»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.
jferrero or xperl7 00:03
00:04 pecastro left, dct left 00:06 levex joined, levex is now known as Guest6493 00:15 pmurias left 00:25 markong left 00:33 mempko joined, benjisomething joined
benjisomething Is there a better way for updating rakudo perl6 than just deleting my old directory and cloning the newer updated one (and proceeding to reconfiguring/building) 00:34
I'm not a fan of using rakudobrew
lookatme of course, there are some pre-build package on the github 00:36
benjisomething where can I get those from? 00:37
lookatme If you are using Linux, here you go: github.com/nxadm/rakudo-pkg 00:38
00:39 imcsk8 left
benjisomething I get this "/opt/rakudo-pkg/bin/perl6: 2: exec: /opt/rakudo-pkg/bin/moar: not found" 00:43
00:43 jberger left
benjisomething yet after checking, both of the binaries are there 00:44
lookatme have you update your PATH ? 00:45
or other things of PATH about Perl6 00:46
benjisomething yeah, I did
export PATH=$PATH:/root/.perl6/bin:/opt/rakudo-pkg/bin
00:47 psch joined
benjisomething :\ 00:51
lookatme :) I don't know 00:53
kraih leont: super efficient message passing between processes and nodes, what the erlang vm is famous for
benjisomething oh wow, I made a dumb error
kraih if rakudo could actually replicate that it would open up a very interesting niche
benjisomething I just had the wrong architecture 00:54
fixed now :)
lookatme yeah, if you have done a fresh installation it should be work
00:54 espadrine left
lookatme without any other setting except zef 00:55
also and the PATH
00:55 aborazmeh joined, aborazmeh left, aborazmeh joined, AlexDaniel joined
Geth perl6-most-wanted: 4fa96a646a | (Zoffix Znet)++ (committed using GitHub Web editor) | most-wanted/modules.md
Change CGI::Session to Data::Session

  irclog.perlgeek.de/perl6-dev/2018-...i_15703491
00:58
perl6-most-wanted: ed52b5256f | (Zoffix Znet)++ (committed using GitHub Web editor) | most-wanted/modules.md
Add back CGI::Session::Auth

  irclog.perlgeek.de/perl6-dev/2018-...i_15703573
01:02
benjisomething how would I remove all of certain types of characters from an array of strings 01:05
what would be the most efficient way
01:06 imcsk8 joined
lookatme for ex ? 01:06
benjisomething ? 01:07
timotimo what kind of "types" are we talking about? 01:08
lookatme I mean can you show some example ? 01:09
timotimo if it's a unicode property, you can use uniprop (be aware that it operates on the first character)
m: say uniprop "h"
camelia Ll
benjisomething any characters in a character array/string 01:10
lookatme m: say uniprop "h1"
camelia Ll
benjisomething will be special characters like "'/\$-=,.
lookatme m: say uniprop "h-"
camelia Ll
timotimo oh, you have an array of characters and you want to use that to filter out some stuff?
benjisomething yeah
timotimo i believe we just recently had a stackoverflow q/a for that
stackoverflow.com/questions/473290...-in-perl-6 - not quite what you wanted 01:12
benjisomething why doesn't $filtered .= trans($ignore.comb X=> ''); work
timotimo i think for the trans you want you'd have to have a pair of two arrays? 01:13
benjisomething hm
wait, I think it works for everything except for `'` and `"` 01:14
timotimo m: my $source = "hello how are you"; my $ignore = "abcdefgh"; say $source.trans($ignore => "") 01:15
camelia llo ow r you
timotimo benjisomething: this seems to work fine
benjisomething OH I know why
There are separate characters for `“` and `"`
woops, yeah this works
psch this reminds me that tr/// and .trans both feel weird with :d 01:16
01:17 unicodable6 left, quotable6 left, unicodable6 joined
geekosaur "“” 01:17
3 of them
01:17 quotable6 joined, ChanServ sets mode: +v quotable6
timotimo not to forget the „ that german has 01:18
geekosaur yeh
and single quote versions
timotimo yeah, ‚‘’ 01:19
.u ,‚
yoleaux U+002C COMMA [Po] (,)
U+201A SINGLE LOW-9 QUOTATION MARK [Ps] (‚)
Xliff timotimo: Yes, that was the first thing I looked at. I don't think there are UI elements in it yet.
timotimo can you tell them apart with your font? i can't
Xliff: it's true :(
geekosaur barely 01:20
Xliff timotimo: I would like to do that, but I need something quick and dirty for another project I am working on.
benjisomething can I check for newlines just by using contains "\n"?
psch m: $_ = "foo"; say tr:d/o/o/ # ISTR :d should ignore the replacement..?
camelia StrDistance.new(before => "foo", after => "foo")
Xliff Maybe fork Curses::UI for Terminal::Print?
timotimo m: say "foo\r\n".contains("\n")
camelia False
timotimo m: say "foo\r\n".contains("\r\n")
camelia True
timotimo benjisomething: you cannot 01:21
benjisomething thanks, lol
isn't \r\n the carriage thing for windows?
psch m: say so "foo\r\n" ~~ /\n/ 01:22
camelia True
psch heh
m: say so "foo\r\n" ~~ /o\n/
camelia True
psch *that* was my point :x
benjisomething still isn't finding newlines for some reason 01:23
;-;
lookatme m: say "foo\r\n".codes 01:25
camelia 5
benjisomething what does `.codes` do
psch returns the number of unicode codepoints 01:26
says docs.perl6.org/routine/codes at least :)
benjisomething mm
b2gills \r\n gets turned into a synthetic 01:27
TimToady m: say "foo\r\n".chars
camelia 4
benjisomething timotimo: Your example screenshots for modules.perl6.org/dist/SDL2::Raw:cpan:TIMOTIMO aren't there anymore
geekosaur perl 6 normally works in terms of "graphemes" (things that *look* like a character). some of these require combining codepoints, so a single Perl 6 "character" may be represented by more than one codepoint
psch m: say "foo\r\n" ~~ /o\xd/ 01:28
camelia Nil
TimToady and note that \n is special in regex, and matches any common newline sequence
psch m: say so "\r" ~~ /\n/ 01:29
camelia Cannot allocate memory
psch uh
TimToady coolio!
benjisomething So how do I find all newlines lol
TimToady a very bug
b2gills m: say so "\r" ~~ /\n/
camelia True
TimToady was there anything different? or did we just see a resource failure? 01:30
benjisomething doesn't look different at all
I tried it a few time on my machine and it returns true every time
timotimo benjisomething: damn, i'm not sure how to make those links proper 01:31
modules.perl6.org/dist/SDL2::Raw:cpan:TIMOTIMO/ - if you use this URL, i.e. with a / at the end, you can click the links in the readme
benjisomething Hey timo
I've made a few raytracers, and I've tried gd/cairo/sdl 01:32
What do you think is optimal for raytracers?
b2gills m: (1..9000).grep({.chr ~~ /\n/}).map: *.uniname.say # all characters that /\n/ matches
camelia <control-000A>
<control-000B>
<control-000C>
<control-000D>
<control-0085>
LINE SEPARATOR
PARAGRAPH SEPARATOR
benjisomething I'd say probably something that just outputs a simple bitmap
timotimo you can get a bitmap on the screen somewhat easily with SDL2. cairo is probably a bit too much, because you can't get a window to display stuff with unless you also use gtk or something similar
and i think gd also won't open a window for you
benjisomething I tried to make a library for exporting bitmaps from direct pixel colors in julia 01:33
but it was a hassle
which do you think is the fastest?
I always have to thread my raytracers like crazy for them to do it in a timely manner
timotimo see the white noise example, it shows how you can access an array of pixel data to show something on the screen
benjisomething alright, ty
timotimo it uses the RGB332 pixelformat so every byte is one pixel and it still has color 01:34
but you can choose something bigger (if you adjust the pitch calculation as commented out)
like ARGB8888 01:35
benjisomething is RGB332 standard 1-255 color
timotimo i don't think so, no 01:36
benjisomething hm
oh, 8-bit has a max of 256 colors? 01:37
I think once you get to 16-bit it goes up to 65535 or whatever that number is 01:38
which is the R1-255 G1-255 B1-255
timotimo yup, that's correct
er, no, wait 01:39
geekosaur generally, yes. but RGB332 means 3 bits for the red channel, 3 for green, 2 for blue. usually you want to use 24 or 32 bit color so you get 0-255 for each color channel (and optionally 0-255 for transparency)
benjisomething ah, yeah, that's what I was thinking
hm
I could make a sweet terminal raytracer that outputs to ncurses lol
timotimo: Can you easily save an sdl surface/screen/whatever to a png/bmp 01:42
I seem to remember you could with png
timotimo probably just need to bind this function: wiki.libsdl.org/SDL_SaveBMP
benjisomething is that already included in SDL2::Raw? 01:43
timotimo don't think so, but it's trivial to add 01:45
though actually 01:46
what you have there is a texture, but you need a surface for that. so you'll also need a conversion function
benjisomething ;-;
sounds annoying
timotimo and there's a CreateTextureFromSurface, but i don't see one for the other way around 01:47
fret not, though. you can CreateSurfaceFrom raw pixel data
wiki.libsdl.org/SDL_CreateRGBSurfaceFrom
benjisomething is there any simple library that can output pixel data to bitmaps?
that would be pretty good
timotimo there's a library that outputs pngs that are not compressed
benjisomething ooh
I wonder how speedy that could be
timotimo github.com/cygx/p6-image-png-inflated 01:48
benjisomething do you think it could work well with raytracers?
timotimo probably. without SDL you probably won't have live preview, but you can just copy image data to a file every few seconds and just refresh that in your image viewer 01:49
benjisomething hmm, that could be cool
and then I could even piece together png frames with ffmpeg
timotimo sure
benjisomething you have a blog right?
