pointer back to the beginning? And is the beginning zero or one? 06:11
samcv 0
well the offset is from the specified location 06:12
06:12 cibs left, eythian_ joined, drrho_ joined
samcv $var.seek(0, SeekFromBeginning) # i think 06:12
06:13 smash_ left, revere left, literal left, kshannon left, drrho left, reveres is now known as revere, g0d355__ left
samcv that's odd. how you specify the seeking tbh. as a SeekType type object. though i already knew our IO stuff had some oddities 06:13
06:13 cibs joined 06:14 john51 left 06:15 john51 joined
Todd_ Sam: that worked. Thank you again! 06:15
$LogFileHandle.seek(0, SeekFromBeginning); for @Log -> $Line { $LogFileHandle.print( "$Line\n" ) }; 06:16
samcv yay
Todd_ By the way, thanks to you guys, I can not use $_ in a for loop. I MUST name the guy!
samcv yay 06:17
for @whatever -> $var { } like that you mean Todd_ ?
06:17 KDr2_c left
Todd_ yes! Makes what I am doign inside the {} far more maintainable as I knwo what $_ is suppose to be! I adore the feature! 06:18
samcv :) so many great perl 6 features
06:21 mcmillhj joined
Todd_ Sam: I am doing a first in, last out file. I read the file into an array, add something(s) to the end of the array, remove some stuff from the beginning or the array (maybe) and overwrite the original. I can see two ways of doing this. 1) remove elements from the beginiing of the array and writing the whole array back to the file, or 2) using a (gulp!) C loop and start writing from an index that skips over the lines to be removed. Yo 06:22
samcv Todd_, you could just ditch the filehandle and use spurt 06:23
and it'll overwrite everything instead of like writing into the file itself
if i'm describing things correctly
seek will seek to a position and it will write from then onward. so if the changed array is shorter than the original one by the number of bytes, it will retain some old data at the end of the file 06:24
06:24 Xliff_ left
Todd_ I want to remove lines from the top of the file and add lines to the bottom 06:24
06:24 Xliff_ joined
samcv so best to use `"Filename.txt".IO.spurt(@array.join("\n"))` 06:24
Todd_ okay, so remove N elements from the beginning of the array first. 06:25
samcv usually simpler to spurt and replace original file unless the file is big enough you want to get fancy
06:25 mcmillhj left
Todd_ right now it is six line. The real file when I migrate it will be close to 1000 lines of text. Mostly "date time, no errors found" type stuff 06:26
If i spurt, do I need to seek first? 06:27
samcv nope
if you use linux it's like doing `cat file > output.txt` 06:28
Todd_ Also, does shift have an option to shift N elements? or is it one at a time?
samcv yeah i think it does
Todd_ cat is a perfect example! Thank you!
samcv should be @array.shift(4) or something
if it doesn't support it i'll eat my hat
Todd_ would it be a "Red Fedora"? Chuckle! 06:29
trying it now!
samcv m: my @a; for ^5 { @a.push('a') }; say @a.shift(3)
camelia Too many positionals passed; expected 1 argument but got 2
in block <unit> at <tmp> line 1
samcv m: my @a; for ^5 { @a.push('a') }; say @a.shift
camelia a
samcv i guess i need to eat my hat now. also no it's a generic metaphorical hat :) 06:30
well Todd_ 06:32
m: my @a; for ^5 { @a.push('a') }; say @a.shift xx 2
camelia (a a)
samcv you can do this
Todd_ Ya, me too. Too many positionals passed; expected 1 argument but got 2
samcv yeah just use xx operator
will perform the command specified number of times and turn results into a Seq 06:33
06:33 mcmillhj joined
samcv m: my @a = ('a'..'z'); say @a.shift xx 5; say @a 06:33
camelia (a b c d e)
[f g h i j k l m n o p q r s t u v w x y z]
Todd_ Yippee! 06:34
what did the xx do?
06:34 darutoko joined
samcv <samcv> will perform the command specified number of times and turn results into a Seq 06:34
Todd_ Like a loop.
samcv so shift's that many times, and the results of the shifts are a sequence of things 06:35
Todd_ please tell me on the code level, you guys are just moving the element pointer and not moving data from one element to another down the line
samcv well it removes that many elements from the start of the array 06:36
and returns them
Todd_ I mean at the assembly code level. Are you just moving a pointer or are you actually moving data? 06:37
samcv similar to how shift will remove the 1st element in the array and return one value. xx performs the command specified times and the result is a Seq object
geekosaur Todd_, that's internals and likely depends on the underlying type
samcv oh
moving data
well at least the array changes things
geekosaur also, "assembly code level"? this is mostly done at nqp level currently iirc
samcv the array has elements removed
06:38 mcmillhj left
samcv and yeah what geekosaur said. it's removed from the array and if you don't use it, it'll probably be removed. if you do use it 06:38
then it'll just not be in the array
but be in some other object
Todd_ I really don't want to do $x[0]=$[1]; $x[1]=$x[2] and so forth. Areyou guys telling the reference that the array starts in a differnt location? 06:39
06:39 wamba left
samcv elements are removed from the start of the array 06:39
06:39 ufobat joined
samcv you are discarding them right Todd_ ? just need them removed? shift can do that. or if you assign the result of shift to a variable you can hold onto that data 06:40
Todd_ into the big bit bucket int he sky. I don't want them. Changing the poitner would be extremely fast. moving 1000 lines or more of text, not so much 06:41
samcv m: my @array = 10, 20; @array[0].WHERE.say; my $var = @array.shift; $var.WHERE.say
camelia 140376078134104
140376078134104
samcv in this case it is exactly the same object that gets put into `$var`
Todd_, i would think shift should be pretty fast. 06:42
Todd_ cool!
