»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.
SmokeMachine m: say % = $*REPO.repo-chain.flatmap(*.loaded>>.Str).grep(*.strats-with: "MyPrefix::").map({::($_)}).map: {.get-my-key => $_} # I'm doing something like this! If you want to use a plugin, load it! 00:01
camelia {}
SmokeMachine star: 00:03
star: say % = $*REPO.repo-chain.flatmap(*.loaded>>.Str).grep(*.strats-with: "Test::").map({::($_)}).map: {.^name => $_} # I'm doing something like this! If you want to use a plugin, load it!
camelia {}
IOninja You have a typo in starts with 00:04
00:05 naxieAlDle joined
SmokeMachine And I didn't load anything... 00:06
00:11 telex left 00:12 telex joined 00:17 risou is now known as risou_awy
SmokeMachine star: use Pod::To::HTML; say % = $*REPO.repo-chain.flatmap(*.loaded>>.Str).grep(*.starts-with: "Pod").map({::($_)}).map: {.^name => $_} 00:17
camelia {Pod::To::HTML => (HTML)}
SmokeMachine star: use Pod::To::HTML; use Pod::To::Text; say % = $*REPO.repo-chain.flatmap(*.loaded>>.Str).grep(*.starts-with: "Pod").map({::($_)}).map: {.^name => $_} 00:27
camelia {Pod::To::HTML => (HTML), Pod::To::Text => (Text)}
naxieAlDle BenGoldberg: lost my internet connection for a while, so couldn't answer
BenGoldberg: for some reason it stringifies every element
m: constant h = %(<a 1 b 2>); enum E (h); say E.enums 00:28
camelia Map.new(("a\t1\nb\t2" => 0))
naxieAlDle m: constant h = %(<a 1 b 2>); enum E (|h); say E.enums 00:29
camelia Map.new(("a\t1" => 0,"b\t2" => 1))
00:29 llfourn left
naxieAlDle BenGoldberg: maybe it should work, I don't know 00:31
BenGoldberg Is that a rakudobug?
naxieAlDle possibly 00:32
BenGoldberg: so yes, submit a ticket and we'll see what happens then :) 00:34
00:34 llfourn joined 00:36 espadrine left 00:38 bjz left 00:50 obfusk left, obfusk joined 00:51 laouji joined 00:52 Sound joined 00:55 laouji left 00:56 irobotnot left 00:59 khw joined
BenGoldberg m: enum Nums (qw,zero one two,); sub foo(-->Int) { two }; say foo; 01:03
camelia two
BenGoldberg m: enum Nums (qw,zero one two,); sub foo(-->int) { two }; say foo;
camelia 2
BenGoldberg m: enum Nums (qw,zero one two,); sub foo(-->Str) { two }; say foo;
camelia Type check failed for return value; expected Str but got Nums (Nums::two)
in sub foo at <tmp> line 1
in block <unit> at <tmp> line 1
BenGoldberg m: enum Nums (qw,zero one two,); sub foo(-->Str()) { two }; say foo;
camelia two
01:03 madochan joined, aborazmeh joined, aborazmeh left, aborazmeh joined, cdg left 01:07 cdg joined 01:12 cdg left 01:14 aborazmeh left 01:18 aborazmeh joined, aborazmeh left, aborazmeh joined 01:21 telex left 01:23 lukaramu_ left 01:39 Ven joined 01:52 laouji joined 01:57 laouji left 01:58 risou_awy is now known as risou 02:02 aborazmeh left 02:08 cdg joined 02:13 cdg left, user__ joined 02:14 Ven left
labster m: say True ⁇ "foo" ‼️ "bar"; 02:14
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused: Found ⁇ but no ‼
at <tmp>:1
------> 3say True ⁇ "foo"7⏏5 ‼️ "bar";
expecting any of:
infix
infix stopper
labster no emoji operators? c'mon guys :P 02:15
02:16 Ven joined
SmokeMachine m: say True ⁇ "foo" ‼️ "bar"; 02:16
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused: Found ⁇ but no ‼
at <tmp>:1
------> 3say True ⁇ "foo"7⏏5 ‼️ "bar";
expecting any of:
infix
infix stopper
02:16 risou is now known as risou_awy 02:17 cibs left, kalkin- left
SmokeMachine m: say True ⁇ "OK" ‼ "NOK" 02:18
camelia OK
02:19 cibs joined
SmokeMachine labster: wrong ‼... 02:19
labster Hm, that's not working on the rakudo I just built. Do I need to use a version or something?
02:19 madochan left
SmokeMachine labster: that was done this week... 02:20
labster Oh duh, I forgot to pull before rebuilding rakudo! 02:21
02:22 Sound left
SmokeMachine :P 02:23
u: ‼️ 02:27
unicodable6 SmokeMachine, U+203C DOUBLE EXCLAMATION MARK [Po] (‼)
SmokeMachine, U+FE0F VARIATION SELECTOR-16 [Mn] (◌️)
02:27 Ven left
SmokeMachine u: double interrogation mark 02:30
unicodable6 SmokeMachine, Found nothing!
02:31 bjz joined
SmokeMachine u: double interrogation 02:31
unicodable6 SmokeMachine, Found nothing!
SmokeMachine u: double
unicodable6 SmokeMachine, U+00AB LEFT-POINTING DOUBLE ANGLE QUOTATION MARK [Pi] («)
SmokeMachine, U+00BB RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK [Pf] (»)
02:31 Ven joined
unicodable6 SmokeMachine, 438 characters in total: gist.github.com/5b903ccc52c5f3dcbf...c1579c0434 02:31
SmokeMachine u: double question mark 02:32
unicodable6 SmokeMachine, U+2047 DOUBLE QUESTION MARK [Po] (⁇)
SmokeMachine m: 'm: say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"'.EVAL 02:33
camelia 5===SORRY!5=== Error while compiling /home/camelia/EVAL_0
Confused
at /home/camelia/EVAL_0:1
------> 3m: say True7⏏5 \c[double question mark] "OK" \c[double
expecting any of:
infix
infix stopper
postfix…
BenGoldberg m: 'm: say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"'.say
camelia m: say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"
SmokeMachine m: 'say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"'.EVAL
camelia 5===SORRY!5=== Error while compiling /home/camelia/EVAL_0
Confused
at /home/camelia/EVAL_0:1
------> 3say True7⏏5 \c[double question mark] "OK" \c[double
expecting any of:
infix
infix stopper
postfix
SmokeMachine m: qq'say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"'.EVAL 02:34
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused
at <tmp>:1
------> 3qq'say True7⏏5 \c[double question mark] "OK" \c[double
expecting any of:
infix
infix stopper
postfix
statement end
SmokeMachine m: qq'say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"'.say
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused
at <tmp>:1
------> 3qq'say True7⏏5 \c[double question mark] "OK" \c[double
expecting any of:
infix
infix stopper
postfix
statement end
BenGoldberg m: "say True \c[double question mark] 'OK' \c[double exclamation mark] 'NOK'".say
camelia say True ⁇ 'OK' ‼ 'NOK'
BenGoldberg m: "say True \c[double question mark] 'OK' \c[double exclamation mark] 'NOK'".EVAL.say
camelia OK
True
SmokeMachine BenGoldberg: thanks! 02:35
BenGoldberg Welcome
02:35 dct left
SmokeMachine I didn't get why the qq version didn't work... 02:36
BenGoldberg m: qq''.say;
camelia
BenGoldberg m: qq'\n'.say;
camelia
BenGoldberg m: qq'say True'.say;
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3qq'say True7⏏5'.say;
expecting any of:
infix
infix stopper
postfix
statement end
statement modifier…
SmokeMachine m: qq|say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"|.say 02:37
camelia say True ⁇ "OK" ‼ "NOK"
BenGoldberg m: qq!say True!.say;
camelia say True
SmokeMachine Is that a bug? Isn't qq working well with '?