I think i asked you months ago
under a different username lol
timotimo i used to blog the perl6weekly, but liz does that almost exclusively since almost a year 01:50
benjisomething ah, so no personal blog then I guess
timotimo i don't have a personal blog or something, but i'll set one up in at least a month or two
benjisomething I see
timotimo the png inflated module is probably not performance-tuned yet, but you can copy the pixel data and save them to disk on a thread with start { ... } 01:51
gotta go to bed now, seeya and good luck! 01:52
benjisomething night night
timotimo github.com/raydiak/pray - if you want to, you can take inspiration from Pray, another perl6 ray tracer. it's not been touched in a long time, though, so might not work right away without some minor changes
01:55 Matthew[m] left 01:56 Juerd joined, salva joined, robinsmidsrod joined, Altreus joined, dustinm` joined, BuildTheRobots joined, mienaikage joined 01:57 evalable6 left, coverable6 left, rgrau left, committable6 left 01:58 Matthew[m] joined, unicodable6 left 02:00 rgrau joined
benjisomething Can someone dumb this down for me `blob8.new(^256 .flatmap: { $_ xx 3, 255 });` 02:17
psch m: say ^256 .flatmap: { $_ xx 3, 255 } 02:18
camelia (0 0 0 255 1 1 1 255 2 2 2 255 3 3 3 255 4 4 4 255 5 5 5 255 6 6 6 255 7 7 7 255 8 8 8 255 9 9 9 255 10 10 10 255 11 11 11 255 12 12 12 255 13 13 13 255 14 14 14 255 15 15 15 255 16 16 16 255 17 17 17 255 18 18 18 255 19 19 19 255 20 20 20 255 21 21 2…
AlexDaniel m: say "foo\r\n" ~~ /o\xd/
camelia Nil
psch benjisomething: looks like "all greys in RGBA" to me
AlexDaniel m: say "foo\r\n" ~~ /o\xd/
camelia Nil
psch m: say ^256 .map: { $_ xx 3, 255 } 02:19
camelia (((0 0 0) 255) ((1 1 1) 255) ((2 2 2) 255) ((3 3 3) 255) ((4 4 4) 255) ((5 5 5) 255) ((6 6 6) 255) ((7 7 7) 255) ((8 8 8) 255) ((9 9 9) 255) ((10 10 10) 255) ((11 11 11) 255) ((12 12 12) 255) ((13 13 13) 255) ((14 14 14) 255) ((15 15 15) 255) ((16 16 …
AlexDaniel yea it's .map(…).flat I think
psch ah, both sides flat
AlexDaniel flatmap discussion: rt.perl.org/Ticket/Display.html?id=130520 02:20
psch AlexDaniel++
geekosaur there is one other cleverness in there, the space before the .flatmap makes flatmap method be called on the Range object produced by ^256 instead of on 256. (I'm not actually fond of that particular precedence hack.) 02:21
psch yeah, i'm a bit iffy about the precedence change for <methodop> with a magic invisible unspace 02:22
i think we already had that before i ducked out though and i didn't complain enough then probably so eh :) 02:23
geekosaur I thought it wasn't a change, just a "clever" use of what already existed
but I strongly prefer readable over "clever", so magic spaces annoy me a bit 02:24
psch m: say ^256\ .flatmap: $_ 02:25
camelia Cannot resolve caller flatmap(Int: Any); none of these signatures match:
($: &block, :$label, *%_)
in block <unit> at <tmp> line 1
psch m: say ^256\ .flatmap: { $_ }
camelia ^1
psch m: say ^256 \ .flatmap: { $_ }
camelia (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86…
psch m: say ^256.flatmap: { $_ }
camelia Potential difficulties:
Precedence of ^ is looser than method call; please parenthesize
at <tmp>:1
------> 3say ^2567⏏5.flatmap: { $_ }
^1
AlexDaniel m: !
camelia 5===SORRY!5=== Error while compiling <tmp>
Prefix ! requires an argument, but no valid term found
at <tmp>:1
------> 3!7⏏5<EOL>
expecting any of:
prefix
benjisomething I much prefer readable over clever 02:26
but then again, I'm not great at reading any perl6 lol
02:26 cdg left
psch geekosaur: i think we got "space before <methodop> changes its precedence" something like two years ago or so? ISTR the reason being method chaining with line breaks..? 02:27
geekosaur hm, possibly
psch can't find a relevant seeming commint right now though
geekosaur I thought that was done differently 02:28
psch m: "foo" . say
camelia foo
geekosaur actually I thought that was not precedence, it was a straight-up syntax error before
02:28 cdg joined
geekosaur . for method call didn't accept *any* whitespace 02:29
you had to use unspace to split it over a line before that
02:33 cdg left, cdg joined 02:37 mempko left, cdg left
AlexDaniel I guess unspace works better anyway? 02:44
because you will not run into precedence differences
02:46 ilbot3 left
psch geekosaur: there was the perl5-ism warning i think? 02:48
m: say "foo" . "bar" # this one 02:49
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of . to concatenate strings; in Perl 6 please use ~
at <tmp>:1
------> 3say "foo" .7⏏5 "bar" # this one
psch geekosaur: but yes, i think it is actually two <methodop>s instead of one with multiple precedences 02:50
geekosaur: so you're right, it was added :) 02:51
geekosaur it's worse than that. iirc foo.bar is parsed differently to start with (the . is not an operator)
but that wasn't the change, it was always that way (one parsed directly, the other an operator)
02:56 ilbot3 joined, ChanServ sets mode: +v ilbot3, jeromelanteri joined 03:05 ponbiki left 03:18 eliasr left 03:21 troys_ is now known as troys 03:30 sjn left, kst left, tinita left, mspo joined, domm joined, kst joined, sjn joined 03:31 clkao joined, olinkl joined 03:32 jhill_ joined, rgrau left 03:33 squashable6 left, bloatable6 left, squashable6 joined 03:35 jdv79 joined, tinita joined
mspo u: a 03:35
03:37 rgrau joined, Actualey` left 03:46 Kaiepi left 03:47 Kaiepi joined 03:57 bloatable6 joined 03:58 unicodable6 joined
AlexDaniel u: a 03:58
unicodable6 AlexDaniel, U+000A <control-000A> [Cc] (control character)
AlexDaniel, U+001A <control-001A> [Cc] (control character)
AlexDaniel, Cowardly refusing to gist more than 5000 lines
AlexDaniel mspo: heh well, that seems to be “a” :)
mspo u: 04:02
unicodable6 mspo, Found nothing!
04:03 jcallen left 04:11 astj_ left, astj joined 04:13 astj left, astj joined 04:14 mtj_ joined 04:20 jcallen joined 04:34 Cabanossi left, dj_goku left 04:37 Cabanossi joined 04:42 benjisomething left 04:43 troys is now known as troys_ 04:44 vichib joined
vichib is there tk bindings for perl6? 04:48
stmuk vichib: I think people tend to use GTK today 04:56
05:07 khw left 05:25 aborazmeh left 05:27 cdg joined 05:29 wamba joined 05:32 cdg left 05:38 bulat_ joined 05:39 mempko joined
bulat_ Hi. 05:39
How do I print to stderr using say?
Thansk.
b2gills m: { my $*OUT = $*ERR; say 'foo' } 05:40
camelia foo
b2gills or just use `note` instead of `say`
AlexDaniel m: $*ERR.say: 42
camelia 42
bulat_ @AlexDaniel, how do I pass multiple args to $*ERR? 05:44
AlexDaniel not entirely sure what you mean by that 05:45
if you want to call say with multiple arguments, then sure you can do it
m: $*ERR.say: 42, 50, 80
camelia 425080
bulat_ Why isn't it possible to call $*ERR.say 42, 50, 80; 05:46
I tried, it errs.
AlexDaniel because that's a method call
ideally you'd write it as
m: $*ERR.say(42, 50, 80)
camelia 425080
AlexDaniel but I like fancy stuff [and don't like parens] so I use : instead ☺
bulat_ So usual "say" is ordinary function? 05:47
AlexDaniel yes
bulat_ Thank you.
05:48 imcsk8 left
AlexDaniel bulat_: fwiw, if you want to use something similar in a function form, then just use `note` as suggested above 05:48
m: note 42, 50, 80
camelia 425080
AlexDaniel it's identical to say except that it prints to $*ERR
05:50 squashable6 left, statisfiable6 left, player joined, committable6 joined, squashable6 joined, statisfiable6 joined, ChanServ sets mode: +v statisfiable6
moritz m: $*ERR.say: 42, 50, 80 06:02
camelia 425080
player Why say so 'a b c' ~~ m:s/a b c/ gives true? 06:07
06:09 nige joined
nige just wanted to pop in to make a suggestion - regarding Perl 5 and Perl 6 06:10
yoleaux 17 Dec 2017 21:50Z <b2gills> nige: There is no entry for any version of Perl other than 1.0 en.wikipedia.org/wiki/Timeline_of_..._languages
06:11 Zoffix joined
nige ^ - thanks b2gills I think lizmat++ sorted that 06:11
Zoffix player: because `:s` makes whitespace significant, not literal. The whitespace after atoms in regex becomes a <.ws> call that can match a bunch of whitespace. It's like using `rule` instead of `token` inside grammars. 06:12
nige I quite like the way the Apache foundation handles its branding 06:13
Zoffix Relevant docs: docs.perl6.org/syntax/:sigspace docs.perl6.org/language/grammars#i...ntry-ws-ws 06:14
nige "Apache" used to be a web server - but it's now much more than that
there are over 150 flourishing subprojects associated with the Apache master brand: Apache Hadoop, Apache Hive etc
is it time to free all the butterflies? ;-) 06:15
the suggestion is is that Perl becomes a master brand - similar to the Apache foundation
I think we need three distinctive runtime brand names 06:16
"perl" the runtime for legacy Perl 5 code 06:17
and two new runtime names - one for Perl 5 and one for Perl 6
for the next releases of these respective languages
each subcommunity could shortlist some names and submit them to Larry - to choose a name for each 06:19
both languages are free to flourish - with their own distinctive brands and sub-cultures 06:20
Zoffix bulat_: $*ERR is just a variable (a dynamic one). It simply contains an IO::Handle object that's open to STDERR. `$*ERR.say: 1, 2, 3;` (by default) is like `my $handle = open :w, IO::Special.new: '<STDERR>'; $handle.say: 1, 2, 3`; Although in most circumstances you can omit parentheses on subroutine calls (and can write `say 1, 2, 3`) they're required on method calls, so `$*ERR.say 1, 2, 3` was erroring out
for you because you missed the parentheses. And as others mentioned, you can use `:` instead of parentheses for trailing method calls and just write `$*ERR.say: 1, 2, 3`. (or just `note 1, 2, 3`, since it uses $*ERR under the hood)
06:21 darutoko joined
nige while at the same time being part of the Perl family and its culture and history 06:21
Zoffix nige: wasn't that already tried in Perl 5 land in the form of "Pumpkin Perl" and wasn't wildly successful?