06:42 mcmillhj joined
samcv m: my @array = lines; say @array.elems 06:42
camelia 88
samcv m: my @array = lines; @array.append(@array); say @array.elems
camelia 176
06:43 Cabanossi left
samcv m: my @array = lines; @array.append(@array) xx 1000; my $t1 = now; @array.shift xx 1000; say now - $t1; 06:43
ok maybe shouldn't have made it so big
camelia (timeout)
samcv m: my @array = lines; @array.append(@array) xx 10; my $t1 = now; @array.shift xx 1000; say now - $t1; 06:44
camelia 0.0022920
samcv m: my @array = lines; @array.append(@array) xx 100; my $t1 = now; @array.shift xx 1000; say now - $t1;
camelia (timeout)
samcv oh yeah. cause uh. it appends each doubly long array to itself. so i'm sure it grows pretty fsat
but yeah it shifted 1000 elements in 0.002 seconds Todd_ so that's pretty fast 06:45
06:45 Cabanossi joined
samcv m: my @array = lines; @array.append(@array) xx 100; say @array.elems 06:45
m: my @array = lines; @array.append(@array) xx 10; say @array.elems
camelia (timeout)
90112
samcv hahah. so xx 10 is 90,000 elemns, can't even imagine what xx100 is at the rate it grows 06:46
Todd_ Then on the assembly code (or p code or whater is is being called) you are just moving an pointer. This is awesome and well though out.
thought
samcv yeah i guess that's an alright c analogy. though it's not 100% accurate. but essentially in perl 6 terms
06:47 mcmillhj left
samcv geekosaur, why don't we have a .shift(10) ? should we? 06:47
parv samcv, would 10 be number of first n elements or an index k? 06:48
samcv it would shift the first 10 elements and return a sequence or something
geekosaur samcv, isn't that splice?
parv ok. 06:49
geekosaur shift was modeled after the shell
samcv geekosaur, well splice implies splicing some other thing in there
as opposed to just removing them
geekosaur but even from the earliest perl version that had it, the some other thing could be empty 06:50
samcv m: my @array = 1,2,3,4; @array.splice(0, 2).say
camelia [1 2]
samcv ok looks like man page can be updated for splice
splice geekosaur ? 06:51
06:51 RabidGravy joined
geekosaur ? 06:51
samcv you mean splice in perl 5 or whatever
geekosaur yes
(but splice predates perl 5 iirc.) 06:52
think it came in in one of the later perl 3 patchlevels?
06:52 mcmillhj joined
samcv so that means we should not allow you to shift(Int) number of elements? 06:53
geekosaur Todd_, btw, the problem with that kind of optimization is that Perl 6 does not require that a thing that quacks like an array actually be an array. Such a pointer swizzling optimization could be provided for things backed by an actual array, but you need a fallback for when it's not.
samcv, well, it could be allowed but you have to think about duplication of work 06:54
Todd_ and that would be why you can develope 10 times faster in Perl than in C.
samcv true geekosaur
geekosaur (also splice is generally slower than the others because it always works from a fixed position)
er 06:55
samcv slower than shift?
geekosaur backwards that
samcv shift is slower?
geekosaur splice is slower because it can start anywhere, shift is faster because it starts at the same place always
samcv k
so that means there may be a reason to allow you to shift X number of elements
geekosaur ...but that's older-perl, where you always used it on an actual array, vs. perl 6 where it might not actually be a vm array
probably want to discuss this with lizmat 06:56
samcv kk
06:56 xtreak left
samcv i mean if it simply performs shift specified number of times, the normal shift will do its work right? 06:56
and act accordingly
Todd_ oh bugger, I keep capitizing My and If. mumble 06:57
samcv heh Todd_
06:57 mcmillhj left
geekosaur .oO { use Slang::Todd_ } 06:57
samcv geekosaur, you can unshift an array 06:58
geekosaur (could probably do that, starting from Slang::Tuxic or something)
Todd_ does .oO correspond to a bad word?
samcv err iugnore me
geekosaur thought bubble
samcv misread something
wait nvm. you _can_ unshift an array of things onto another array
so would make sense if you can shift a certain number of things off an array
06:58 domidumont joined 06:59 xtreak joined
samcv that may be a more compelling reason than my others for adding this. will talk with lizmat 06:59
geekosaur right. the questions are (a) older-perl decided not to do them because of overlap with splice; (b) while .shift(N) can be optimized when backed by an actual array, it ends up degrading to a loop when not so backed 07:00
Todd_ uh oh. Back to my Modula2 days. I just declared `my Int $NewNumLines;` Bad Todd. No soup for me!
parv thanks samcv (about your proposed shift function, regardless of that being available later) 07:01
geekosaur possibly other questions as well, depending on where this fits in types vs. roles like Positional
samcv Todd_, the capitalization thing you mean? 07:02
07:02 mcmillhj joined
samcv yeah of course geekosour 07:02
geekosaur (I suspect, the way things are currently set up, there is no good way to optimize for the array-backed case if you put it in Positional)
07:02 brrt joined
geekosaur (but that's likely a lizmat question) 07:02
07:03 rindolf joined 07:04 mst joined 07:07 mcmillhj left
Todd_ `My $var` and `If $var > 0`. Drives me nuts 07:08
I am capitilizing the first letter of a sentence 07:09
07:13 mcmillhj joined
Todd_ at least I am FINALLY remembering to add my semicolons at the end. 07:14
I am going to call it a night
Thank you guys for all the wonderful help!
07:17 mcmillhj left 07:20 brrt left 07:25 nadim_ joined
samcv night Todd_ 07:28
07:28 wamba joined
Todd_ :-) 07:28
07:28 Todd_ left
Woodi hallo #prl6 :) 07:34
parv holla 07:35
Woodi samcv: if you want more CPU power I will strongly suggest desktop - laptop cpus always are weeker. aaaand on top of that shiny, paper-thin laptop with moderate cpu :) 07:36
samcv i have a desktop. i never use it though
07:37 mcmillhj joined
parv was admiring iPad Mini w/ keyboard few hours ago. Don't tell to Woodi noone. 07:37
Woodi samcv: other option is remote work/testing. but not for business stuff, IMO.