m: qq'True' 02:38
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3qq'True7⏏5'
expecting any of:
infix
infix stopper
statement end
statement modifier
statement modifie…
SmokeMachine m: qq/True/
camelia WARNINGS for <tmp>:
Useless use of constant string "True" in sink context (line 1)
SmokeMachine m: say qq"True" 02:39
camelia True
SmokeMachine 23:37 <SmokeMachine> m: qq|say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"|.EVAL 02:40
m: qq|say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"|.EVAL
camelia OK
02:40 bjz left
SmokeMachine m: say qq'' 02:41
camelia
SmokeMachine m: say qq'a'
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3say qq'a7⏏5'
expecting any of:
infix
infix stopper
postfix
statement end
statement modifier
SmokeMachine rakudobug? 02:42
ilbelkyr m: sub qq'x () { 42 }; say qq'x
camelia 42
SmokeMachine Hum! 02:43
Ok...
m: say qq ' bla ' 02:44
camelia bla
SmokeMachine m: say qq 'bla'
camelia bla
02:45 bjz joined
naxieAlDle SmokeMachine: by the way, why ⁇‼ in particular? 02:45
like, why not something else?
SmokeMachine m: qq 'say True \c[double question mark] "OK" \c[double exclamation mark] "NOK"'.say 02:46
camelia say True ⁇ "OK" ‼ "NOK"
02:46 ilbot3 left
SmokeMachine naxieAlDle: what do you mean? 02:46
Those nicks are getting complex... 02:47
02:47 imcsk8 left
naxieAlDle SmokeMachine: wasn't it you who implemented ⁇‼ ? 02:47
02:47 ilbot3 joined, ChanServ sets mode: +v ilbot3
SmokeMachine naxieAlDle: yes 02:48
naxieAlDle SmokeMachine: out of all unicode things you can add, why did you decide to work on ⁇ ‼ ? :)
02:48 ZzZombo joined, ZzZombo left 02:49 ZzZombo joined
SmokeMachine I thought that was too much coincidence both parts of the operator exist as Unicode char to do not do that! 02:50
naxieAlDle: ^^
naxieAlDle: and that will help on golf... 02:51
02:53 user__ left
naxieAlDle SmokeMachine: how so? 02:54
most golf tasks only care about bytes
?? is two bytes
u: ‘⁇’.encode.elems.say
unicodable6 naxieAlDle, U+2018 LEFT SINGLE QUOTATION MARK [Pi] (‘)
naxieAlDle, U+2047 DOUBLE QUESTION MARK [Po] (⁇)
naxieAlDle, 20 characters in total: gist.github.com/7fe9640cd507887483...b5504563ed
naxieAlDle m: ‘⁇’.encode.elems.say
camelia 3
naxieAlDle ⁇ is 3
m: ‘⁇‼’.encode.elems.say
camelia 6
naxieAlDle m: ‘??!!’.encode.elems.say
camelia 4
naxieAlDle 🤷 02:55
SmokeMachine m: ‘⁇‼’.elems.say
camelia 1
SmokeMachine m: ‘⁇‼’.chars.say 02:56
camelia 2
02:56 Ven left
SmokeMachine Unicode language means Unicode golf... 02:56
And that was the first thing that I tried the Unicode version and didn't work... 02:58
I thought that, for completeness, would make sense to do that... 02:59
naxieAlDle ≤ ≥ ≠ :P
SmokeMachine I didn't try... :P
naxieAlDle: why are you changing the anagram you are using as nick? 03:01
03:01 ZzZombo left
naxieAlDle SmokeMachine: I wanted some change 03:02
and given that Zoffix does it every freaking month, I guessed it won't hurt anything :)
03:04 ZzZombo joined, ZzZombo is now known as Guest93070, Guest93070 left, Guest93070 joined, Guest93070 is now known as ZzZombo
SmokeMachine m: "SmokeMachine".comb.pick(*).join.say #should I? :P 03:05
camelia chMkmSaeione
03:16 noganex joined 03:19 noganex_ left 03:20 imcsk8 joined 03:25 ufobat_ joined, ZzZombo_ joined, ZzZombo_ left, ZzZombo_ joined, ZzZombo is now known as Guest54591 03:26 Guest54591 left 03:28 wamba joined 03:29 ufobat left
BenGoldberg m: .say for lines().map: /(.*?) /; 03:49
camelia No such method '!cursor_start' for invocant of type 'Str'
in block <unit> at <tmp> line 1
naxieAlDle m: lines().map: /./ 03:57
camelia No such method '!cursor_start' for invocant of type 'Str'
in block <unit> at <tmp> line 1
naxieAlDle m: lines.map: /./
camelia No such method '!cursor_start' for invocant of type 'Str'
in block <unit> at <tmp> line 1
naxieAlDle m: (‘hello’,).Seq.map: /./
camelia No such method '!cursor_start' for invocant of type 'Str'
in block <unit> at <tmp> line 1
naxieAlDle BenGoldberg: well, that's a bug of some sort 03:58
even if that is not supposed to work, it is definitely LTA
BenGoldberg Do you want to rakudobug it?
naxieAlDle BenGoldberg: well, would be nice if you do it :) 03:59
03:59 risou_awy is now known as risou
naxieAlDle m: say (‘hello’,).Seq.map: * ~~ /./ 04:00
camelia (「h」)
naxieAlDle this is what should happen, right?
04:10 cdg joined 04:14 cdg left 04:16 risou is now known as risou_awy 04:50 mawkish__ joined 04:51 khw left
b2gills m: say [~] %*ENV<ME>.comb.pick(*) 04:53
camelia isbll2g
04:55 Cabanoss- joined
SmokeMachine m: say (‘hello’,).Seq.map: /./ 04:57
camelia No such method '!cursor_start' for invocant of type 'Str'
in block <unit> at <tmp> line 1
SmokeMachine m: say (‘hello’,).Seq.map: {/./} 04:58
camelia (/./)
SmokeMachine m: say (‘hello’,).Seq.map: {/./;}
camelia (/./)
SmokeMachine m: say (‘hello’,).Seq.map: -> $_ {/./}
camelia (/./)
04:59 araraloren joined, Cabanossi left, Cabanoss- is now known as Cabanossi 05:05 bjz left 05:07 bjz joined
BenGoldberg m: module Foo { }; use Foo; 05:25
camelia ===SORRY!===
Could not find Foo at line 1 in:
/home/camelia/.perl6
/home/camelia/rakudo-m-inst-2/share/perl6/site
/home/camelia/rakudo-m-inst-2/share/perl6/vendor
/home/camelia/rakudo-m-inst-2/share/perl6
CompUnit::Reposit
05:52 snowblack joined 05:59 adu joined
snowblack question, $ perl6 gives me prompt. How do I load a file with definitions I have writen in a file? 05:59
05:59 risou_awy is now known as risou
araraloren What prompt ? 06:02
geekosaur I think they want to know how to load a file from the repl 06:04
...which should be require, I think?
masak morning, #perl6 06:05
06:06 perlawhirl joined
araraloren o/ 06:07
06:07 llfourn left 06:08 BenGoldberg left
snowblack how to load a file into the repl, yes. filename is k1, require "./k1" 06:09
results in "Could not find ./k1 in: [a list of directories]; how to make it look in current dir? 06:10
geekosaur mm, was afraid of that :/ 06:11
araraloren Add module search path ?
geekosaur use lib '.'