nige HTH - just wanted to share this idea
I remember hearing about that - I wonder it might help if both languages rebrand at the same time 06:22
- I'm really just referring to the runtime name at this point 06:23
Zoffix nige: in the perfect world, perhaps. The Perl 6 rebranding discussion is very difficult. I imagine it'd be much much larger for Perl 5. And to do both at the same time seems to be the sum of both 06:24
nige I think "Perl" as a brand has outgrown it's association with a particular runtime - like Apache outgrew it's initial project 06:25
Zoffix nige: that kinda complicates the process, because in Perl 6 land the runtime already has a different brand: "Rakudo". It's the language name that's the issue
nige Yes - I hear your Zoffix - I know this is a BIG ask
when I run perl6 at the command line I use "perl6" not rakudo at the moment 06:26
it's what people type in for the runtime - that I think need separate branding 06:27
Zoffix I type "6" at the command line when I use Rakudo.
nige would be difficult to trade mark that ;-) 06:28
666 is already taken for one
like all good brands - honesty is important 06:30
"Perl" is more than perl and it's more than perl6
Zoffix And what is it? 06:31
nige it should become the master brand - because honestly that's what it is IMHO
similar to Apache - it's the master brand for lots of subprojects - it's a natural progression of the brand 06:32
TMTOTMTI
There's More Than One Trademark To It ;-)
Zoffix cues in relevant remarks from TimToady and goes back to hacking on guts: www.youtube.com/watch?time_continu...5t8qaAGw9w 06:33
06:33 Zoffix left, domidumont joined 06:38 domidumont left
nige thanks for the link - it sounds like TimToady++ is open to marketing "Perl" under different names - or have I misunderstood? 06:39
06:39 domidumont joined 06:47 parv joined
AlexDaniel nige: that's my understanding also 06:50
nige TimToady++ also made the point there - that deep down he sees it as all Perl - which I think points to "Perl" as the master brand 06:54
06:55 lowbro joined, lowbro left, lowbro joined
nige now we *just* need some distinctive marketing names (aka brand names) for the respective runtimes commands - so there is no collision at the command-line 06:56
IMHO - three runtime names are required: "perl" for legacy Perl 5 and a completely new name for the perl5 runtime command going into the future and a new name for perl6 07:01
07:02 hoelzro left
nige I can almost hear a collective groan ^ at the suggestion above - it doesn't have to be painful - and could reinvigorate each subculture/project 07:05
El_Che morgen 07:07
did we all have a good night rest? 07:08
(or afternoon nap :) )
Zoffix: Perl 6 could take the high road and rename it self without expecting Perl 5 to rename it self 07:16
as I see it, Perl 6 can pull off being a "new" language, historical Perl can. Also Perl 5 can benefit from the positive Perl association (backwards compatible, installed everywhere, ducktape), while Perl 6 does not (and is damaged by the undeserved bad rep that some people have of perl 5) 07:20
if the 2 lang narrative does not hold, I don't see the point of 2 langs with the name Perl in it 07:21
(and now I'll stop annoying people) 07:24
07:30 melezhik left
nige El_Che I think it's fair to say that the status quo is annoying people - we may need to invest some positive energy in getting to a good place 07:30
El_Che nige: the fragile status quo seem to be gone yesterday 07:31
(not that the status quo was a desirable state) 07:34
the new status quo (nothing changed with the langs) now with people with pitchforks it's maybe an opportunity for both langauges 07:35
I hope it could be, at least
07:36 rindolf joined 07:38 parv left, parv joined
TEttinger perl, pearl, purl 07:39
moritz PRL!
TEttinger hehe PRRRL
07:40 mspo left
TEttinger praline is a nice thing to name after 07:40
I can't remember exactly what it is but I like it in ice cream (is it candied pecans?)
was perl initially an acronym? 07:41
PHP is that weird recursive acronym now
Parlance might be interesting, a term for a sorta personal section of a language 07:42
El_Che TEttinger: Praline is a small stuffed chocolate 07:43
TEttinger also, lance, everybody likes medieval weapons
07:43 zakharyas joined
El_Che TEttinger: I am pretty sure moritz would agree with "Lenz" 07:44
TEttinger Internet Experler
Internet Experler 5.5
"worst of all worlds"
El_Che Machine Learning Big Data Microservice Container-based runtime
TEttinger ExaNanoNeuralScale 07:45
why Lenz? 07:46
something butterfly themed... Monarch, Luna, Clearwing 07:48
El_Che It's a famous Perl 6 book author
moritz nah, don't go naming stuff after me before I'm dead 07:49
TEttinger Luna is too close to Lua
El_Che moritz: if it would solve our situation it may be considered...
07:49 zakharyas left
moritz after I'm dead, I no longer care, and you can do as you want :) 07:49
TEttinger Nectar (butterflies love it) 07:50
07:50 zakharyas joined
El_Che moritz: the solution may be hastened :) 07:50
TEttinger you don't have to be dead to not care
you could get hooked on drugs stronger than perl 6 07:51
07:53 wamba left
moritz somehow I don't approve of the direction your thoughts are going :-) 07:55
TEttinger you could binge-watch every show on netflix 07:56
after 10 seasons of various "Real Housewives of $SOMEWHERE_RICH", you probably wouldn't care about anything 07:58
moritz doesn't sound appealing to somebody who writes programming books in his free time :-) 07:59
08:00 domidumont left 08:03 aindilis left, aindilis joined
player How to change working dir? 08:07
moritz player: typically you run indir "some directory", { code that runs in that directory }
docs.perl6.org/routine/indir 08:09
player It says NOTE: that this routine does NOT alter the process's current directory (see &*chdir) 08:10
It seems that indir doesn't change working dire, only variable.
08:11 parv left
player Only code in closure sees change PWD 08:15
08:15 parv joined
player after indir finishes, working dir is the same as it was before calling indir 08:15
08:16 hobbs joined, hobbs left, hobbs joined
moritz player: but if you use code inside of indir that uses the working directory (like accessing files, launching external processes etc.) they use the changed directory 08:20
player: so the question is really: why do you want to change directory? 08:21
08:25 troys_ is now known as troys 08:26 Actualeyes joined, stux|RC-- left 08:27 troys left, stux|RC joined
moritz the problem with chdir is that it's process-global, and not limited to the current thread 08:31
and so you should try to avoid it
player @moritz, I want to change to other directory, and execute git commands there. 08:36
parv bagh! 08:37
geekosaur as moritz said, it handles that
there's also things like the JVM backend becoming confused if you change directory to where it can't find its jars etc., so chdir is emulated 08:38
player Yes, but after indir, the program is still in old working directory.
And not the new one.
geekosaur the *interpreter core* is
your code will behave as if it is tin the new directory
what are you doing that would not accept this? trying to do syscalls manually? (there's a proper way to do that as well) 08:39
...in fact there shouldn't be a way to escape the abstraction, aside from badly written NativeCall. your program should not see a difference 08:40
if you absolutely believe you must micromanage everything, never write in anything but assembly language
player I want to be in cloned git repository after given section. 08:41
hastebin.com/unilepoqog.pl
geekosaur ok, abstraction goes over your head. 08:42
08:42 scimon joined
scimon Morning all. I hope everyone is feeling happy. 08:43
08:44 scimon left
player @scimon, hi. 08:45
@geekosaur, so what should I do?
geekosaur aapparently, stick to languages tht don;t do abstraction 08:46
player Returning back to my old working directory restricts my program seriously.