samcv though the cpu i'm looking at is 2x the speed of the one i have now. so that is plenty good enough for my uses 07:38
Woodi samcv: even x220 is haevy, and T series is even better ;)
samcv though wish the t540p were lighter and smaller. x220 is smaller and lighter. well heavy for today i guess
but t540 is like more than double the weight it feels like
Woodi that's why I proposed paper-thin laptop :) but this excludes powerfull cpu 07:39
samcv also if networking goes out how will i work if i remote into something. i cannot have that
07:39 kurahaupo__ joined
Woodi and in last days we had first Ryzen bugs :) 07:40
samcv desktops are power inefficient as well. kind of wasteful to keep on
Woodi samcv: just GPUs are power ineficient :)
samcv need to see if the quad core that was an option on the t540p had a different heatsink/fan. 07:41
07:42 mcmillhj left, tojo joined, bjz joined
samcv Woodi, well the whole motherboard + power supply it all adds up 07:43
plus cpu's not designed to be as power efficient because when you don't have a battery, minimum tdp is not as important for the motherboard and chipset as a whole
Woodi samcv: but I just planning to switch from X series back to T. or even from whole Lenove thanx to backdors... China ppls realy know how to destroy their own good product...
samcv backdoors? 07:44
Woodi smash: Intel have 65 Watt desktop cpus
fish-something...
samcv i like sushi
Woodi samcv: I'm afraid of sushi :) it's raw food... 07:45
u-ou how do I use the bot?
Woodi huggable: hug u-ou
huggable hugs u-ou
u-ou lol ty
but how do you run perl? 07:46
6
Woodi u-ou: there are some docs I think, will try to find
u-ou huggable: 1 + 1
huggable u-ou, nothing found
Woodi m: say 1+1
camelia 2
u-ou cool thanks
Woodi np :)
u-ou i like sushi too!
kurahaupo__ Woodi: sushi is to cooking as a bikini is to tayloring: there's nowhere to hide if you make a mistake, so it's generally more trustworthy than ordinary food 07:47
Woodi I like butter cream from raw eggs :)
samcv u-ou, how to run perl6?
as in how to install? or like documentation or tutorials on programming
u-ou m: say 1+1
camelia 2
u-ou how to do that :P
Woodi kurahaupo__: I know it's safe :) just never tried becouse it's feel scary :)
samcv u-ou, you just did it! 07:48
u-ou I know 07:50
because Woodi showed me how
samcv i wasn't sure what the question was though :P 07:51
07:52 g4 joined
u-ou that's why I told you lol :p 07:52
07:53 kurahaupo__ left, kurahaupo__ joined
u-ou what's an idiomatic way to insert a Pair into a hash? just %h{p.key} = p.value? 07:55
07:56 mcmillhj joined
u-ou s/p/$p/g 07:57
07:57 kurahaupo__ left
parv found Go to require drastic reorientation (feels low level & verbose; Java-like) than Perl 6 or Python when coming from Perl 5, shell; may try Go again when have more free time to pay attention. 07:59
08:01 brrt joined 08:02 mcmillhj left 08:06 mcmillhj joined
u-ou ahh, method push 08:07
08:10 xinming_ is now known as xinming 08:11 mcmillhj left 08:16 wdna joined
u-ou pony 08:16
08:20 ChoHag left
Woodi some story: for years I was looking on thinkpads business-grayish displays with pity... but not so long ago I got "new" 99% sRGB, wide screen with Reader Mode. I turned ON this becouse on my previous gaming monitor I used reader mode all the time (discovered it few months ago) - with this I can look on monitor without half-close eyes... some days later in the office I look on thinkpad and something started 08:22
to bother me... "Why it is so gray? ...
... I din't turn readr mode on... Do it even have such mode?" And then I got some DING moment: It's build in! TaDa! My dear ThinkPad !
ok, maybe it's not good story...
anyway, with 430 or 440 old keyboard is gone... one more reason to drop thinkpads... is it realy so big problem to keep best keyboard in new models ??? 08:24
TEttinger heh 08:26
08:26 brrt left
TEttinger yeah, the old thinkpads were supposed to be amazing, my brother had one that I think was an IBM thinkpad, dropped it from arm to concrete multiple times while walking to class, around 2008-2012 era. it still runs 08:27
very slowly, but it still runs
timotimo my first one was a t23, it was fantastic 08:28
08:28 mcmillhj joined
timotimo i ran gentoo on it for no good reason 08:28
08:31 ChoHag joined
timotimo i'm replacing my mouse with one that can keep a drag going and usually only clicks once when i press the mouse button once! 08:32
Woodi I play CS1.6 on T41 fo years... legs on the table, laptop on legs, me on sofa :> too bad for me that 41 was last model with some part un-soldering from heat and non-rigid base :) 08:33
08:33 mcmillhj left
moritz just returned from a small cross-country skiing tour o/ 08:34
Woodi hi back moritz :) 08:35
moritz: nice blue picture you got :)
timotimo huh, you can't cross the country on skiis!
08:35 dakkar joined
moritz depends on the country :-) 08:36
08:36 Peter_R left
timotimo clearly there's only one The Country 08:36
moritz and it's blue? 08:37
08:37 Peter_R joined 08:38 zakharyas joined 08:40 mcmillhj joined
moritz perlgeek.de/images/view-beitostolen.jpg 08:40
timotimo now i get it 08:41
08:42 Wanderer68 joined
Woodi nice and blue but I wasn't talking about that - don't have spy software installed in moritz++ camera :) I means picture with origami butterfly and blue background :) 08:43
08:43 Cabanossi left
moritz Woodi: yes, I thought so, but couldn't resist :-) 08:44
Woodi to the topic: flow based programming could be relevant to hyper/race... it even use buffers like lizmat++ prototype... 08:45
08:45 mcmillhj left, Cabanossi joined 08:48 wamba left
lizmat Files=1180, Tests=55954, 192 wallclock secs (11.65 usr 4.55 sys + 1151.96 cusr 103.22 csys = 1271.38 CPU) 08:50
Woodi and as I understand Sponge approach it gets methods and seq and work on it in paraller but internally. that's one of the ways of working with data - prepare all data, define work, execute. it have adventage that it [can] know size of work. but I thinked hyper/race was more like adding paralleism to pipe, size of elements to transform is not know, you can only request more (maybe with buffering internally)
samcv can i get bisectable to run a code from a gist or something? anyone know
Woodi samcv: not me. but you have a gist ? you can try :) 08:51
samcv lol i suppose
bisectable6, gist.github.com/samcv/0df9fcc0c119...fbee6f24d7
bisectable6 samcv, It looks like a URL, but mime type is ‘text/html; charset=utf-8’ while I was expecting something with ‘text/plain’ or ‘perl’ in it. I can only understand raw links, sorry.
samcv bisectable6, gist.githubusercontent.com/samcv/0...e/concat.t
bisectable6 samcv, Successfully fetched the code from the provided URL.