06:11 cdg joined
araraloren use lib may be work 06:11
geekosaur I am not sure there's a better way at present
snowblack use lib "." require "./k1" does execute k1, but doesn't seem to have remembered the definitions in the file 06:15
06:16 cdg left 06:17 risou is now known as risou_awy 06:18 geekosaur left
araraloren use lib "path"; use "filename without extension"; 06:19
Will be work
06:19 geekosaur joined 06:20 geekosaur left 06:21 mawkish__ left 06:22 mawkish__ joined, geekosaur joined
araraloren snowblack, eg. ls output: module.pm6 In ERPL: use lib "."; ↓ use module; ↓ Then you can use class or variable definitions in module.pm6 06:23
06:30 adu left 06:33 llfourn joined 06:43 ufobat_ left 06:57 ZzZombo joined, ZzZombo left, ZzZombo joined 06:58 ZzZombo left 06:59 ZzZombo_ left 07:00 CIAvash joined 07:02 zakharyas joined 07:07 darutoko joined
snowblack still not working 07:07
cat k2.pm6 07:08
sub sample($x) {$x * $x}
use lib "."; 07:09
require "k2";
say sample(45345);
require "k2"; 07:10
Could not find k2 in:
/black1/home_alt/carl/btr/carlthuban/math/projecteuler.net/060
etc
Is there not a simple example somewhere? 07:11
07:18 dalek left 07:28 ZzZombo joined, ZzZombo left, ZzZombo joined 07:47 zakharyas left 07:52 cibs left 07:54 cibs joined 07:55 laouji joined
timotimo IOninja: if i remember correctly, "closing" was a sort of crutch that was needed to make sure things get called in the right order for cleanup. i might be100% wrong here, but i think it was kicked when supplies got a major internal overhaul at some point 07:57
bisectable6: my $s = Supplier.new; my $tap = $s.Supply.tap(-> $v { say "the value is $v" }, done => { say "Supply is done" }, closing => { say "Tap closed" }, quit => -> $ex { say "Supply finished with error $ex" }, ); $tap.close
bisectable6 timotimo, On both starting points (old=2015.12 new=e114d52) the exit code is 0 and the output is identical as well
timotimo, Output on both points:
timotimo bisectable6: my $s = Supplier.new; my $tap = $s.Supply.tap(-> $v { say "the value is $v" }, done => { say "Supply is done" }, done => { say "Tap done" }, quit => -> $ex { say "Supply finished with error $ex" }, ); $tap.close 07:58
bisectable6 timotimo, On both starting points (old=2015.12 new=e114d52) the exit code is 0 and the output is identical as well
timotimo, Output on both points:
timotimo well, done only runs when the supplier closes it, not the tap
so that's probably fine 07:59
naxieAlDle all: my $s = Supplier.new; my $tap = $s.Supply.tap(-> $v { say "the value is $v" }, done => { say "Supply is done" }, done => { say "Tap done" }, quit => -> $ex { say "Supply finished with error $ex" }, ); $tap.close
timotimo IOninja: note however that an on-demand supply has a closing argument and attribute. you don't pass it via tap, though
07:59 laouji left
committable6 naxieAlDle, gist.github.com/156e4cc6b1547bf653...e2e3b55900 07:59
naxieAlDle bisect: old=2015.07 my $s = Supplier.new; my $tap = $s.Supply.tap(-> $v { say "the value is $v" }, done => { say "Supply is done" }, done => { say "Tap done" }, quit => -> $ex { say "Supply finished with error $ex" }, ); $tap.close 08:00
bisectable6 naxieAlDle, Bisecting by exit code (old=2015.07 new=e114d52). Old exit code: 1
naxieAlDle, bisect log: gist.github.com/fcc9da46c9bc422e66...29e6ea04b9
naxieAlDle, (2015-11-26) github.com/rakudo/rakudo/commit/6b...30409bb74f
naxieAlDle bisect: old=6bd80ec9cc my $s = Supplier.new; my $tap = $s.Supply.tap(-> $v { say "the value is $v" }, done => { say "Supply is done" }, done => { say "Tap done" }, quit => -> $ex { say "Supply finished with error $ex" }, ); $tap.close
bisectable6 naxieAlDle, On both starting points (old=6bd80ec new=e114d52) the exit code is 0 and the output is identical as well
naxieAlDle, Output on both points:
08:00 risou_awy is now known as risou
naxieAlDle ah right 08:01
anyway, I have no idea what you were trying to find 08:02
08:04 mawkish__ left, mawkish__ joined
naxieAlDle timotimo, IOninja: is it this? github.com/rakudo/rakudo/commit/a8...00318142c5 08:07
had to do this to find it: bisect: old=6bd80ec9~50 new=6bd80ec9^ my $s = Publisher.new; #`(rest is same) 08:08
08:08 mawkish__ left, mawkish__ joined
timotimo looks right on the money! 08:09
when someone fixes up that piece of documentation, it'd be nice to point people towards supply blocks 08:14
supply/react blocks
naxieAlDle m: say ‘hello’ 08:17
camelia hello
08:17 risou is now known as risou_awy
timotimo oh no 08:17
is hack hanging again?
ugh, it absolutely is 08:18
08:19 Util_ left, PerlJam left, Undercover left, sergot left, SourceBaby left, synopsebot6 left 08:20 dalek joined, ChanServ sets mode: +v dalek, synopsebot6 joined, ChanServ sets mode: +v synopsebot6
timotimo it'd be great if we could get util, perljam, and sergot off of hack.p6c.org and onto irc.p6c.org 08:20
cool, that's a downtime of just a little bit over an hour 08:21
i just randomly wanted to sign into hack and there it was
i just be magic
naxieAlDle well, occasional reboot is good!
timotimo it's not a good kind of reboot, though 08:22
it's the kind where you just flip the power switch off
because it can't properly reboot ... the hard drives have just full-on disappeared in those cases
naxieAlDle I've only rebooted my server once in two years, not a good way to do it also :P 08:23
timotimo ah 08:24
yeah, maybe :)
unless of course you've subscribed to ksplice or what it's called
08:28 synopsebot6 left, synopsebot6 joined, ChanServ sets mode: +v synopsebot6
naxieAlDle not only that, but I'm also not sure if everything that is supposed to come up on boot actually does so 08:28
08:29 diakopter left
timotimo i changed synopsebot a tiny bit so that it hopefully understands Geth doc links now 08:29
someone wanna change some docs to try it? ;)
08:32 xinming_ left 08:33 xinming joined 08:34 domidumont joined, CIAvash left 08:36 gk_1wm_su joined 08:40 domidumont left, gk_1wm_su left, parv joined, domidumont joined 08:59 mawkish__ left 09:00 mawkish__ joined 09:04 xinming left 09:05 rindolf joined, xinming joined 09:06 vike left 09:08 mawkish__ left, mawkish__ joined, eroux joined
Geth doc/≤≥≠: 150de2b6a0 | (Aleks-Daniel Jakimenko-Aleksejev)++ | 2 files
Docs for ≤ ≥ ≠ unicode ops
09:10
timotimo it won't trigger for branches because those don't get built on the docs site
Geth doc: AlexDaniel++ created pull request #1234:
Docs for ≤ ≥ ≠ unicode ops
09:11
09:11 mawkish__ left 09:12 mawkish__ joined 09:15 Actualeyes joined, mawkish__ left 09:16 mawkish__ joined, sena_kun joined 09:20 mawkish__ left, mawkish__ joined 09:21 naxieAlD` joined 09:22 naxieAlD` is now known as naxieAlDle`
naxieAlDle` :| 09:22
09:22 RabidGravy joined 09:23 zakharyas joined, naxieAlDle left, LeCamarade left, naxieAlDle` is now known as naxieAlDle 09:24 Sound joined 09:26 cibs left
Geth whateverable: 13cbe9a899 | (Aleks-Daniel Jakimenko-Aleksejev)++ | 6 files
Sync OUTPUT with style in docs

Resolves issue #101, tests changed accordingly.