Oh
geekosaur why is there a difference between "the runtime makes sure things happen in the correct directory" and "the program always forcibly changes directory even if that breaks the jvm or another thread"
how do you ever program in other languages that happily do the latter? 08:47
08:47 abraxxa joined
geekosaur (yes, this *is* a common source of errors. people don;t get the difference between thread and process attributes.) 08:47
(...at some point I need to write up the haskell version of that faq, in fact. I end up fielding it often enough on irc...) 08:49
08:49 scimon joined
scimon And then I got a kernel patch. 08:49
player You mean every thread has its own working dir?
geekosaur and, sorry, I am more than a bit grumpy. kidmney stone and the drugs aren't helping in *any* sense
08:49 wamba joined
scimon So.... I had a thought last night while I looked at the chat logs after going to bed. 08:49
geekosaur player, perl 6 acts like it; that's why it's not *actually* changing dir unless it must. no other language does 08:50
scimon geekosaur: sorry to here that :(
geekosaur point big that you can act like every thread has its own directory and it will Just Work
*being
scimon Anyhoo. I thought of a name, because we love the "Why not change the name" just call Perl 6 (drumroll) V6
player Thanks. 08:51
scimon V stands for V (definitely not Version) Just V.
parv scimon, "V" as in "V for V just V", oui? 08:52
scimon Perl can have a version 6 major release if they like. V6 versions use base 36 and the first stable one was 'c' (so after v6.z we go to v6.10) 08:53
Oui :)
Just a random thought. I'm still trying to work out why there's so many Perl devs who are so angry about it. I mean I was working with 5.6 in my first job (so I didn't have to go through the 4 -> 5 release) but I've been working with perl for years now. 08:55
(Maybe that's it though. I have generally been working with it not being too involved with the community. For all it's talk it has not always been easy to get into.) 08:56
stmuk I think the perl 5 devs are blaming perl 6 for the decrease in popularity of perl 5. Even although Perl 6 exists exactly as a solution to this problem as predicated in 2000
El_Che scimon: if the name it's chaged, keeping the version is a terrible idea (for perl6 and perl5) 08:57
moritz player: but the indir *around* the code calling git
stmuk I don't think changing the names of anything will make any difference whatever
scimon But the 6 (in v6) isn't the version. It's just part of the name. Like the ++ in C++
El_Che scimon: if you're explaining the name because it's ambigous, you're doing it wrong :)
stmuk: I honestly think it will 08:58
stmuk the Naming Discussion is a complete waste of time
scimon I mostly thought of it so all the scripts thyat say use v6; at the top don't need cahnged ;)
El_Che stmuk: I can't count the numbers of people I have met that won't try Perl 6 because it's Perl
stmuk: and the number of people that think Perl5 is dead because it's stuck to a major version 08:59
stmuk: experiences are subjective, some people have different experiences, but this is what I've met so far
scimon (As I say, I've been coding in Perl for 15+ years, and I've been having the arguments with management for much of the later time. It's not Perl6 that have hit Perl5's popularity it's node and Java nad python (not in that order)
Problem is majot version changes are hard. Heck python 2 -> 3 is still causing issues. 09:00
El_Che scimon: people here know that :)
scimon Anyway. I don't really care about the name. *Because* I don't care I'd be fine with changing it. 09:02
I'd rather work on selling the language on it's strengths and making useful libraries for people to use. 09:03
El_Che scimon: the name is hurting Perl 5 more than Perl 6. I really see the argument there that Perl 5 people make
scimon: it hurts Perl 6 too (a lot), but that's Perl 6's decision
stmuk it seems impossible to have any serious discussion about the future of perl 5 and 6 without the same tired arguments coming up time and time again 09:04
El_Che stmuk: why do you think it's that?
stmuk: because nothing is ever solved
stmuk: the 2 langs narrative offered a truce for some time. 09:05
stmuk: it had the advantage of setting the naming discussion on the side for some time 09:06
stmuk there is much bitterness in some parts of the community (look at p5p)
I'm sticking to "sister"
El_Che stmuk: bitterness, true
but that's often the other side of the coin of love
stmuk the idea of bring back "v5" is sadly lost in the noise 09:07
El_Che stmuk: it was a fragile compromise at best...
stmuk I guess its the "official line" still being on the perl6.org website 09:08
although people can express their own opinions
El_Che it is, as long as we don't sound as comical ali :)
DrForr Larry has to allow a new name; until then the discussion is somewhat moot, but needs to go on so that the important player (note the lack of plural there) knows it's a proble. 09:11
09:11 lookatme left
El_Che Larry can change his mind 09:11
stmuk maybe a slang of perl 1 "use v1" would be a starting point called "munie" or something :)
El_Che thinking that the discussion/frustration will go away until then is quite optimistic
the sister narrative was very fragile, but the best compromise so far. If it's not supported anymore, a new solution will be needed. 09:12
Civil war while sharing the same name is Monty Python humour 09:13
09:20 cdg joined 09:22 domidumont joined 09:23 dakkar joined
kraih change the name to Python 6 :p 09:23
09:24 cdg_ joined, wamba left
scimon It's a thought. 09:24
;) 09:25
El_Che kraih: you troll :)
09:25 cdg left
TEttinger there's a grain of truth there 09:25
El_Che kraih: you could port Mojo to Perl 6 so people could say "Perl5 lost its mojo"
scimon I guess because I've basically spent a lot of time not involved in this, I don't have the scars from the earleir attempts (I tried to get involved a few times but could never get my head around how to run it) I am still optimistic. 09:26
I feel that comment may not go down weel.
*well
TEttinger if perl6 is really so different from perl5, why not name it something completely without bias, like COBOL_6 09:27
El_Che which one? yours? my joke?
TEttinger python's a start, we can go worse
huf call it north perl 09:28
scimon Well kraih seemed a bit negative last night.
TEttinger NBRE, Nothin' But Regular Expressions
scimon I feel he was burnt in the past. Which sucks.
TEttinger Sigilicious
09:28 cdg_ left 09:29 Aceeri_ joined
TEttinger sig.ilicio.us 09:29
scimon Like I say, I'm an optimist. This is the first time in years I've really felt like I've found a language that works with my brain (which possibly says something about my brain) I've always liked Perl5 but I've possibly spent too much time dealing with scary scary scary codebases.
09:32 wamba joined
scimon Has to be said I'm mostly focussed on the "Making a useful set of libraries for Perl6" part of the equation. 09:32
And if some of them look a lot like Perl5 libs making it easier for people to start using them. Great. 09:33
09:42 pmurias joined
pmurias El_Che: re sister languages narrative, there is nothing for Perl 6 to gain in opposing Perl 5, and currently things like Python are more of a Perl 5 "replacement" than Perl 6 09:46
09:47 Kaiepi left 09:48 Kaiepi joined
lizmat afk for a few hours& 09:49
09:52 bulat_ left, player left
kraih the sentiment i hear most often in the Perl 5 community today is hope for Perl 6 to just fade into obscurity 09:53
that's a change from yesterday's "just rename Perl 6 and let both languages prosper"
jast hey, renaming it to Python was my idea :/ 09:54
09:56 kraih left 09:57 ponbiki joined 09:59 pmurias left 10:14 markong joined 10:22 wamba left
DrForr The forerunner was '6lang' last time I bothered to check; just use that and roll with it. 10:24
El_Che not a fan of it, but better than what we have now 10:25
10:26 TEttinger left 10:28 pmurias joined
moritz 6lang can't convince me 10:29
stmuk I thought liz's post was about "v5" and not renaming! :) 10:30
El_Che stmuk: wasn't the working title "let's open this can of worms"? :)
stmuk apparently! 10:31
wasn't the "v5" support removed from rakudo anyway? 10:32
10:36 wamba joined
stmuk 10:36
rt.perl.org/Public/Bug/Display.html?id=130834
DrForr Well, we have 'use v6;' so the version syntax is still available. I think that got set aside in order to get something out the proverbial door. 10:37
stmuk that rt was probably an error in hindsight
El_Che hindsight? sir, do you have some time to talk about handsight? :P 10:38
by now, we need doctor who
DrForr I'm sure she'd set this straight. 10:39
stmuk by pointlessly running around and spouting a lot of dull techno babble? 10:40
El_Che stmuk: I haven't watched the episodes with the new dr yet, but it sounds you're not a big fan
DrForr I don't think they've been released yet. I saw the regeneration teaser last week or so, that's been it so far. 10:41
stmuk I grew up with the 70s ones and quite liked some of the shows from a decade or so ago but the recent ones have been as dull as ditchwater
El_Che there is a christmas special waiting on my disk
stmuk: they should rename the show 10:42
El_Che : mic drop
:)
jast doctor what
El_Che doctor .WHAT
stmuk Nurse Who :P 10:44
geekosaur hasn't been so fond of the reborn series either. then again, it's also not the movie.
which was more Doctor WAT 10:45
stmuk I like Blink and Talons of Weng Chiang (although I was 11) 10:46
El_Che I have no history with the old series, pretty fond of the reboot (as it's the only one I really know, except an old episode left and right) 10:48
10:48 wamba left 10:49 wamba joined 10:57 domidumont left 11:05 domidumont joined, parv left 11:07 aborazmeh joined, aborazmeh left, aborazmeh joined 11:10 cdg joined 11:14 cdg left
scimon Old enough to remember watching the old shows (Tom Baker onwards) as a kid. Really wish they could have givne Peter Capaldi some better scripts. I am hopefully for the new series. 11:23
(And when is the Doctor going to learn to regenerate outside? Sheesh) 11:24
El_Che didn't he run out of regeneration tokens?
geekosaur wasn't 10 given a new set? 11:30
El_Che maybe it's not really a bearer token
and he can just renew it
scimon So yeah, he ran out of regens. Then River song gave him all of hers... or the time lords gave him some at Trensalor... or something. Then he became Peter "I'm a bit moody and dark" Capaldi.... 11:35
(I like stories sorry)
El_Che scimon: good thing your getting marries. No way you're picking someone up with that story :) 11:36
geekosaur never got to see those eps
wondering if he's more like 2, or 6
scimon She thinks I'm cute and smart. I try really hard to not disabuse her of these crazy notions.
El_Che food 11:37
scimon: haha. congratz btw
scimon Cheers.
11:39 stmuk_ joined 11:41 stmuk left
DrForr scimon: Wasn't Baker's outside at the Jodrell Bank telescope? 11:41
scimon Yup, he fell of it.
*off 11:42
11:52 aborazmeh left 12:08 wamba left
geekosaur there's also the Troughton edge case 12:10
(outside the TARDIS at least)
12:12 pmurias left, pmurias joined 12:14 zakharyas left 12:18 Aceeri_ left
Altreus is that the one in the film 12:26
that we cunningly ignore unless required
not the new film, the old one
DrForr I thought that was Colin Baker -> Paul McGann? Though I'll admit to be somewhat out of the loop. 12:27
Altreus Anyway I thought the TARDIS was instrumental in the regeneration nonsense 12:28
and that's why he always regenerates inside it
I say he
they :P
12:28 vichib left
scimon Dalek Invasion Earth? Peter Cushing wasn't it? 12:29
Paul McGann was after the little chap... not Tom Baker...