Woodi yay
samcv \o/
bisectable6 samcv, Bisecting by exit signal (old=2015.12 new=43e0902). Old exit signal: 1 (SIGHUP) 08:52
08:52 TheovdH joined
samcv oh no 08:52
TEttinger ?
samcv will have to narrow the bisect i think
TEttinger 2015.12, heh
samcv as soon as it tells me when it stopped exit code 1
though it is a .t file so maybe it failed 1 test? or just totally was broken. idk 08:53
lemme check exit code
yeah i get exit code 1 currently. so maybe it's always been this way
bisectable6 samcv, bisect log: gist.github.com/a293e9e03bc844d693...afdd42fdf3
samcv, (2016-03-20) github.com/rakudo/rakudo/commit/f6...c131744403
samcv evalable6, gist.githubusercontent.com/samcv/0...e/concat.t 08:55
evalable6 samcv, Successfully fetched the code from the provided URL.
(exit code 1) ok 1 -
ok 2 -
ok 3 -
ok 4 -
ok 5 -
ok 6 -
ok 7 -
ok 8 -
ok 9 -
ok 1…
samcv, Full output: gist.github.com/f3dedf562375bb8517...16dbab7b72
samcv bisectable6, help gist.githubusercontent.com/samcv/0...e/concat.t
bisectable6 samcv, On both starting points (old=2015.12 new=43e0902) the exit code is 1 and the output is identical as well
samcv, Output on both points: «04===SORRY!04=== Error while compiling /tmp/IFjR0Uw1S0␤Confused␤at /tmp/IFjR0Uw1S0:1␤------> 03help https:08⏏04//gist.githubusercontent.com/samcv/0df9f␤ expecting any of:␤ colon pair»
samcv bisectable6, help 08:56
bisectable6 samcv, Like this: bisectable6: old=2015.12 new=HEAD exit 1 if (^∞).grep({ last })[5] // 0 == 4 # See wiki for more examples: github.com/perl6/whateverable/wiki/Bisectable
samcv bisectable6, old=2016.04 gist.githubusercontent.com/samcv/0...e/concat.t
bisectable6 samcv, Successfully fetched the code from the provided URL.
samcv, Bisecting by exit code (old=2016.04 new=43e0902). Old exit code: 0
samcv hopefully i was not the one who broke it 08:57
bisectable6 samcv, bisect log: gist.github.com/eced94a0890dd13e19...c6bdb8aaf9
samcv, (2017-03-22) github.com/rakudo/rakudo/commit/02...b23afc1f96
samcv heh i broke it :P
hehe
08:58 wamba joined 08:59 mcmillhj joined
Woodi samcv: don't worry :) 09:01
09:01 jraglin left 09:03 wamba left 09:04 mcmillhj left, nadim_ left
Woodi btw. I know it's very old school to have project code "tree" like in old days in Linux but I see a lot of merges with a lot of changes... maybe it would be good to slow down a little and having branches that are parallel to head for some time, for testing purposes ? 09:04
09:08 jonadab left 09:15 shdb left 09:17 shdb joined 09:18 cibs left 09:20 cibs joined 09:21 mcmillhj joined 09:26 mcmillhj left
samcv well this is just something we don't have test coverage yet. and as long as it's not close to the end of the month 09:29
change seemed sane and fine and i even added tests to support it. but today added some more tests for more exotic cases like korean text (three codepoints per visible character aka grapheme) and \r\n newlines 09:30
timotimo i really prefer short-lived branches over long-lived branches
samcv yes
09:30 aborazmeh joined, aborazmeh left, aborazmeh joined
samcv i use personal branches though 09:31
timotimo don't show anybody! :)
samcv then merge in or cherry-pick the commits into master
heh timotimo
09:32 mcmillhj joined
timotimo ugh, i have only looked at moar-gdb for a minute and a half and i'm already feeling like quitting 09:32
masak don't give up! 09:35
you can do this! :+1:
(hi, #perl6)
(good... antenoon)
09:36 movl left 09:37 mcmillhj left
lizmat m: dd ^10 .grep({ FIRST say "first"; LAST say "last"; $_ %% 2 }) # looks like FIRST phaser not working either :-( 09:40
camelia (0, 2, 4, 6, 8).Seq
lizmat m: dd ^10 .map({ FIRST say "first"; LAST say "last"; $_ if $_ %% 2 }) # map does 09:41
camelia first
last
(0, 2, 4, 6, 8).Seq
timotimo i now have written something in python that uses exceptions to cause a fall-back to be returned from a decorated function further up the stack
not sure if terrible or genius
09:43 Cabanossi left, tojo left
lizmat sometimes I wonder whether we shouldn't disallow phasers in grep altogether 09:43
afk&
09:45 Cabanossi joined 09:46 mtsd joined
masak lizmat: huh! phasers in grep... 09:48
lizmat: on the *one* hand, it seems consistent with control-flow keywords working in grep...
masak stops to think if there is a counterargument
lizmat: on the *other* hand, if it already works in .map, then that's at least an inconsistency, the kind we prefer to avoid ;) 09:50
so, I guess, either un-implement it in .map (with a sufficiently stern motivation as to why You Did Not Actually Want phasers in .map and .grep), or implement it in .grep (with a mixture of pride and fear that that even works) 09:51
007 chickens out on this one, big time. it just says "no" to anything lexotic.
see github.com/masak/007/issues/145#is...-232278144 09:52
09:53 TheovdH left, TheovdH joined, mr-foobar joined 09:54 mcmillhj joined 09:55 n1ce joined, Sound joined 09:57 lukaramu joined 09:59 mcmillhj left 10:07 bjz_ joined
spider-mario the release notes for rakudo 2017.03 say: 10:07
“It includes bugfixes and optimizations on top of the 2015.12 release of Rakudo, but no new features”
are they supposed to refer to 2015.12?