Committable will now show exit code in a funny way (with nested
  «»). This has to be changed in the future.
09:27
09:28 cibs joined, perlawhirl left 09:40 gk_1wm_su joined, gk_1wm_su left 09:41 bisectable6 left, committable6 left, benchable6 left, unicodable6 left, evalable6 left, statisfiable6 left 09:42 unicodable6 joined, ChanServ sets mode: +v unicodable6, bisectable6 joined, benchable6 joined, evalable6 joined, committable6 joined, ChanServ sets mode: +v bisectable6, ChanServ sets mode: +v benchable6, ChanServ sets mode: +v evalable6, ChanServ sets mode: +v committable6, statisfiable6 joined, ChanServ sets mode: +v statisfiable6, travis-ci joined
travis-ci Doc build passed. Aleks-Daniel Jakimenko-Aleksejev 'Docs for ≤ ≥ ≠ unicode ops' 09:42
travis-ci.org/perl6/doc/builds/207859179 github.com/perl6/doc/commit/150de2b6a01b
09:42 travis-ci left 09:44 FROGGS joined 09:48 dct joined 09:49 naxieAlDle left
moritz good morning 09:59
10:00 zakharyas left 10:01 risou_awy is now known as risou
lizmat moritz o/ 10:02
moritz lol I blogged: perlgeek.de/blog-en/perl-6/2017-01...icode.html
lizmat whee! :-) 10:04
10:06 ifim left 10:12 Ven joined 10:13 Sound left
nebuchadnezzar \o/ 10:15
I thought english use “” to quote…
RabidGravy boo! 10:17
10:18 risou is now known as risou_awy 10:29 Ven left 10:30 sammers joined, parv left 10:45 TEttinger left 10:49 cdg joined 10:52 ZzZombo_ joined, ZzZombo_ left, ZzZombo_ joined, ZzZombo is now known as Guest61025, ZzZombo_ is now known as ZzZombo, labster left 10:54 Guest61025 left 10:55 mawkish__ left 10:56 mawkish__ joined 11:11 lukaramu joined, vendethiel left 11:15 Ven joined
RabidGravy wonders idly if POSIX mqueues would play nicely 11:17
there's only one way to find out 11:27
11:27 LeCamarade joined
El_Che boom 11:28
11:34 pmurias left 11:35 Ven left
RabidGravy oh va_args :( 11:36
11:42 mawkish__ left
IOninja AlexDanial: s/Geth/GitHub/ reports the name inconsistently. Eventually Geth will use the rakudo/CREDITS file to fish out the names tho 11:49
Gee, we're now adding ≤ ≥ ≠ ops -_- 11:53
I wish people were as eager to fix the 1,500 bug tickets for existing features. 11:54
DrForr Just like typical jobs :/ No time to fix bugs, always time to add new features. 11:55
11:56 laouji joined 12:01 laouji left, bjz_ joined 12:02 bjz left 12:06 mawkish__ joined, llfourn left 12:09 llfourn joined 12:25 lep-delete left 12:27 lep-delete joined 12:36 Sound joined 12:37 bjz_ left 12:38 bjz joined
IOninja I notice we call the colon form on this page "Filetest operators": docs.perl6.org/type/IO$COLON$COLON..._operators 12:50
However, that syntax is much more general and is a property of Pair.ACCEPTS 12:51
m: say 31337 ~~ :is-prime
camelia True
IOninja It calls the method with the name of the .key, and I don't see that listed under Pair.ACCEPTS 12:52
12:52 mr-foobar joined
IOninja s: :foo, 'ACCEPTS', \(42) 12:52
MasterDuke huh, never knew it was a general syntax
12:53 SourceBaby joined, ChanServ sets mode: +v SourceBaby
IOninja s: :foo, 'ACCEPTS', \(42) 12:53
SourceBaby IOninja, Something's wrong: ␤ERR: Cannot resolve caller sourcery(Str, Capture, :foo); none of these signatures match:␤ ($thing, Str:D $method, Capture $c)␤ ($thing, Str:D $method)␤ (&code)␤ (&code, Capture $c)␤ in block <unit> at -e line 6␤␤
12:53 Undercover joined, ChanServ sets mode: +v Undercover
IOninja s: Pair.new("foo", True), 'ACCEPTS', \(42) 12:53
SourceBaby IOninja, Sauce is at github.com/rakudo/rakudo/blob/e114...air.pm#L36
IOninja Right, it calls the method and checks its .Bool matches the .Bool of LHS of ~~ 12:54
m: say 0 ~~ :abs
camelia False
IOninja really
m: say 0.abs
camelia 0
IOninja m: say 0.abs.Bool
camelia False
IOninja m: say 0.Bool
camelia False
IOninja :/ 12:55
m: say 0.Bool === 0.abs.Bool
camelia True
IOninja Oh, doh
m: say 0 ~~ :!abs
camelia True
IOninja :)
IOninja writes up Pair.ACCEPTS 12:56
Umm... I'll open an Issue. There several other candidates too... 12:58
DrForr Incidentally I'm going to start writing Perl6::Tidy today, with the basic editing I can at least get a few things reformatted. 13:01
El_Che \o/ 13:02
DrForr Also submitted a ML talk for YAPC::NA. 13:04
13:13 g0d355__ joined 13:14 ZzZombo_ joined 13:15 mawkish__ left 13:16 ZzZombo left, mawkish__ joined
MasterDuke DrForr++ and Perl6::Critic shortly to follow? 13:17
DrForr Maybe. I expect testing ::Tidy will show up a few design flaws I'll have to rectify. 13:18
El_Che ML as in OCaml? 13:22
DrForr Machine Learning. 13:23
sena_kun El_Che, machine learning.
El_Che thx
tbrowder hi, #perl6 13:24
IOninja DrForr: ML with Perl 6?
\o
sena_kun hi 13:25
tbrowder ref bug fixes, what about a bug smashing party at YAPC::NA 2017?
13:26 cpage_ joined
DrForr With 1500+ I expect itd be a verification party :) 13:26
tbrowder well, as teenagers say, "whatever!" 13:27
verification is good, too
DrForr Definitely.