No no. I think nowadays he regens inside so the huge light show can explode the tardis so they have an excuse to redecorate. 12:30
12:31 stmuk joined, jeromelanteri left
Altreus Well you do have to keep your obsolete disguise fashionable 12:31
geekosaur Altreus, when the Time Lords forced him to regenerate and trapped him on Earth (2ns -> 3rd Doctor) 12:33
although I am not recalling that sequence and never did manage to see the actual episode so it might have happened in the TARDIS
12:33 stmuk_ left
DrForr There are plenty of regeneration supercuts, and that one'll be fairly early on :) 12:37
scimon :) Righto. 20 mins before my internal perl6 coding dojo. 12:41
DrForr Must hear how you arranged this at some time... 12:42
12:43 eliasr joined 12:46 pilne left, Zoffix joined
Ulti just read some of the back log... anyone who's livlihood is so fragile that Perl 5 not being selected by managers utterly affects their ability to live, need to *strongly* reconsider how they are thinking about their career as a programmer regardless of the status of Perl 12:48
I think this is really the major issue there are "Perl 5 guys" left in Perl 5 in a condensed state and its a very strange state of a community
I love Perl 6 but this year I've been doing 100% Python for work, the two years before that it was Perl 5 with a bit of JS 12:49
at no point have I been unemployed waiting for Perl 6 to be something Im interested in or want to program in 12:50
DrForr Most of my runs of unemployment have been waiting for companies to finish the (rather protracted, in my case) hiring process. 12:53
stmuk I haven't seen any arguments in the backlog or on reddit which were new or different to the ones I've been hearing for years
Ulti the period of time that P5 had a super strong hold on large scale commerical products was almost twenty years past and it was a short period of time.... its not like Perl ever had Java or C level usage and commercial mindset
stmuk it seems any mention of perl 6 to the wider community just triggers "renamers"
Ulti if you want software jobs for the rest of your life pick up a book on Java or C++ 12:54
Perl is a tool to use for the rest of your life that makes life immensely enjoyable and easy
stmuk: yeah and the name I think probably did matter, if you changed it now it would make a lot of people happy... and then deeply unhappy 12:55
Zoffix El_Che: the "high road"? As far as I'm concerned we're still on schedule to create a marketing alias for the name with 6.d release, as was always planned. People who yell the loudest about naming discussions being pointless and renames unachievable expect everything to change overnight, as soon as Proposal Name #13232 is announced by someone.
Ulti which is probably why it should happen, so that the community can move past it 12:56
El_Che Zoffix: glad to hear 12:57
Zoffix: however, an alias does not help Perl 5 12:59
stmuk liz's post wasn't about renaming I'm not sure why we allowed the agenda to be changed 13:00
El_Che Zoffix: so it's not a solution for all the Perl community (which we assume it exists)
stmuk: because the issue wasn't resolved and a sneeze would have teared it open again
stmuk a vocal minority of perl 5 people hate perl 6 and you won't be able to make them happy 13:01
El_Che true 13:02
but it's not them I care about
13:02 wander joined
El_Che the malaise can not be reduce to that group 13:02
stmuk you can call perl 6 Flibbertigibbet and people will still be angry 13:03
El_Che Perl 5 and 6 *do* have a problem. Liz proposed her solution.
jast what exactly is the goal with renaming, anyway? I'm still not really sure
Zoffix El_Che: it does help. It's an opportunity for the renamers camp to prove the current name is as detrimental and the new name is as beneficial as they claim. If the hypothesis is true, the alias will naturally become the One True Name for the language through its mere use.
El_Che Not everyone agrees with the solution, but most acknowledge there is a problem
Zoffix: I see the plan. But it's still very Perl 6 centric: "we liberate the name only when whe are sure we benefit, otherwise screw you" could be an interpretation 13:05
lizmat FWIW, I haven't heard anything new in the comments so far
DrForr It'll be interesting to see what the undercurrents are at FOSDEM. 13:06
lizmat only reaction on p5p by Zefram: www.nntp.perl.org/group/perl.perl5...48848.html
El_Che lizmat: because it's the same discussion that has never gone away
lizmat: bad timing for FOSDEM, though
Ulti P5 and 6 having a unified runtime with full deep interop would be kind of amazing, especially if XS to some extent could be auto shimmed to NativeCall style stuff
stmuk El_Che: it's nothing new and will probably be forgotten by FOSDEM 13:07
Zoffix El_Che: I only care about Perl 6 as far as renaming goes. It's purely a way to market Perl 6 more efficiently rather than some concession to the Perl 5 camp.
lizmat Ulti: I'm not saying that that will happen, I'm saying it *may* happen
stmuk virtualsue will have Camelia soap bars at FOSDEM :)
lizmat Ulti: but we need the CPAN Butterfly Plan first
El_Che Zoffix: I disagree there. I am closer to lizmat's POV on that part: we're still a Perl community (or I want it to be like that)
pmurias lizmat: why is the CPAN Butterfly Plan a requirement? 13:08
El_Che Zoffix: therefor I want solutions than favour all the community
lizmat because it would need to be done for a Butterfly Perl 5 Project
Ulti lizmat: yeah it could the existing v5 is a good blob of evidence its possible
lizmat even if you *would* create a Perl 5 runtime as a slang, you would still need oodles of modules that live in Perl 5 core and on CPAN
El_Che Zoffix: would you be happy with Perl 5 jumping to 7 and make perl 6 looks like yesterdays news? Win-win for Perl 5. 13:09
Zoffix: that's not the road we want to walk
DrForr Triple-life modules now, 5 core, 5 CPAN and 6 CPAN :)
jast do it like the browsers, bump version to 40 or something
stmuk maybe try an easier language as a slang first perl 1 or something like was tried with Punie and parrot 13:10
13:10 noganex left
DrForr Drop the 5 and go the Java route :) 13:10
jast Perl EE? :)
lizmat stmuk: I see a different approach between use v5 and BP5P:
El_Che Perl 6 could be the new JavaScript
lizmat v5 tried to create a full fascimile of Perl 5 from the start 13:11
with the entire Perl 5 test-suite
Zoffix El_Che: I don't believe the existence of Perl 6 is Perl 5's biggest problem. There's a reason Perl 6 was started in the first place. They're now trying to implement more and more of Perl 6's features (signatures, smartmatch, `when` semantics) but without causing incompatibility breakage that Perl 6 accepted as inevitable and their users are freaking out (
www.nntp.perl.org/group/perl.perl5...48369.html ). Perl 6 is just their boogeyman IMO
pmurias lizmat: some of the XS-using internals exposing modules would need to be adapted in some way for Butterfly Perl 5, but things like DBI wouldn't
13:11 noganex joined
lizmat I see BP5P as more of a "make baby Perl 5 work and take it from there" approch 13:11
jast is there any initial work in that direction yet? 13:12
lizmat pmurias: why wouldn't DBU need to be adapted / migrated ?
stmuk perl 6 is certainly the perl 5 scapegoat
El_Che Zoffix: of course you're right. But it's *not* a purely --or even-- technical discussion
stmuk I thought the original "v5" slang approach a good one
perl6advent.wordpress.com/2013/12/...16-slangs/ 13:13
Zoffix El_Che: I might be happy if they rename to Perl 7, yeah. It'd give the Perl 6's language alias more desirability. But it'd be totally stupid for Perl 5 to be renamed to Perl 7 **right now** as they don't have anything nice to offer to justify a major release.
jast as someone who has used perl 5 a lot and who isn't heavily invested in either language, I see that p5 has kept many of its quirks (syntactical and otherwise) and languages in the same genre have left theirs behind (and many new languages have appeared)... and if anything has caused a decline in p5's popularity, it's probably that
pmurias lizmat: I forgot the DBI drivers have XS parts, but stuff like DBIx::Class could be just loaded by Butterfly Perl 5
stmuk froggs had over 4000 perl 5 tests passing 13:14
lizmat pmurias: indeed
Zoffix stmuk: yeah, but which Perl 5 does "use v5" enable? 13:15
lizmat stmuk: but use v5 was about also supporting all of the "difficult" perl 5 code, the code that typically lives inside modules
CPAN modules
pmurias it's interesting to see that TruffleRuby on the JVM just run the Ruby XS equivalent on the JVM using their LLVM-on-JVM implementation
stmuk Zoffix: not sure what you mean 13:16
El_Che Zoffix: agreed
lizmat afk for 30 mins or so&
Zoffix stmuk: there are differences in Perl 5 versions. If I write `use v5;` what version gets loaded? 13:18
DrForr 'use v5 :ver<22>' :) 13:20
pmurias Zoffix: emulating the behavior of a latest Perl 5 without a version declaration seems best
stmuk Zoffix: an arbitary recent one with attempts to catch up every few releases 13:21
its a moving target (although slow moving target) 13:22
Zoffix Doesn't that defeat the purpose of porting [old] modules to Perl 6?
13:22 zakharyas joined
Zoffix i.e. if I use a module that hasn't been updated for latest smartmatch changes, `use v5` will crap out on it and I have no options left. With Inline::Perl5, I can just lock my application to a specific version of Perl 5 13:23
(and yes, I know those changes were reverted now, but the point stands)
pmurias Zoffix: Perl 5 is assumed to be back compatible
El_Che pmurias: you missed the smartmatch discussion :) 13:24
Zoffix pmurias: But that's not reality.