10:07 TEttinger left
spider-mario oh, right, that was the “official ‘1.0’ release”, wasn’t it? 10:08
(wow, already)
10:08 bjz left 10:09 mtsd left
timotimo right 10:10
10:11 aborazmeh left
Woodi samcv, timotimo: I was proposing there to cherry-pick or merge head into branch with new big feature and test it fer few days, maybe a week with normal development patches. and then merge into head. but that require that testing would be done by someone or automatically, without it there is no point in one more step on the way... 10:13
10:13 aborazmeh joined, aborazmeh left, aborazmeh joined
samcv well we have a full suite of tests in roast repository 10:13
but if it's not catched by that, often it can be catch by people using the dev build, then it's fixed before release and tests are added so it never breaks again 10:14
10:15 labster left 10:16 mcmillhj joined
Woodi samcv: I probably thinked that "catching" by ppls doing dev make one more temporal problem for them. but not saying everything is wrong now, no... 10:17
10:21 mcmillhj left 10:26 mxco86 left 10:29 parv left 10:31 bjz joined 10:32 bjz_ left 10:38 eythian_ is now known as eythian, mcmillhj joined 10:42 literal_ is now known as literal 10:43 mcmillhj left 10:54 yqt joined 10:57 Cabanossi left
samcv Woodi, turns out it wasn't me who broke it :P 10:58
i merged it though. so. i guess maybe i'm to blame too. idk lol
it is all okay though :) tests are now better
11:00 Cabanossi joined 11:01 mcmillhj joined 11:02 abraxxa joined 11:06 mcmillhj left 11:10 pmurias joined
pmurias hi 11:10
masak \o
11:12 mcmillhj joined
moritz o/ 11:16
samcv yes 11:18
11:19 mcmillhj left
samcv ok fix is now in MVM master will get somebody to bump nqp and mvm version for rakudo since i'm going to bed 11:20
11:22 xtreak left, bjz left 11:23 mcmillhj joined
moritz can do (I hope :-) 11:25
timotimo good night! 11:27
11:28 mcmillhj left 11:30 robertle left 11:43 bjz joined 11:45 robertle joined 11:47 mcmillhj joined 11:48 cibs left 11:50 cibs joined, duncan_dmg joined 11:51 wamba joined 11:53 mcmillhj left 12:00 mcmillhj joined 12:03 awwaiid left 12:04 m44st4 left 12:05 mcmillhj left 12:14 mcmillhj joined 12:17 llfourn left 12:19 mcmillhj left 12:28 mcmillhj joined 12:32 mcmillhj left 12:34 gdonald joined
[Coke] finds open.segment.com/fellowship (grant for open source developers) 12:36
12:36 kyan left 12:42 effbiai left 12:43 faraco joined 12:47 mcmillhj joined 13:00 aborazmeh left 13:01 awwaiid joined 13:10 sufrostico joined, sufrosti1o joined, sufrosti2o joined 13:20 ufobat left 13:22 skids joined 13:31 cschwenz joined 13:32 cibs left
cschwenz given that IO::Notification.watch-path() watches the given path (docs.perl6.org/type/Supply#method_...tch-path), how do you 1) list all of the watched paths, and 2) un-watch a path? 13:33
13:34 cibs joined 13:35 lukaramu_ joined
jnthn For 1 you'd have to track yourself; For 2, should be handled by closing the tap 13:36
cschwenz thanks jnthn :-) 13:37
jnthn (On 1, we don't have that info around to hand back, so we'd have to explicitly do so also)
13:39 lukaramu left 13:44 Cabanossi left, cibs left 13:45 Cabanossi joined 13:46 cibs joined 13:47 MilkmanDan left 13:48 MilkmanDan joined
timotimo it'd be cool if there were a way to watch a path recursively ... but at least linux doesn't offer that afaik 13:52
jnthn If recursive dir is for module-space then this surely is also :) 13:54
13:54 wdna left
timotimo fair enough 13:57
14:00 cschwenz left 14:03 wamba left 14:08 curan left 14:10 Ven joined, Ven is now known as Guest65550 14:11 Guest65550 is now known as Ven` 14:12 AlexDaniel left 14:13 sufrosti2o left, sufrostico left, sufrosti1o left 14:17 ChoHag left 14:19 sufrostico joined, ChoHag joined
pmurias why do we use 72 characters as the wrapping limit for commit messages? 14:21
Ven` git convention IIRC. 14:23
masak 80 - 4 spaces indent - 4 spaces for symmetry
[Coke] RT #131048 - this a Linenoise bug, I wonder? 14:24
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131048
masak Ruby has a new "absent" operator: medium.com/rubyinside/the-new-abse...3ef6cd0b99 -- but I struggle to understand what I would use it for 14:31
14:36 tokomer joined
[Coke] the main example seems to be "does not contain", but that seems better done for literals with a ! .contains 14:36
pmurias the example that matches /* */ style comments is the one that makes sense in the blog post 14:42
Ven` masak: it removes the need for some lookaheads 14:43
negative* lookaheads
14:46 nhywyll joined 14:47 nhywyll left 14:51 jonas1 left
pmurias Ven`: is it really useful for anything other than matching something ending with a multi character delimiter? 14:56
15:02 cibs left 15:03 cdg joined
Ven` pmurias: it's useful when you want to make sure e.g. a delimiter is not present 15:04
<<I want \d+ \d+ with something inbetween that isn't "ven">>
15:04 cibs joined 15:07 wamba joined
pmurias Ven`: how do you combine the absent operator with \d+? 15:08
Ven` \d+(?~ven)\d+
15:12 bjz left
b2gills m: say '123 ven 32' ~~ /(\d+) (.*) <!{ $1.contains('ven') }> (\d+) / 15:13
camelia ｢123｣
0 => ｢12｣
1 => ｢｣
2 => ｢3｣
b2gills It would take more work to get the same semantics as (?~ ) 15:14
Ven` ^ that's very very ugly and probably doesn't have very good performance due to backtracking (?) 15:15
b2gills m: #`(There is space available) say '123 ven 32' ~~ /(\d+) <- 'ven'> (\d+) / 15:18
camelia 5===SORRY!5===
Unrecognized regex metacharacter < (must be quoted to match literally)
at <tmp>:1
------> 3available) say '123 ven 32' ~~ /(\d+) <-7⏏5 'ven'> (\d+) /
Unable to parse regex; couldn't find final '/'
at <tmp>:1
------> …
15:34 cdg_ joined 15:35 tokomer left 15:37 cdg left 15:50 sufrostico left
perlpilot The implied * in (?~absent) bothers me for some reason. 15:53
15:57 Cabanossi left 15:58 Wcool joined
Wcool hi 15:58
faraco hi
Wcool Does anyone know how to assign STDOUT or STDIN to a (file) variable? 15:59
like perl5: my $fh_out = *STDOUT;
jnthn The handle is in the $*OUT varialbe 16:00
(And $*ERR and $*IN)
So my $fh_out = $*OUT