13:28 cdg left 13:33 lukaramu_ joined
tbrowder i have discovered the power of str.subst-rw and would like to add some more practical examples in the docs if no one objects. i am using it to replace some old c code that uses c strings and temp buffers to work through individual chars and do transformations including adding sets of chars in place of zero or more chars at desired positions. such use as 13:33
insertion and growing the Str i don't think is clear in the docs.
s/subst/substr/ 13:34
moritz sure, go ahead 13:36
13:36 lukaramu left 13:38 espadrine joined 13:40 Ven joined 13:45 laouji joined 13:46 FROGGS left 13:52 ZofBot_ joined, ChanServ sets mode: +v ZofBot_ 13:53 ZofBot_ left, ZofBot left, ZofBot joined, ChanServ sets mode: +v ZofBot
masak an OT question and a trip down memory lane: 13:58
does anyone recall the name of a DOS program that had a browsable list of all the 386 assembly instructions, and all the MS-DOS interrupts?
it had a surprisingly polished ASCII interface. you navigated it mostly with the arrow keys. 13:59
IOninja :o 14:00
.oO( pretty sure I wasn't born at the time.... )
14:01 Ven left 14:02 risou_awy is now known as risou
masak I've been searching half of the morning. confused/frazzled search results tell me it *might* have been called DOSREF 14:04
14:06 llfourn left, Ven joined
tbrowder masak: are you sure it wasn't part of MS's asm package? 14:07
masak I won't rule it out that it was 14:10
that'd explain why it was so polished
also, the confused/frazzled search results seemed to indicate something similar
it's hard to tell, because at that age I didn't exactly buy software off the shelf 14:11
DrForr I'm guessing that at that age I was actually buying software off the shelf. In the form of cassettes. :) 14:12
tbrowder i had it on floppys at one time but i'm afraid i may have dumped it. i'll look when i get homemin a couple of weeks. 14:13
masak \o/
tbrowder there are some dos code sites around... 14:14
El_Che tbrowder: you kept floppies?
wow
tbrowder also the drdobbs site i think is still up
El_Che has an usb floppy reader, but no floppies :)
DrForr has the hole punch for the other side somewhere in storage. 14:15
IOninja damn, simply commenting on code in PR is not enough anymore. You must also "submit changes". Just accidentally noticed my comment from 4 hours ago wasn't ever submitted :( 14:16
El_Che Last time I saw a floppy was on an electronic piano at a firend
friend
tbrowder yes, for a lng time, 5.25 in as well 3.5--wife shamed me into dumping a lot of that--was easy just keeping it on a closet shelf
14:17 mawkish__ left 14:20 risou is now known as risou_awy
masak I kept floppies too, with my first computer 14:20
El_Che the single medium I haven't thrown out are music cd's and dvd's (never went to BD) 14:22
all the rest went in the trash, mostly 3.5 floppies and VHS and music cassettes
stmuk masak: maybe a form of en.wikipedia.org/wiki/Ralf_Brown's...rrupt_List 14:24
I seem to remember a TSR version of that (or something similar) 14:25
masak: HELPPC21.ZIP? 14:30
masak HELPPC21! 14:31
that rings a bell!
stmuk++!
also, lazyweb++
together we are schtrong :D
found it immediately here: www.cs.nyu.edu/~yap/classes/machineOrg/helppc/ 14:32
so happy right now <3
stmuk I still have lots of DOS stuff under dosbox :) 14:33
masak I'm running around old parts of the web, planning what to put together onto a vintage PC for my son, who is turning 3 in December
14:34 wamba left
IOninja Are you planning to raise him on 1980s technology so he doesn't beg for latest smartphone when he becomes a teen? :) 14:35
masak yes to the plan, no to the reason :)
El_Che masak: were you a sociology major? 14:36
masak I just found the tools I grew up with suprior to today's tools
IOninja :)
masak El_Che: am not. I majored in bioinformatics.
14:36 Util joined
El_Che masak: sound like The Borg to me 14:36
:)
masak not entirely unlike the Borg, actually. 14:37
El_Che "sorry son, your life was a sociology experiment". /masak opens the door and shows the 21st century
:)
IOninja :D
masak firstly, the 21st century isn't all that awesome
El_Che and specimen 327a, you can call me Dr. Masak now
b2gills I have a copy of MS Quick C along with the manuals less than 10 feet away from me 14:38
El_Che I threw away my Dos 5.2 books last year
masak secondly, I find there's something alluring and seductive about a computer whose stack I can grok completely
today's Internet-connected browser-based full-stack things don't have that quality 14:39
14:39 yqt joined
El_Che I a lots of litature and human sciences book, but I don't tend you keep IT books that are too outdated 14:39
b2gills If you want some ISA Ethernet cards I have some
masak and so, I'm stuffing everything that gave me the warm fuzzies onto as old an IBM PC as I can possibly find, and giving it to my son
El_Che my girls (7 & 9) love the old games I installed in the living room
masak anyone else here grew up on Turbo BASIC? 14:40
El_Che stuff from the 80s and 90s
masak .oO( SELECT CASE INKEY$ )
b2gills I grew up on learning how to program with batch files
stmuk gorilla.bas :)
El_Che hah
masak: I don't miss fighting with trumpet to get my tcp/ip stack or editing autoexec.bat to free mem to play games though 14:42
b2gills the only Basic I programming in was Visual Basic in the mid to late 90s ☹
MasterDuke gorilla.bas++
masak heh, everyone knows about QBASIC and/or QuickBasic, it seems
I wonder how it was that I came across Turbo BASIC so much earlier 14:43
MasterDuke i started with QBasic, then moved to Turbo Pascal
El_Che I had turbo pascal at the uni
b2gills If you need spare parts like a power supply, I still have some
MasterDuke then tried Java, and it turned me off of programming completely for several years (until i got to college) 14:44
stmuk I mostly played with Turbo Pascal and C
14:44 mawkish__ joined 14:46 llfourn joined
El_Che masak: so the idea is also to install old OS'es on hw? 14:49
masak you mean rather than running DOS in DOSBox or VMWare? yes.
timotimo i'm not sure what the difference between QBASIC and QuickBasic is
but i definitely had one of the two
masak timotimo: for one thing, QuickBasic could compile to .EXE 14:50
moritz timotimo: quickbasic allowed compilation to .exe files, QBAISC didn't
masak moritz: stop that :P
MasterDuke jinx
El_Che masak: you go for the full experience. Nice
timotimo OK, that sounds like i had QBASIC.
moritz masak: you stop that, OK? :-)
stmuk QBASIC was the stripped down version which shipped with DOS 5 and QuickBasic was the full product
masak moritz: would you s... argh
moritz for the record, masak and I haven't met in person for nearly a year 14:51
masak there was also a fun relationship between QBASIC.EXE and EDIT.EXE (or EDIT.COM, don't remember)
yeah, moritz and I have maintained our quantum resonance for months now
moritz qbasic had this fun thing where when you started to write a function, it would automatically open a new window
masak the EDIT.* was a thin shell that delegated to QBASIC.EXE telling it to run an editor without the BASIC programming stuff :P 14:52
moritz but when you saved the file, it was an ordinary, single text file
masak moritz: it also preceded Visual Basic with the "feature" of autocorrecting your code to its own capitalization
stmuk I think edit.exe was qbasic.exe at first and became another program (but could be wrong)
masak feels like a very MS-y feature, for some reason
moritz masak: ah fun. I noticed the similarities, but I just assumed that QBASIC used EDIT.COM, not the other way 'round :-) 14:53
masak could've been the other way around, I can't be 100% sure
but I think QBASIC was the big file
moritz re MS-y feature, yes. Until about 3 years ago, MS tried very hard to make choices for developers 14:54
stmuk it was "qbasic /edit" at first
moritz if you bought in to the whole stack, things were great. But if you tried to compiled projects without visual studio, people looked very funny at you 14:55
stmuk FreeDOS is quite nice now
moritz I'm currently trying to come up with a small project with Inline::Python, which I could use for my "Perl 6 by Example" book 14:56
any suggestions?
masak moritz: please be more specific re "small project"
what are your parameters?