And another point to it: we currently do have a portion of what you propose implemented: :P5 regex modifier. It's poorly tested, buggy, and last I saw seemed to miss some features going back as far as 5.10. Considering this small part of the language doesn't have enough maintainers, is it reasonable a slang for the entire language has more viability? 13:25
Zoffix has to go do @things
\o
13:25 Zoffix left
stmuk running perl 5 under perl 6 is clearly desirable (and has always been a stated aim) but I doubt its currently practical (at least until speed issues and much else is fixed) 13:31
13:33 wamba joined
AlexDaniel I see everyone has a lot of energy today. You guys should check these out: github.com/rakudo/rakudo/issues?q=...2%9A%A0%22 13:36
AlexDaniel is struggling with cold
DrForr has been struggling with backache :/ 13:37
El_Che has been struggling with this Perl 5 & 6 thing 13:38
AlexDaniel not saying that 5/6 thing is not important, just saying that some help with the blockers will… help :) 13:40
13:55 cdg joined 13:56 wamba left 14:13 bisectable6 left 14:14 bisectable6 joined 14:17 mspo joined 14:19 natrys joined, nige left
scimon DrForr: It's mostly on my lunch break :) 50% increase in attendee's this week (up to 3!) Took our new dev through some of the ways Perl5 and 6 differ as he'd never done any :) 14:19
b2gills [backlogging] As for re-branding, either both projects should re-brand at the exact same time or they should keep the identities they have and push forward together. I'm tired of seeing the bickering happen between my two favorite languages. 14:20
scimon AlexDaniel: I don't think my Perl6 fu is strong enough yet. And my nqp fu is worse. 14:21
AlexDaniel strong enough for what?
scimon Well looking at the open issues. 14:22
14:29 efg left
lizmat commute to Niederrhein.PM meeting& 14:30
14:30 lizmat left 14:31 lizmat joined, lizmat left 14:39 releasable6 left, greppable6 left, releasable6 joined, greppable6 joined, ChanServ sets mode: +v greppable6 14:43 pmurias left 14:49 pmurias joined 14:52 zakharyas left 14:53 zakharyas joined 14:57 zakharyas left, zakharyas joined 14:59 nige joined 15:23 bart__ joined
scimon (Coding dojo included a small detour into how to playing Magic the Gathering and why I wrote a perl6 script to handle card interactions) 15:28
15:31 lowbro left 15:38 vike1 joined, vike left 15:39 pmurias left 15:41 bart__ is now known as brrt 15:51 ilbot3 left 15:52 ilbot3 joined, ChanServ sets mode: +v ilbot3 16:01 wamba joined 16:02 cdg left, cdg joined 16:03 notbenh left 16:05 zakharyas left, zakharyas1 joined 16:12 mempko left, brrt left
scimon I've got it! 玑 6 Or TLFKAP6 16:17
If it worked for Prince.... 16:18
16:19 MilkmanDan joined
timotimo prinsix 16:29
scimon Actually the more I think about it 玑 (or Ji) is a nice name. 16:34
Ok jī to be correct.
jast a.k.a. ungoogleable 16:35
scimon Because A) it's unicode! Yay! B) no mention of 6 so everyone who is complaining we stole the 6 can be quiet and C) it means irregularly shaped pearl which is fun. 16:36
And.... it's ungoogleable .
Sheesh everyone's got some kind of issue. ;)
jast I was thinking... if Perl is taken, how about Dimond ;)
scimon You mean 🔹 16:43
:D
On vague serious having a unicode variant of the name would be kinda awesome..... I think I need more sleep. 16:44
jast no, that's a diamond
there is... 🦋
none of these render in my IRC client, but I think that's right 16:45
scimon Di🔹mond
So yours didn't for me but the small blue diamond did.
jast spoiler: it's a butterfly
scimon Oooooo
jast not *the* butterfly, but close enough
(vaguely unrelated side note, my problem with Camelia is that I always think Kademlia) 16:46
scimon Di⨝ ond
ilmari ruby has already taken 💎 16:50
scimon See!!! 16:52
jast no, ruby has taken something which apparently doesn't have a unicode code point yet :) 16:56
16:56 cdg_ joined 17:00 cdg left 17:06 troys joined, cdg_ left
El_Che yeah, that's what you want, picking a new referring to project seens as stalled from the outside (perl5 and ruby) 17:07
17:09 cpage joined 17:11 cpage_ left
El_Che rakudo is a great name, worry about non-existing implimentations later. It's pretty clear there is only one working implementation and that no new ones are coming until it picks up 17:11
(it about picking a crappy name (6lang, psix, etc) when you have a perfect one already that's recognized) 17:12
17:14 hoelzro joined
El_Che I would so buy a rakudo tshirt, hopefully with a dragon on it :) 17:16
17:21 domidumont left 17:23 benchable6 left, coverable6 joined, benchable6 joined 17:24 khw joined 17:27 scimon left, abraxxa left
tbrowder off topic: which parent has a child’s wedding coming up? 17:32
tony-o both parents, most likely 17:34
tbrowder er, only one is here i think...anyhow, i created some things for my daughter’s wedding in raw postscript such as the wedding service program and a souvenir postcard. i will be happy to share the PS files which can easily be modified to suit the upcoming event. 17:40
email me at [email@hidden.address] if interested. 17:41
17:42 dakkar left
perigrin tony-o: I would suspect *all four* parents 17:42
17:42 sena_kun joined
sena_kun m: role A[::Ret] { method a(--> Ret) { Ret.new } }; class B{}; class C does A[B] {}; say C.a; 17:47
camelia Died with X::TypeCheck::Return
in method a at <tmp> line 1
in block <unit> at <tmp> line 1
sena_kun it totally looks like a bug to me.
where should I file the issue(if it's a bug after all), on github or RT?
tony-o perigrin: if it was children's, maybe
jnthn Looks to me also. Rakudo bug tracker
(the GitHub one is probably easiest) 17:48
sena_kun ok, github one then~
tony-o but a child is only half of the wedding, and has only 2 (at least biologically) parents
[Coke] stares at tony-o & perigrin.
perigrin right but he didn't specify which child ... and for each set of soon-to-be-inlaws it's only one *child* from their perspective
tony-o if that's the case they still wouldn't consider the parents of the other child 17:49
perigrin so grep { $_->child->wedding_date <= soon() } @parents ... would return 4 instances ... no?
17:49 lizmat joined
tony-o $_->child->parents would be only 2 .. 17:50
perigrin right
tony-o tbrowder: we're at an impasse
it's 2|4
perigrin well ... modern marriage customs 17:51
tony-o that's a good point
perigrin it's >0 with some kind of a normal distribution
mst grep { any($_->children)->wedding_date <= soon() } @parents ?
17:51 natrys left
tbrowder ok, cut to the chase: the offer stands regardless of slang errors! 17:51
mst though you'd presumably want to reduce the @parents to @families fi- no that won't work if there's been a divorce and remarriage 17:52
has to be @relationships I guess or something
tony-o do step-parents stick around for the children after divorce? 17:53
perigrin well I was thinking @parents was just grep { $_->has_children } @people ... and then we'd sort it from there 17:54
17:54 khw left
mst in less dysfunctional families one would expect both parents to attend the kid's wedding even if they're in other relationships by that point 17:54
17:55 setty1 joined, domidumont joined
tony-o there's a lot to sort out 17:56
perigrin @people.grep({ .has_children }).grep({ any(.children).wedding_date <= soon() }) ... with apologize because I don't *actually* know Perl6
apologies
mst: trog had all three of his parents there
geekosaur considers trying that around CMU. or MIT. 17:57
perigrin as did I actually at mine
geekosaur doesn't think he wants to work out the sufficiently general form >.>
tony-o perigrin++
17:57 troys is now known as troys_
gfldex `any(.children».wedding-date)` looks better to me 17:58
perigrin tbrowder: suffice to say for some kind of probably gausian distribution around 2-4 ... the offer stands.
18:01 zakharyas1 left 18:04 zakharyas joined 18:05 khw joined
[Coke] doc site question: we have some stuff we manage with 'make' and some stuff we manage inside of the giant htmlify; What's the current thought on going all-make vs. all p6? (I tend to prefer Make because it's a common, shared, horrible thing which I already understand, but I get that this is a great project for us to eat our own dogfood. 18:06
(even if we have to figure out what that looks like first)
gfldex [Coke]: make is tricky on Windows (I do like make too tho) 18:08
ilmari strawberry perl ships make
gfldex if we acutally want to move away from all external deps (what is very well possible) then there is good reason to start early 18:09
[Coke] we can make something that works with nmake or gmake (or other make)
gfldex i didn't write there is no make on Windows. It's just full of traps.
[Coke] yup.
gfldex I'm pretty sure we could turn the Makefile we have into a .p6 without much work 18:10
[Coke] having said it out loud, I think p6 is probably a better way to go. I'll start poking around in a branch. 18:11
gfldex [Coke]++ 18:12
mspo mv Makefile Makefile.p6
what could go wrong?
gfldex .oO( In Perl 6 we make our own luck! ) 18:13
[Coke] sees he had a "makefile" branch from ages past that can probably die. 18:20
18:21 imcsk8 joined
Geth perl6-lwp-simple: wbiker++ created pull request #24:
HTML header names with mixed case fix
18:21
18:21 wander left 18:23 cdg joined 18:27 cdg left, cdg joined 18:30 sjoshi joined 18:32 troys_ is now known as troys, noganex left 18:33 noganex joined 18:34 darutoko left
Geth perl6-lwp-simple: eb43079ba4 | Wolf++ (committed by Zoffix Znet) | 2 files
HTML header names with mixed case fix (#24)

  * Transform HTML header names to lowercase before looking for location or charset
18:37
18:40 reportable6 left 18:41 nativecallable6 left, reportable6 joined, nativecallable6 joined, Scimon joined 18:43 cdg left 18:44 cdg joined 18:50 ryn1x joined 18:51 sjoshi left 18:53 zakharyas left, Kaiepi left, _Kaiepi joined
ryn1x Is there a built in way to read from a Buf while advancing a pointer through it? Eg. I want to read 20 bytes from a Buf, and then the next time I want to read 4 bytes starting from where I left off at byte 20 and so on. The closest I can find is .subbuf and then keeping track of $from manually. 18:56
yoleaux 31 Dec 2017 12:47 MST <TimToady> ryn1x: we don't do Lisp-style cons lists by default because the head/tail model is bad for parallel processing of lists, so we keep them at a higher abstraction level; however, you can make tail-sharing lists if you like by using => for cons (it's even right associative for that reason)
18:58 _Kaiepi left 19:00 Kaiepi joined 19:04 Zoffix joined
Zoffix ryn1x: you could make use of lower-level Iterator interface 19:04
m: my $b := Buf.new(1..100).iterator; say $b.pull-one xx 10; say $b.pull-one xx 20;
camelia (1 2 3 4 5 6 7 8 9 10)
(11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30)
TimToady bets subbuf is faster as the batches get larger though 19:07
might even be faster for batchsize of 1...