Wcool great, cheers
16:00 Cabanossi joined
faraco question, is this a good practice just for STDOUT? perl6 -e 'my $message = q:x/ls -a/; say $message' 16:01
I mean, TIMTOWTDI right? 16:02
16:02 tokomer joined
gfldex faraco: &say will be a bad idea in many cases 16:04
16:09 mcmillhj left 16:10 jraglin joined
faraco can I see an example, if you don't mind. 16:10
16:11 Wcool left 16:12 nhywyll joined 16:14 mcmillhj joined
lucs Holy cow! JH Conway will be in Montreal to give a conference in a couple of weeks. 16:16
lucs won't miss it
DrForr loos at his Math::Game module :) 16:17
16:19 mcmillhj left
[Coke] uses perl 6 to safely truncate a utf8 string to fit into a 4000 byte varchar without having to worry about cutting a character into pieces. 16:23
DrForr senses mySQL. 16:25
16:27 mcmillhj joined
mst as if a million developers cried out, and were suddenly silently trunca 16:29
16:32 mcmillhj left, araujo left 16:33 mcmillhj joined
[Coke] DrForr: Oracle 16:33
I was truncating to 4000 chars at first, but that failed because it's 4000 bytes. 16:34
DrForr mst++
16:37 tokomer left 16:38 mcmillhj left 16:40 zakharyas left 16:44 domidumont left, araujo joined 16:50 mcmillhj joined 16:55 Sound left, mcmillhj left, thowe_work joined 16:58 Cabanossi left 17:00 abraxxa left, Cabanossi joined 17:02 mcmillhj joined 17:08 xzhao left
gfldex faraco: say will truncate lists at 100 elements or output whatever an overloaded .gist returns on the object you call say with 17:10
17:13 xzhao joined, xzhao left 17:15 xzhao joined, kurahaupo__ joined 17:16 m44st4 joined
m44st4 Hi ! I'm currently trying to use socket in perl6, but i'm in front of a problem: when i use $socket.recv, i only receive the first 12 chars, i have to $socket.recv again to receive the rest. Does anyone have an idea about that? 17:17
17:19 rindolf left
[Coke] m44st4: are you passing any args to the .recv call? 17:25
m44st4 i tried without and with a length
but it looks like it stopped receive after the first whitespace
geekosaur if it's non-blocking, as libuv's sockets almost certainly are, all bets are off as to how many reads / how much data 17:26
if you were trying to rely on read lengths, on a stream socket, you are doing it wrong
m44st4 i don't understand what you mean 17:28
i thouth recv just receive data until \n
geekosaur I would expect it to mimic the C recv() call
which just does binary data, it cares not about newlines 17:29
m44st4 hmm ok
but i did not understand the reason why it stop receive 17:30
17:31 rindolf joined
geekosaur because there is no guarantee that (a) the sending kernel (b) the receiving kernel (c) any network equipment between them will retain any notion of how it was sent 17:32
if you write out a line all at once in the sender, it's entirely possible for each character to be received in its own packet and potentially a separate recv() call needed for each.
m44st4 ok! that's clear :) 17:33
thanks
geekosaur if you need packets to be retained as packets, use UDP. (but I don't think libuv supports UDP sockets yet, so rakudo doesn')
you might also just want to use normal handle-based I/O and .lines
17:37 duncan_dmg left 17:41 raschipi joined
jnthn There is basic UDP support in IO::Socket::Async, fwiw :) 17:42
(And libuv does support UDP)
geekosaur heh 17:43
mst takes a moment to bemoan the absence of unix sockets
geekosaur also, while it's possible for that case I mentioned to happen, I'd probably be taking a close look at the network equipment if it actually did that 17:44
17:44 Cabanossi left 17:45 Cabanossi joined
geekosaur (that doesn't mean yours indicates a problem; lots of stuff can happen. 12 octets + header makes a bit more sense for a packet than 1 octet + header) 17:45
17:49 andrzejku joined 17:50 zenon joined
zenon p6: say "生"~&"死"; 17:51
camelia 愛
17:58 cpage left 17:59 cpage joined 18:00 dakkar left, n1ce left 18:01 sufrostico joined 18:02 n1ce joined
faraco p6: say "生"~&"死".ords 18:09
camelia
18:11 cdg_ left 18:12 faraco left 18:16 mcmillhj left, Wanderer68 left 18:18 mcmillhj joined 18:22 cibs left 18:23 mcmillhj left, gdonald left 18:24 mcmillhj joined, AlexDaniel joined, cibs joined
Geth doc: 787882b6ab | (Zoffix Znet)++ | doc/Language/5to6-perlfunc.pod6
Update `pipe` info

IO::Pipe does not offer what Perl 5's `pipe` routine does.		 18:27
18:30 AlexDaniel left
Geth doc: 61ef0f6a4e | (Zoffix Znet)++ | doc/Language/5to6-perlfunc.pod6
s/suttle/shuttle
 18:30
18:33 Ven` left, domidumont joined, Sound joined 18:35 raschipi left
samcv can somebody tell me why i get this error? gist.github.com/samcv/0df9fcc0c119...fbee6f24d7 18:35
Variable '@ords' is not declared. line 29
but it is declared. 18:36
geekosaur in perl 5 it'd often mean a missing semicolon before it 18:37
18:38 eroux joined
samcv maybe you can dl and tell me what's going on. it's happened other times, and i've done really weird things to make it go away like restructuring all the code 18:39
this could be different than that, i really have no clue why
18:41 nhywyll left
geekosaur oh 18:41
line 23
you're missing the replacement part from the S:g
(boy is that a lousy error to report for that)
18:43 Cabanossi left 18:45 Cabanossi joined
andrzejku hi:) 18:47
I find a leak in Perl6 probably documentation leak 18:48
18:52 Tonik joined 18:53 cibs left 18:55 cibs joined 19:00 cibs left 19:01 cibs joined
Geth doc: 69d32dab7a | (Zoffix Znet)++ | doc/Type/IO/Handle.pod6
[io grant] Remove IO::Handle.z

IO::Handle never had this method.		 19:06
samcv thanks Geth
geekosaur,
19:06 domidumont left 19:08 FROGGS joined 19:15 Tonik left 19:17 TEttinger joined, sufrostico left
Geth doc: d02ae7d5b2 | (Zoffix Znet)++ | doc/Type/IO/Handle.pod6
[io grant] Remove IO::Handle.rw and .rwx

IO::Handle never had these methods		 19:18
19:21 ugjka is now known as thasia
samcv ty much geekosaur 19:21
19:21 darutoko left 19:22 thasia is now known as ugjka
geekosaur andrzejku, details? 19:22
andrzejku geekosaur I need a secret document how to smuggle Perl6 in your company 19:23
I think many people just don't know how to