14:56 nadim joined
nadim Good afternoon all. 14:57
moritz masak: I actually thought about that in the past: gist.github.com/moritz/25db24be2e9...332b365c6b 14:58
14:58 skaji left
masak nadim: ahoj 14:58
nadim I go away for a few months and 1000+ commits have happened on P6 , Great work
masak: hej på dig ;)
moritz so, the concept or feature I want to illustrate is the Inline::* family of modules 14:59
14:59 skaji joined
IOninja huh... I though El_Che was nadim 14:59
masak moritz: evaluate poker hand; parse S-expression; count words in article and return reading time (words/100 minutes)
nadim IOninja: I am sneaky but not that sneaky 15:00
MasterDuke moritz: perlpilot has a couple blog posts in the works about using Inline::Python for matplotlib 15:01
timotimo moritz: does it make any sense to go for PyQt for that example?
moritz masak: none of those are things that I would naturally use one of the Inline:: modules for
IOninja nadim: oh, I guess I was misreading this twitter handle then.... twitter.com/nxadm
moritz timotimo: maybe, if PyQt code can be compact
MasterDuke which seems like a good example, using something Python is particularly good at (or has a good module/library for)
masak moritz: print a file with lines numbered (getting the width of the numbers column right -- involves log(L)/log(10))
timotimo it can be, really depends on how much stuff you want to build ;) 15:02
masak er, log(L+1)/log(10)
moritz MasterDuke: not a bad idea; maybe extract some stats from the IRC logs and graph them
masak what MasterDuke++ said :)
MasterDuke oops, it was perlawhirl i think
sena_kun moritz, pygments maybe? 15:03
stmuk Python has NAG libaries doesn't it?
timotimo nags you about what?
nadim I am fixing Data::Dump::Tree after a few bug reports, no change to the code really some "use some::module". one of my test examples dumps a call frame and groaked on dumping Test and Data. Anyone knows what those contain? I will ahh a Stash handler to avoid the error but I'd like to know what those are.
stmuk timotimo: whitespace? 15:04
15:04 risou_awy is now known as risou
moritz fwiw I'm currently planning to have one chapter about Unicode (see today's blog), and one on Inline::Python, and then I'm mostly done with the manuscript 15:04
timotimo haha :)
but really, what is NAG?
moritz something nmerical?
*numerical
numerical algebra, i'd think
timotimo numerical alge bra?
does that include pandas? 15:05
moritz yes
the problem with that is the domain knowledge
stmuk yeah some numerical libs from fortran
timotimo what was the one called where you write formulas in abstract and manipulate them? that's not pandas, right?
right, numpy and scipy have lots of fortran code in them
moritz I can either assume every reader knows why need a SVD, or spend 3 pages explaining it 15:06
neiter feels very satisfactory
stmuk I remember web scraping with python's beautifulsoup as well
moritz sena_kun's idea about pygments has some merits; need to think a bit about it
masak python is supposed to be good at parsing RSS/Atom, too 15:07
timotimo oh how hard can RSS/Atom be, it's just xml! /s
stmuk yeah its simple like SOAP 15:08
masak timotimo: though I note the sarcasm, ISTR the answer is quite interesting
(hint: people aren't good at writing RSS/Atom) 15:09
like, *really* not good
nadim moritz: any simple clarification about what Data and Test are in a Stah?
moritz nadim: what's a Stah?
sena_kun moritz, there are no "big" alternatives, it is a simple and useful task and the API is super-simple. Though, explanation of lexers/parsers can be, well, quite arguable.
moritz
.oO( Alternativlos )
15:10
stmuk I think there are raw socket libraries in python used in security as well
moritz sena_kun: before pygments I used vim with Text::VimColor for hilighting 15:11
(with obvious disavantages, but i works)
timotimo yeah, i imagine those feeds you'll encounter on the web are as bad as some web sites (i.e. html code) used to be just a few years ago
stmuk ah "scapy" 15:12
sena_kun moritz, well, either is fine, I guess, it is just an example. Anyway, I just think that this type of libraries(robust, useful, well-known yet small) should be searched for. :) 15:13
moritz so maybe the example could be to point at a feed aggregator, extract some statistics, and plot the result using matplotlib
sena_kun: yes, agreed
timotimo "requests" is a python library that regularly gets lots of praise 15:15
sena_kun ++ for requests 15:16
stmuk pygame
moritz timotimo: aye; hence the idea to get data from the IRC logs to parse, and extract some stats
timotimo to be quite honest, pygame isn't very good. quite popular, though.
nadim moritz: a Stash (I am going to re-hate this keyboard soon) 15:17
15:17 naxieAlDle joined
moritz nadim: sorry, I still think I'm missing some context here 15:18
MasterDuke timotimo: should i report the code we were playing with last night as a MoarVM issue?
timotimo probably. if only as a discussion platform for how it should be dealt with 15:19
15:19 kurahaupo__ joined
MasterDuke k, will do 15:19
stmuk there are also probably plenty of official python libraries to Google APIs where no perl ones exist :/ 15:20
15:20 risou is now known as risou_awy
nadim Moritz: dd callframe(); that contains a My of type Stash. the Stash is a hash and it has two elements Data and Test. I wonder what those are/contain. also when I send them as an element to a sub, the sub receives them as two elements. I saw this before when I put my nose where I shouldn't; but that happens when one writes a dumper. I am going to write a handler for Stash but I'd like to know what Test and Data are. 15:23
moritz nadim: Test sounds like the module that 'use Test;' imported 15:25
nadim: and Data is likely another package or module, for example if you imported Data::Dump or so, you have Data in your lexical environment 15:26
MasterDuke timotimo: github.com/MoarVM/MoarVM/issues/545, please make any edits you'd like
timotimo wrote a little summary of what the irc discussion contained 15:27
15:28 sufrostico joined
lizmat m: class A {}; say callframe.my<A> # nadim 15:28
camelia (A)
timotimo at least i guess that that's what you linked to, i didn't actually click the link %)
15:29 llfourn left
RabidGravy my my 15:30
ZzZombo_ m: my \my '!!!';say my 15:31
camelia 5===SORRY!5=== Error while compiling <tmp>
Term definition requires an initializer
at <tmp>:1
------> 3my \my7⏏5 '!!!';say my
ZzZombo_ m: my \my='!!!';say my
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed my
at <tmp>:1
------> 3my \my='!!!';say my7⏏5<EOL>
ZzZombo_ m: my \my:='!!!';say my
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed my
at <tmp>:1
------> 3my \my:='!!!';say my7⏏5<EOL>
ZzZombo_ m: my $my='!!!';say $my
camelia !!!
ZzZombo_ :(( 15:32
15:33 laouji left 15:34 laouji joined, laouji left 15:35 laouji joined
naxieAlDle ZzZombo_: what's wrong? :) 15:37
ZzZombo_: oh I see. Cannot create a constant called “my” 15:38
m: my \my = 42; say ::<my>
camelia 42
naxieAlDle hehe
cannot use, rather
15:42 cdg joined 15:53 dogbert17 left 15:54 risou_awy is now known as risou
naxieAlDle github down -_- 15:56
IOninja works fine for me 15:57
naxieAlDle “We track these errors automatically, but if the problem persists feel free to contact us. In the meantime, try refreshing.”