Zoffix m: my $b := Buf.new(1..100); say $b.splice: 0, 10; say $b.splice: 0, 20; 19:09
camelia Buf:0x<01 02 03 04 05 06 07 08 09 0a>
Buf:0x<0b 0c 0d 0e 0f 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e>
Zoffix ryn1x: brainfart... .splice is better than what I originally suggested
rindolf hi all 19:10
sup?
TimToady Zoffix: assuming you don't mind destroying the buffer, of course 19:11
Zoffix nods 19:12
TimToady hmm, curious
m: my $b := Buf.new(1..100); say subbuf-rw($b,0,10) = Buf.new() for 1..2
camelia Buf:0x<01 02 03 04 05 06 07 08 09 0a>
Buf:0x<0b 0c 0d 0e 0f 10 11 12 13 14>
TimToady that works too, I guess, but splice is more better anyway
19:15 noganex left 19:16 noganex joined 19:17 Ven`` joined
ryn1x Thanks Zoffix and TimToady ! Going to try those out now. 19:19
TimToady m: my $b := Buf.new(1..100); say $b.rotor(10).perl 19:20
camelia ((1, 2, 3, 4, 5, 6, 7, 8, 9, 10), (11, 12, 13, 14, 15, 16, 17, 18, 19, 20), (21, 22, 23, 24, 25, 26, 27, 28, 29, 30), (31, 32, 33, 34, 35, 36, 37, 38, 39, 40), (41, 42, 43, 44, 45, 46, 47, 48, 49, 50), (51, 52, 53, 54, 55, 56, 57, 58, 59, 60), (61, 62…
TimToady or that
m: my $b := Buf.new(1..100); say $b.rotor(1,2,4...*).perl 19:21
camelia ((1,), (2, 3), (4, 5, 6, 7), (8, 9, 10, 11, 12, 13, 14, 15), (16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31), (32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62,…
TimToady m: my $b := Buf.new(1..20); say $b.rotor(1,2,4...*,:partial).perl # oopsie 19:24
camelia ((1,), (2, 3), (4, 5, 6, 7), (8, 9, 10, 11, 12, 13, 14, 15)).Seq
TimToady m: my $b := Buf.new(1..20); say $b.rotor((1,2,4...*),:partial).perl
camelia ((1,), (2, 3), (4, 5, 6, 7), (8, 9, 10, 11, 12, 13, 14, 15), (16, 17, 18, 19, 20)).Seq
TimToady m: my $b := Buf.new(1..20); say $b.rotor(1,2,4...*):partial.perl
camelia ((1,), (2, 3), (4, 5, 6, 7), (8, 9, 10, 11, 12, 13, 14, 15), (16, 17, 18, 19, 20)).Seq
TimToady
.oO(TYL...)
19:25
m: my $b := Buf.new(1..20); say $b.rotor(1,2,4...*) :partial.perl 19:26
camelia ((1,), (2, 3), (4, 5, 6, 7), (8, 9, 10, 11, 12, 13, 14, 15), (16, 17, 18, 19, 20)).Seq
Zoffix :o
TimToady m: my $b := Buf.new(1..20); say $b.rotor(1,2,4...*) :partial .perl 19:27
camelia ((1,), (2, 3), (4, 5, 6, 7), (8, 9, 10, 11, 12, 13, 14, 15), (16, 17, 18, 19, 20)).Seq
TimToady probably parsing that way
TimToady has wondered whether for 6.d we should special-case the adverbs placed directly after a postfix/postcircumfix (without intervening space) to always apply to that operator regardless of precedence 19:29
since that's usually what people intend 19:30
and I doubt there's much code out there that relies on the adverb leaping a postfix to find a prefix or infix yet 19:31
we could probably detect any such now and warn "this will require whitespace before the adverb in the future" 19:33
Zoffix m: my %h = <a b>; say !%h<a>:exists
camelia Unexpected named argument 'exists' passed
in block <unit> at <tmp> line 1
Zoffix So that this would DWIM?
TimToady yes 19:34
19:34 someuser left
TimToady and .rotor():partial would also continue to work since method calls are officiallly postfixes 19:34
19:36 vichib joined
TimToady if that looks weird, consider how STD would allow you to switch from parentheses to colon-form in midstream: 19:36
m: $*OUT.say(1,2,3): 4,5,6
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused
at <tmp>:1
------> 3$*OUT.say(1,2,3):7⏏5 4,5,6
expecting any of:
colon pair
Zoffix Opened this as R#1418 19:39
synopsebot R#1418 [open]: github.com/rakudo/rakudo/issues/1418 [RFC] Special-case the adverbs placed directly after a postfix/postcircumfix
19:41 lichtkind joined
lichtkind does anyone know the nickname of pierre vigier? thanks 19:44
19:46 someuser joined
ryn1x is there a naming convention for subroutines that mutate their arguments? 19:47
mst prefix the name with 'footgun-'
ryn1x ha 19:49
TimToady just pick a suitable destructive verb: clobberate($x) 19:50
lizmat ryn1x: many use the "set_" prefix 19:51
this meme is actually used in some of Perl 6 bootstrap internals
TimToady with methods, you can use .=foo to indicate $x = $x.foo 19:52
19:52 Ed__ joined
TimToady m: my @a = <a b c>; @a.=rotate; say @a 19:53
camelia [b c a]
ryn1x great suggestions. thanks.
lizmat m: my $b := { ... }; $b.^set_name("Foo"); dd $b.^name
camelia "Foo"
19:53 khw left 19:54 cdg left
lizmat m: my $b := { ... }; $b.set_name("Foo"); dd $b.name # less damaging :-) 19:54
camelia "Foo"
19:56 zakharyas joined, Ven` joined
lizmat decommute& 19:57
19:57 lizmat left
ryn1x lizmat: I've been meaning to ask you if you ever recieved the thank you post card I sent? I very thankful for the stuffed Camelia, books, stickers, and pins you sent me a while back! I am now using Perl6 everyday at work (at a US National Laboratory) and as a part time CS student. 19:58
Zoffix \o/ 19:59
19:59 domidumont left
masak ryn1x: awesome! 20:01
and lizmat++
Ed__ Hi, just installed Perl6 2017.06 on Raspberry Pi, so far working. Unfortunately perl6-debug-m test.pl6 leads to: “Could not find Debugger::UI::CommandLine at line 1 in” followed by several directories...I’m grateful for anz hint... 20:04
20:05 lichtkind left
DrForr Ed__ - it's a separate module, you'll need to install it, ideally using 'zef install Debugger::UI::CommandLine' 20:05
20:06 khw joined
ryn1x .tell lizmat ^ 20:07
yoleaux ryn1x: I'll pass your message to lizmat.
Ed__ ...Thanks, I thought I did that already, but will check again!
DrForr If you think you've already installed it, try checking for CommandLine.pm on the paths it listed, maybe it failed to install. 20:08
Zoffix Ed__: the debugger is busted in 2017.06. It was fixed in 2017.07 20:09
The fix FWIW: github.com/rakudo/rakudo/commit/6d4691fb 20:10
20:12 Ed__ left, Ven` left, Ven` joined 20:16 Ed__ joined
Zoffix Ed__: don't know if you received last message before you disconnected, but debugger is broken on 2017.06. You need 2017.07 or later or to apply this patch manually: github.com/rakudo/rakudo/commit/6d4691fb 20:17
20:18 troys is now known as troys_
Ed__ !Thanks Zoffix 20:19
...will have a look at it! Thanks so much!! 20:20
20:24 Ed__ left 20:25 Zoffix left 20:30 TEttinger joined 20:39 pilne joined 20:48 dudebbn__ left 20:51 evalable6 joined, ChanServ sets mode: +v evalable6
mspo TimToady: clobbernate(x) sounds more fun 20:54
20:58 comborico1611 joined
comborico1611 Any Emacs users present? I installed company-web from melpa. I then enabled it, but I see no results. I'm not sure how to locate the package settings in Emacs. 21:02
21:14 Ven` left 21:15 Ven` joined 21:19 setty1 left
comborico1611 There must be some setting overriding the package . . . O Emacs . . . 21:26
21:29 dj_goku joined
sena_kun comborico1611, did you look at github.com/osv/company-web#setup? 21:29
of course, you need to `eval-buffer` your .emacs after setting appropriate configuration(or just restart emacs, but for what reason) 21:30
21:35 combo joined
comborico1611 sena_kun: I read it, but I dont' know how to access Emacs Config. 21:36
21:39 troys_ is now known as troys
sena_kun comborico1611, why not just `.emacs` file in your home directory? My question may be rude, but did you read up some tutorials if you're a beginner with emacs? E.g. www.masteringemacs.org/article/beg...e-to-emacs is a good start. 21:39
comborico1611 I'll try that. Thank you! I read the manual in Emacs itself. I then watched some videos. I also read Linux User's Guide, which covered Emacs, but I don't think Greenfield covered this. 21:41
I did browse that link you posted just yesterday.