geekosaur generally you just start writing stuff in it, preferably to show coworkers how you'd easily solve some problem that's harder or requires more work in other languages 19:24
andrzejku geekosaur personally I know one way which I practice with Perl that was "write Perl script" every time when you need write somekind of script and commit it silenty
I commit this way some good scripts before my coworker see it
geekosaur I don't like that way because it gets you in trouble when someone else has to change it and has no idea what is going on
you really need to *sell* it, not try to sneak it in; you're just antagonizing your coworkers otherwise 19:25
timotimo oh samcv is already back :)
samcv :O
yes
andrzejku geekosaur ye but right now I am the worst case, my coworkers doesn't like perl they don't know perl6 and there no official perl6 compiler on production servers 19:26
geekosaur [Coke]'s example is a good one, most common languages will have problems of one kind or another with truncating by whole Unicode characters instead of by bytes
(assuming utf8 representation)
andrzejku geekosaur I think working underground is the best thing
commit perl everywhere you can 19:27
before your manager admit
geekosaur until someone gets back at you by committing idris code and making you maintain it :)
andrzejku or even more agressive way, rewrite existing scripts for refactoring or fixing bugs and commit perl
geekosaur or cobol
andrzejku that's no problem I can maintain perl it is even better 19:28
the problems become when people ask this "Can I write this in Perl?"
and the people say: "NO"
so I think working underground is the best thing which we can do for Perl 19:29
become Perl6 partisanen
19:29 setty1 left, jdmmmmm joined 19:30 jdmmmmm left
[Coke] I think if you do that "underground", you're going to end up alienating more people against perl. 19:31
Using any perl here, I end up using it for things have a high time pressure on me, and usually an implicit one-use. if I need to use it for something longer laster, I make sure I get buyin on the technology side that I'm going to use a language that no one else here is necessarily an expert on because of the time issue. If you want longer term adoption, you're going to need buyin. if not, they'll just be cur 19:33
sing your name as they rip our your code.
geekosaur that exactly, yes 19:34
that's why I said you should be selling it, not sneaking it
*show* how you can do something more easily in perl 6. otherwise you annoy everyone else and potentially get fired for not being a team player or something 19:35
andrzejku geekosaur are you from USA? 19:37
geekosaur yes
andrzejku geekosaur so you don't know how IT look in my country
geekosaur but you can get in trouble by annoying coworkers regardless of country
19:37 lichtkind joined
geekosaur maybe not get fired for it, but still, people will find ways to get back 19:37
[Coke] Yup. do what you will, we're just giving you our advice.
andrzejku in my country IT is pretty different 19:38
we earn shit here
we do all things for USA/Germany etc. etc.
so the people in projects change very quickly
and the most companies here is like that 19:39
we just do cheap it work for you
so perl6 here didn't exist and it is hard to talk with someone about it
[Coke] perl6 doesn't exist anywhere. :)
we're not that different in that regard, at least. 19:40
lichtkind in my mind it exists :)
andrzejku many US companies come here because they want to rewrite Perl into another language fast and cheap
so it is pretty different situation
and more difficult though 19:41
the rich US programmers will never understand poor programmers 19:43
19:57 tinita_ left
samcv andrzejku, what country? 19:59
19:59 Cabanossi left 20:00 Cabanossi joined
andrzejku samcv Poland 20:01
samcv andrzejku, on the internet we are all equals :) 20:02
andrzejku we pretend to be equal
lizmat
.oO( eating our own dogfood if we're lucky )
 20:03
20:06 cdg joined
andrzejku I think if the one country will go into it independent way 20:07
then I am pretty sure the Perl Golden Era will come
20:13 atweiden-air joined
mst geekosaur: I would love an excuse to work on idris for money 20:13
20:14 labster joined
geekosaur you would, I probably would, most other folks likely to give it the side-eye :) 20:14
atweiden-air the method `make` docs say: "Sets the AST to $ast, and returns it." 20:15
can/should a grammar-actions method like `method abc($/) { make ~$/ }` be `is pure`?
lichtkind hai lizmat (h)
lizmat lichtkind o/
lichtkind andrzejku: what exactly ou think rich programmer dont understand?
atweiden-air jw for code optimization's sake
lichtkind :)
timotimo i dont think we will ever optimize "is pure" methods 20:16
znless they are private methods
but the only thing "is pure" currently will do is execute the function at compile time and substitute its result in place of the call 20:17
so make is definitely not something that would wotk with that in general
atweiden-air timotimo: right, ok 20:18
about use of `make` in grammar-actions method... is it mutating the AST behind the scenes, or are the methods like `method abc($/) {...}` pure functions?
timotimo yiu mutate the ast attribute on the match obj passed in via $/ 20:20
20:20 bjz joined, pecastro joined
timotimo i.e. make 1 is the same as $/.make: 1 20:20
20:21 Sound left
timotimo which behaves as if it were $/.ast = 1 20:21
atweiden-air so what should be the return type of a method that ends with a `make`?
i've experimented and it appears to be returning a container 20:22
timotimo shrug :)
RabidGravy anyone know enough abiout SSL to see if there is something *funny* about httpbin.org/ ?
timotimo put --> Nil in the sig
that makes it easy :D
also better performance, though the overhead of the grammar engine is immense 20:23
atweiden-air timotimo: ix.io/pd2
timotimo "returns" is going to go awqy 20:24
away
please use -->
20:24 espadrine joined
atweiden-air with --> the return type does not make any difference 20:25
timotimo: why retire `returns`?