naxieAlDle bangs on F5
IOninja: try this: github.com/rakudo/rakudo/pull/1032/commits
hm, indeed, other parts of it are working 15:58
16:01 mawkish__ left
lizmat FWIW, I've seen that error several hours ago as well 16:02
16:02 mawkish__ joined 16:04 araraloren left, khw joined
moritz nine: I'm having some trouble with Inline::Python, see gist.github.com/moritz/766d05b4e64...f09e80aea8 16:09
first, requests.get($url, headers => %hash) complains that headers isn't a valid named argument 16:10
second, $response.text says unicode isn't callable
I get the impression that accessing attributes doesn't work through the normal method call syntax, because it always tries to invoke the result? 16:11
16:12 MilkmanDan left, MilkmanDan joined, cale2 joined
moritz 'get' here is defined as 'def get(url, params=None, **kwargs):' 16:14
cale2 This should print out "Talk", but instead it prints out "Sub". Any idea why? glot.io/snippets/enplb6nk6h
timotimo .WHO gives you the package that belongs to the object 16:15
rather than the package the object belongs to
moritz &greet.package is maybe what you want 16:16
cale2 ahhh. I'm brainstorming about how to making real docstrings in Perl6, but I think those methods aren't going to be the key
moritz: you're right! 16:17
16:19 Sound left
cale2 Theoretically, Pod is already parsed on a per-file basis. Like it organizes your docs into the groups of files that you've already created, right? 16:19
timotimo pod parsing itself knows nothing about other files, i don't think 16:20
i mean besides the file it's currently parsing
all cross-file stuff lives in either Pod::To::HTML or htmlize.p6
cale2 I'd like to make something like ex-doc in elixir or crystal-lang.org/docs/conventions/..._code.html 16:21
Where is grabs all of the `#|` or `#=` Pod and groups it by packages. Then when it renders, it has all of your packages listed on a sidebar on the left 16:23
16:26 llfourn joined
cale2 I guess I can use $?PACKAGE and $?MODULE for that too though 16:26
IOninja m: (("a" x 1000000) x 1073741824).chars.say 16:29
camelia 0
IOninja huh?
16:30 llfourn left 16:31 mawkish__ left
naxieAlDle 6c: (("a" x 1000000) x 1073741824).chars.say 16:31
IOninja m: (('a' x 1000000) x 1000000).chars.say
camelia 3567587328
committable6 naxieAlDle, ¦6c (16 commits): «0» 16:32
16:32 wamba joined
IOninja Filed: rt.perl.org/Ticket/Display.html?id...xn-1452590 16:34
cale2 m: package Stuff { module Talk { say $?PACKAGE ~ '::' ~ $?MODULE } }
camelia Use of uninitialized value of type Stuff::Talk in string context.
Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful.
in block at <tmp> line 1
Use of uninitialized value of type Stuff::Talk in string contex…
cale2 package Stuff { module Talk { say $?MODULE } }
m: package Stuff { module Talk { say $?MODULE } }
camelia (Talk) 16:35
cale2 m: package Stuff { module Talk { say $?PACKAGE } }
camelia (Talk)
16:39 andrzejku joined
Geth Inline-Python: 71bd44d76f | (Stefan Seifert)++ | 3 files
Support named arguments in wrappers for imported Python packages
16:39
nine moritz: ^^^
Geth doc: 35e69981b2 | (Tom Browder)++ | doc/Type/Str.pod6
amplify the substr-rw examples
16:42
moritz nine: thanks! 16:55
moritz runs zef upgrade Inline::Python 16:57
17:00 araujo left
moritz nine: can I run getattr from python in Perl 6 code somehow? 17:02
timotimo just call ."__getattr__(...)"? 17:04
er
."__getattr__"(...) 17:05
17:05 n1ce left 17:06 Ven left, n1ce joined
timotimo twitter.com/whitequark/status/8379...48/photo/1 <3 17:06
IOninja why quotes?
m: class Foo { method __getattr__ { say "hi" } }.__getattr__ 17:07
camelia hi
nine m: use Inline::Python; my $py = Inline::Python.new; $py.call("__builtin__", "getattr", "a", "title") 17:08
camelia ===SORRY!===
Could not find Inline::Python at line 1 in:
/home/camelia/.perl6
/home/camelia/rakudo-m-inst-2/share/perl6/site
/home/camelia/rakudo-m-inst-2/share/perl6/vendor
/home/camelia/rakudo-m-inst-2/share/perl6
CompUn…
nine err... moritz: use Inline::Python; my $py = Inline::Python.new; $py.call("__builtin__", "getattr", "a", "title")
moritz: or submit a PR exposing the py_getattr sub as a getattr method on the Inline::Python object :) 17:10
naxieAlDle timotimo: that's pretty cool! 17:15
17:20 skids joined 17:24 laouji left, g-k--1-w-m--s-u- joined 17:25 g-k--1-w-m--s-u- left, laouji joined
moritz nine: might do 17:25
17:27 llfourn joined 17:32 llfourn left
timotimo naxieAlDle: isn't unicode fun. 17:34
naxieAlDle well, this particular one makes sense 17:35
except for the name, of course…
timotimo so, er, what is it good for again?
it's for when you want to have two strings that look the same but don't eq? 17:36
naxieAlDle if you don't want stuff to be normalized
or that, yes :DDDD
I didn't think about it this way :DD
timotimo not only look the same
but also each grapheme is composed of the same combiners each time
naxieAlDle sooo, if you simply change the order of your combiners, then this character will also prevent them from being reordered 17:38
amazing stuff
m: say ‘a͏̈’.chars 17:39
camelia 1
naxieAlDle m: say ‘a͏̈’
camelia a͏̈
naxieAlDle we already support it? 17:40
m: say ‘a􏿽xCC􏿽x88􏿽xCD􏿽x8F􏿽xE2􏿽x80􏿽x99 17:41
camelia ä͏
naxieAlDle m: say ‘a􏿽xCC􏿽x88􏿽xCD􏿽x8F􏿽xE2􏿽x80􏿽x99.chars
camelia 1
naxieAlDle mhm
m: say “\c[CGJ]”.uniname 17:42
camelia COMBINING GRAPHEME JOINER
MasterDuke IOninja, naxieAlDle: rt.perl.org/Ticket/Display.html?id=130924 seems related to rt.perl.org/Ticket/Display.html?id=127971 and rt.perl.org/Ticket/Display.html?id=127973 17:44
17:45 cale2 left 17:47 espadrine left 17:55 risou is now known as risou_awy 18:13 cibs left 18:14 cibs joined
IOninja ugh... stupid game. Killed off one of the most lovable characters :( Now I'm sad. 18:17
18:18 laouji left
tadzik which game? 18:18
IOninja Witcher 3 18:19
18:19 pmurias joined
pmurias timotimo: isn't pygame good in it's niche of having a well documented way to make a (toy) 2d game? 18:19
18:21 drbojingle_ joined 18:22 espadrine joined
IOninja MasterDuke: yes, merged 130924 into 127971. But 127973 doesn't really look like a bug, if native `str` don't use the rope thing. If you do nqp::indexingoptimized($a).chars to flatten the ropes first, it's slow AF too. 18:22
timotimo pmurias: it's ... all right i guess 18:23
18:24 drbojingle_ left 18:26 llfourn joined 18:31 llfourn left 18:32 laouji joined
pmurias timotimo: anything better in it's niche? 18:37
j75 Any reason why this code takes 30 seconds to run? I would expect it to take about 10 seconds. perl6 -e 'my @promises ; for ^100 { @promises.push: start { sleep (^10).pick } } ; await(@promises) ' 18:42
Rakudo version 2017.02-157-g4707360 built on MoarVM version 2017.02-18-g5f9d698
lizmat j75: the default is a max of 32 threads
when you start something, it is merely pushed onto a queue for processing when a thread becomes available 18:43
IOninja Default is 16.