Wondering if it was a good book. For some reason I kept thinking its author was the creator of Emacs.
sena_kun well, M-x tutorial is very good for basic hotkeys, but it doesn't cover a lot of things. The blog post which link I've posted here covers package management, basic configuration and explain things like buffers, modes, etc. 21:42
comborico1611 sena_kun: I tried C-f ~/.emacs Nothing came up but a new file (blank file)
sena_kun because you did not create your config(yet). :) 21:43
comborico1611 Okay. I'll go back to it with more intent.
Thank you for the help!
sena_kun it's raw, there are no pre-built configs for you or something like that.
np
s/explain/explains/ 21:44
21:45 Scimon left
comborico1611 Hmm? /explain? 21:51
21:51 zakharyas left, espadrine joined
mspo I'd like to use emacs but don't know if I could not-vi 21:52
sena_kun just a silly grammar mistake I do every time when I debug things at night and talk to people in English at the same time. :/ 21:53
El_Che mspo: does your family been known to have genetics (evolution or manufactured) that allows for 12 fingers?
comborico1611 Haha!
21:53 khw left
mspo El_Che: that explains my inability to emacs 21:54
sena_kun >inability to emacs
sounds like a serious illness.
"Oh God, call M-x doctor, quickly!"
gfldex we call it "differently gifted" 21:55
comborico1611 sena_kun: Do you have any recommendations for Emacs books / articles ? 21:58
timotimo FWIW you can make up for 2 missing extra fingers by having two foot switches
comborico1611 Haha. Like a piano! 21:59
timotimo that's probably also a good idea just in general
alternatively, 10 foot switches like an organ
well, organs probably have more than 10?
comborico1611 I have no idea. I do recall seeing some paddles down there.
sena_kun I bet it'd be much slower than this caps right under your pinky though. If you're not a black metal drummer or something. 22:00
mspo I'm mostly concerned about my wrists
but when I looked at vimscript I thought "maybe emacs isn't that bad"
timotimo haha
yeah, vimscript isn't a very pleasant language
sena_kun comborico1611, about books: I dunno, to be honest. It was not yesterday as I started with it, but from what I can recall: default emacs tutorial + some articles(I think the one I posted is good enough). Then goes some googling like "How to make your emacs shine". And a lot of time to discover "must have" plugins, installing it, etc. 22:02
timotimo sena_kun: any thoughts on spacemacs?
oh. my. god.
sena_kun Getting used to hotkeys is like 2 weeks iirc, but after that you literally cannot use arrows. I have emacs keys for navigation in firefox and other things, just because... Hmm.
timotimo weechat allows you to grab lines you inputted in any buffer back with ctrl + up arrow 22:03
this changes *everything*
sena_kun timotimo, I tried it like 3 or something years ago, but it's too big. I have my own set of features I need(some modes, some preferences) plus I don't want any "fancy" stuff they have.
comborico1611 sena_kun: Thanks! 22:04
timotimo right
comborico1611 I'm off to supper or a little more study. Bye, guys!
sena_kun my .emacs is like ~200loc and I feel this is more than enough. Though I don't use features like auto-complete, etc. Default $language-mode is ok. 22:05
22:05 comborico1611 left
sena_kun comborico1611, also see github.com/emacs-tw/awesome-emacs 22:05
22:05 combo left
sena_kun ugh. 22:05
.tell comborico1611 also see github.com/emacs-tw/awesome-emacs
yoleaux sena_kun: I'll pass your message to comborico1611.
22:05 khw joined 22:06 combo joined
combo sena_kun: Alright! Thanks. I hate reading from Github, though. Always seems confusing. I'm more of a book guy than a README guy. Hence my problem learning this stuff. 22:07
(this is combo1611, by the way.) Bye, guys!
22:07 combo left
sena_kun well, I tried to read Stallman's book, but... I saw things I didn't want to know about, like registers and some features nobody remembers about for last 10 years or something. :) 22:08
22:11 troys is now known as troys_, mempko joined 22:18 wamba left 22:25 pmurias joined 22:38 ryn1x left 22:39 rindolf left
a3f How does an array of Ints type look like in a sub parameter list? I want to type my MAIN sub. 22:39
moritz sub f(Int @a) { ... } 22:41
but, that won't work for MAIN 22:42
because that signatures says the caller needs to pass in an Array[Int]
jnthn @a where .all.isa(Int) may well do it
moritz not just an array where all elements happen to be Ints
a3f Aha. So is there an alternative syntax? jnthn Doesn't work for me. I get the usage message 22:45
or just ask for Str and split it up myself?
timotimo it doesn't work because you get IntStr objects 22:46
they don't .isa(Int)
jnthn Really? 22:47
m: say IntStr.isa(Int)
camelia True
timotimo m: say <123>.isa(Int)
camelia True
timotimo oh, huh?
jnthn IntStr multiply inherits from Int and Str iirc
timotimo m: [IntStr.new(1, "1"), IntStr.new(2, "2"), IntStr.new(3, "3")].all.isa(Int).say
camelia False
timotimo m: IntStr.new(3, "3").all.isa(Int).say
camelia False
gfldex m: sub s(@a where .all ~~ Int){ say 'here' }; s([1,2,3])
camelia here
timotimo m: IntStr.new(3, "3").isa(Int).say
camelia True
timotimo another problem is that MAIN will only let you use positional arrays like that if you make them slurpy 22:48
jnthn Oh...of course
a3f I am fine by slurping all elements into the single array. I just want to make sure all arguments are Ints
timotimo oh, jnthn, it'll .isa on the Junction, rather than distributing
gfldex m: sub s(*@a where .all ~~ Int){ say 'here' }; s(1,2,3)
camelia here
jnthn Oh, .isa doesn't distribute over the junction 22:49
haha
m: sub s(*@a where .all ~~ Int){ say 'here' }; s(<1>,<2>,<3>)
camelia here
timotimo that's why that works, then
jnthn Yeah, what gfldex++ suggested is the way to go
a3f Ye, that works. Thanks all. How's the * in front of the @a called? 22:50
timotimo i just say "slurpy parameter" 22:51
jnthn Slurpy. Because it slurps up all available arguments
timotimo there's multiple kinds of slurpy for positionals, too
gfldex there are quite a few nice examples for MAIN signatures in this book: github.com/kyclark/metagenomics-bo...perl6/main 22:52
a3f gfldex Thanks. looks helpful. 22:53
El_Che MAIN is pretty awsome
thx to lizmat we also have this beauty: my $*MAIN-ALLOW-NAMED-ANYWHERE = True; 22:55
This will allow to pass named parameters anywhere instead on the non-unixy way of only at the beginning
so you can do stuff like ./prog action --param1=foo --param2=bar 22:56
22:56 sena_kun left 22:57 lizmat joined 22:58 traxex_ joined
tadzik aww yiss 22:59
23:00 s0me0n3-unkn0wn left 23:01 traxex left 23:05 Zoffix joined, cdg joined
Zoffix jnthn: what's the criteria for choosing which ops to mark in this commit? R#1413 is fixed by marking `return_o` with :useshll... Wonder which other ones might need it github.com/MoarVM/MoarVM/commit/db...bf3b4afb59 23:06
synopsebot R#1413 [open]: github.com/rakudo/rakudo/issues/1413 [regression][⚠ blocker ⚠] Weird issue with `zef install`
El_Che lizmat: 118 comments on reddit. Pretty impressive 23:07
lizmat yeah :-) 23:08
yoleaux 20:07Z <ryn1x> lizmat: ^
jnthn Zoffix: Ah, darn 23:10
If we mark that op :usehll the entire reason for introducing it becomes pointless
Zoffix :(
23:10 preaction joined
jnthn Could you note your findings on the ticket? 23:11
Zoffix Sure
I got "golf" too (it's a giant enum)
jnthn I think maybe we can do an analysis for return_o and the invokee in inlince.c
gah, inline.c
Like, invoke_o + return_o = ok to inline even if cross-HLL, but otherwise don't do it 23:12
23:12 s0me0n3-unkn0wn joined, sunnavy joined
jnthn In the long run we can go one better than that 23:12
If you fancy trying that, feel free, otherwise I should have some time tomorrow to have a look 23:13
Zoffix I'll leave it to you. I don't know this area of the codebase and just bruteforced the location of the bug :)
jnthn That's still very helpful :)
The opt in question made every closure clone around 15% cheaper, so I'm quite keen to fix it up properly :) 23:14
Zoffix :)
jnthn They're still a bit pricey, alas 23:15
Zoffix Commented 23:17
Zoffix &
23:17 Zoffix left 23:26 khisanth_ left 23:34 troys_ is now known as troys
Xliff 118 comments on reddit? Where? 23:40
El_Che Xliff: www.reddit.com/r/perl/comments/7r1..._community 23:41
23:41 cdg left
El_Che 124 by now, it seems lizmat is still awake :) 23:41
lizmat yeah :-)
Xliff Thanks, El_Che 23:42
23:43 Ven` left
El_Che is it me or is reddit awful to track new comments? I set it on "new" and search for " minute" 23:43
23:46 khisanth_ joined
Xliff "Please stop hurting Perl 5. Your project will never be widely adopted and keeping the same name is hurting real workhorse." 23:48
/o\
lizmat yeah, stuff like that 23:49
23:54 khw left
Xliff Can you run a compiled Perl6 package without perl6? 23:54
Ala: Running the moarvm code directly from moarvm.
timotimo regular perl6 modules won't do that on their own i don't think 23:55
lizmat Xliff: you still need the Perl 6 boostrap 23:56
*bootstrap
Xliff OK. I am trying to think of a way users can bypass stage parse. Consider a frozen codebase that doesn't need to be parsed, and can just run the precompiled code. 23:57