RabidGravy I do hope someone is going to send me 65 pull requests or there will be a lot of broken
timotimo heh. 20:26
i can have a try
you can ask TT why. i dont remember the exact reason except parsing makes it a bad thing 20:27
20:29 setty1 joined
timotimo like with "returns $foo or so? 20:29
ill drive home now
so no irc for a bit 20:30
20:30 cibs left 20:32 cibs joined
RabidGravy ah openssl s_client -connect httpbin.org:443 20:38
CONNECTED(00000003)
139647781410680:error:14077438:SSL routines:SSL23_GET_SERVER_HELLO:tlsv1 alert internal error:s23_clnt.c:769:
gfldex is zef creating the wrappers in /share/perl6/site/bin/ ? 20:43
20:44 Cabanossi left 20:45 Cabanossi joined 20:50 AlexDaniel joined 20:53 lichtkind_ joined 20:56 lichtkind left 21:00 skids left 21:04 atweiden-air left 21:05 araujo left, araujo joined 21:09 mcmillhj left 21:10 FROGGS left
AlexDaniel committable6: why are you still here? 21:12
committable6 AlexDaniel, ¦why: «Cannot find this revision (did you mean “all”?)»
21:13 Cabanossi left 21:15 sufrostico joined, Cabanossi joined 21:21 bjz left 21:24 cibs left, n1ce left, n1ce_ joined 21:26 cibs joined 21:37 rburkholder joined 21:38 setty1 left 21:43 Cabanossi left 21:44 andrzejku left 21:45 Cabanossi joined 21:47 cibs left 21:48 cibs joined 21:50 eroux left 21:51 eroux joined
gfldex lolibloggedalittle: gfldex.wordpress.com/2017/03/23/fo...nvenience/ 21:54
jferrero m: my Str enum Mass (uno => "3", dos => "2"); say dos.value 22:03
camelia 2
jferrero m: Str enum Mass (uno => "3", dos => "2"); say dos.value 22:04
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3Str7⏏5 enum Mass (uno => "3", dos => "2"); say
expecting any of:
infix
infix stopper
statement end
statement m…
22:06 cibs left
gfldex jferrero: the type of an enum value is infered from the first value of the first pair 22:06
22:07 cibs joined
jferrero m: use NativeCall; enum Mass (uno => int32(1), dos => 2); say dos.value 22:07
camelia ===SORRY!===
Cannot invoke this object (REPR: P6int; int32)
jferrero thanks, gfldex 22:08
22:10 n1ce_ left
jferrero gfldex: I need to defined int32 NativeCall type for enum constants. Is possible? 22:10
22:11 mcmillhj joined 22:12 RabidGravy left, Cabanossi left
gfldex I have very little knowledge about NativeCall 22:13
jferrero: if I would want to change that I would go and read other ppls code 22:14
22:15 Cabanossi joined 22:16 mcmillhj left
jferrero From enum docs: "As value types Numerical and Str are supported." So... not. 22:16
AlexDaniel jferrero: well, it does not mention natives :) 22:17
22:19 mcmillhj joined
geekosaur iirc it is intended that that will change someday but currently it's not possible to handle anything else? 22:20
22:25 pytuger joined 22:27 mcmillhj left, n1ce joined, wamba left 22:30 espadrine left 22:31 n1ce_ joined 22:32 pytuger left, committable6 left, benchable6 joined, committable6 joined, evalable6 joined, ChanServ sets mode: +v benchable6, ChanServ sets mode: +v committable6, ChanServ sets mode: +v evalable6, bisectable6 joined, unicodable6 joined, ChanServ sets mode: +v bisectable6, ChanServ sets mode: +v unicodable6, statisfiable6 joined, ChanServ sets mode: +v statisfiable6 22:33 statisfiable6 left, unicodable6 left, benchable6 left, bisectable6 left, evalable6 left, committable6 left, n1ce left 22:36 committable6 joined, ChanServ sets mode: +v committable6, bisectable6 joined, benchable6 joined, unicodable6 joined, ChanServ sets mode: +v bisectable6, evalable6 joined, ChanServ sets mode: +v benchable6, ChanServ sets mode: +v unicodable6, ChanServ sets mode: +v evalable6, statisfiable6 joined, ChanServ sets mode: +v statisfiable6
Geth routine-map: 21daef5c27 | (Zoffix Znet)++ | 2 files
Fix incorrect report of `.file`

  ... for routines whose proto is not in the same file as the routine
itself.		 22:36
timotimo good catch
Geth routine-map: 98c45fba9c | (Zoffix Znet)++ | 2 files
Include compiler name and version on HTML page
 22:37
22:37 bloatable6 joined, ChanServ sets mode: +v bloatable6 22:41 Voldenet left 22:42 riatre_ is now known as riatre 22:47 Voldenet joined, Voldenet left, Voldenet joined 22:49 pytuger joined 22:50 BenGoldberg joined 22:51 pytuger left, pytuger joined 22:56 pytuger left, Sound joined 23:08 pytuger joined, pytuger left, pytuger joined 23:14 skids joined 23:17 zenon left 23:24 llfourn joined 23:26 cdg left 23:28 TheovdH left 23:31 rindolf left, cibs left 23:33 cibs joined 23:35 pytuger left 23:36 rburkholder left 23:38 cibs left 23:40 cibs joined 23:43 rindolf joined 23:46 lukaramu_ left 23:50 pytuger joined 23:53 yqt left 23:55 pmurias left 23:58 adu joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!