lizmat use v6.d.PREVIEW might make this run faster
IOninja: ah, ok :-) 18:44
IOninja github.com/rakudo/rakudo/blob/nom/...er.pm#L195
lizmat as in v6.d.PREVIEW a sleep doesn't block a thread (if I remember correctly)
j75 thanks! 18:48
18:48 BenGoldberg joined
jnthn In 6.d sleep is still the real sleep 18:57
So will block a real thread
For non-blocking, you'd do `await Promise.in(10)`
MasterDuke heh, and i bet optimizing `Int :$!max_threads = (%*ENV<RAKUDO_MAX_THREADS> // 16).Int` to `Int :$!max_threads = (%*ENV<RAKUDO_MAX_THREADS>.?Int // 16)` will make it run even faster! 18:58
IOninja doesn't follow the optimization
m: dd %*ENV<RAKUDO_MAX_THREADS> 18:59
camelia Any %ENV = Any
IOninja m: dd %*ENV<RAKUDO_MAX_THREADS>.^lookup('Int')
camelia Mu
IOninja m: dd %*ENV<RAKUDO_MAX_THREADS>.Int
camelia No such method 'Int' for invocant of type 'Any'
in block <unit> at <tmp> line 1
IOninja nevermind
MasterDuke i was being facetious, but meant saving calling `16.Int` 19:00
IOninja It's slower too
lizmat jnthn: how about adding a :dontblock to sleep() ?
jnthn lizmat: Eww 19:01
lizmat or simply make sleep not block at all ?
jnthn lizmat: I'd rather sleep continues to mean what it traditionally has
lizmat but Perl 6 is not traditional, especially in that respect :-)
jnthn That's not really true. We provide high-level things and low-level things for when people need them. 19:02
lizmat perhaps we need a rem-sleep() :-) 19:03
await Promise.in(10) seems poorly huffmann coded as an alternative to sleep(10) looking towards the future 19:04
jnthn Or we could just use the thing that already works fine? :)
19:04 darutoko left 19:05 regreg left
timotimo pmurias: not sure. would you consider love2d to be in that niche at all? 19:06
19:07 wamba left 19:16 g0d355__ left
Geth doc: 180aaeb300 | (Aleks-Daniel Jakimenko-Aleksejev)++ | 2 files
Docs for ≤ ≥ ≠ unicode ops
19:19
naxieAlDle “AlexDaniel committed with AlexDaniel” pfffffft
e: say 5 ≤ 10
evalable6 True
moritz jnthn, lizmat: have you considered a delay() function which is basically a non-blocking sleep? 19:23
the Perl way is to have separate names for separate operations after all 19:24
lizmat delay would work for me 19:25
I actually worked a long time in a programming language where sleep was called delay :-)
so I should have thought of that
jnthn Does it happen often enough to huffmanize it? 19:26
19:27 llfourn joined
moritz does sleep happen often enough to huffmanize it? 19:32
19:32 llfourn left
moritz Thread.delay(5) # better? 19:32
IOninja Promise.in: 5 19:33
lizmat await Promise.in(5)
you need the await
moritz IOninja: but it won't work as the last statement
because that might not be in sink context, so no auto-await
so it's kind of a dangerous pattern 19:34
BenGoldberg sink Promise.in: 5?
. o O (sink is one fewer char than await)
IOninja What's "huffman coded"? The wiki talks about some sort of encoding:/ 19:35
BenGoldberg Huffman coded is perlspeak for commonly used functions having shorter names.
19:35 Cabanossi left
lizmat things you use more often should be shorter 19:35
IOninja Ah
19:35 Cabanossi joined 19:36 Ven joined
BenGoldberg m: wait; 19:36
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
wait used at line 1. Did you mean 'await'?
BenGoldberg . o O (wait 5) 19:37
lizmat I guess that semantically close enough to await
perhaps "await 5"
as short for "await Promise.in(5)" ?
jnthn It's also been proposed that non-awaitable things that are awaited would just be passed through, fwiw 19:39
lizmat m: await 5 19:40
camelia Must specify a Promise, Channel, or Supply to await on (got a Int)
in block <unit> at <tmp> line 1
BenGoldberg That would be confusing, if something changed from non-awaitable to awaitable (like 5)
lizmat it currently doesn't get passed along
jnthn I know that
I'm saying it's been proposed :)
m: use v6.d.PREVIEW; await 5
camelia Must specify an Awaitable to await (got a Int)
in block <unit> at <tmp> line 1
jnthn Notice how the error is differnt in 6.d
Because we now have made it "anything that does the Awaitable role"
IOninja m: sub prefix:<⏲> { await Promise.new: $^a }; ⏲5; say now - INIT now; 19:41
camelia Default constructor for 'Promise' only takes named arguments
in sub prefix:<⏲> at <tmp> line 1
in block <unit> at <tmp> line 1
lizmat m: use v6.d.PREVIEW; await 5 but Awaitable
camelia Method 'get-await-handle' must be implemented by Int+{Awaitable} because it is required by roles: Awaitable.
in block <unit> at <tmp> line 1
IOninja m: sub prefix:<⏲> { await Promise.in: $^a }; ⏲5; say now - INIT now;
camelia 5.00847205
IOninja hehe, cool
BenGoldberg How much code depends on sleep() doing an actual low-level sleep, and blocking the thread instead of scheduling it for later?
I.e., what would break? 19:42
lizmat .u ⏲
yoleaux U+23F2 TIMER CLOCK [So] (⏲)
BenGoldberg .u clock
yoleaux U+20D5 COMBINING CLOCKWISE ARROW ABOVE [Mn] (◌⃕)
U+20D9 COMBINING CLOCKWISE RING OVERLAY [Mn] (◌⃙)
U+21B7 CLOCKWISE TOP SEMICIRCLE ARROW [So] (↷)
tbrowder ugexe: i'm trying to create a module release with the preferred info in the META6.json file, and can't seem to find the proper pointer to the META6.json file in the release to enter in the ecosystem. Can you tell me how to find it? Also, i want to show both release urls: zip and tar.gz. that seems to pass the META6.json test. What.
19:42 felher left
lizmat BenGoldberg: my feeling is that it would probably unbreak a lot of things :-) 19:42
19:42 felher joined
tbrowder do you think about the desirability of showing both? 19:43
BenGoldberg sleep($n, :block-thread) for the current behavior? 19:44
jnthn It'd put sleep into the category of things where you don't know what thread you're going to be on after it
BenGoldberg Ahh, so that's the problem.
naxieAlDle u: clock
unicodable6 naxieAlDle, U+20D4 COMBINING ANTICLOCKWISE ARROW ABOVE [Mn] (◌⃔)
naxieAlDle, U+20D5 COMBINING CLOCKWISE ARROW ABOVE [Mn] (◌⃕)
naxieAlDle, 64 characters in total: gist.github.com/5abdd8f19be98b8abb...1ab4d8034a
jnthn Which doesn't matter for *most* code 19:46
ugexe tbrowder: while looking at your META6.json file on github go to "history" and choose the top (most recent) item. that will reopen it using its hash-id, and from that page you just click 'view raw'. That will be the raw view of that specific commit meta6.json file 19:47
jnthn I'm just a bit wary about changing the semantics of something that has well-understood ones, is all.
BenGoldberg . o O (sleep($n, :no-context-switch))
tbrowder ugexe: thanks! 19:48
ugexe there is no mechanism for using multiple source urls yet 19:49
BenGoldberg u: ' ' clock ' '
unicodable6 BenGoldberg, U+0027 APOSTROPHE [Po] (')
BenGoldberg, U+0020 SPACE [Zs] ( )
BenGoldberg, 13 characters in total: gist.github.com/a8a929ee1831d4b5a4...ad73b120b9
BenGoldberg u: /' ' clock ' '/
unicodable6 BenGoldberg, Regexes are not supported yet, sorry! Try code blocks instead
nadim timotimo: thanks for the answer
BenGoldberg u: { .index: " CLOCK " } 19:50
unicodable6 BenGoldberg, Found nothing!
BenGoldberg u: { .uniname.index: " CLOCK " }
unicodable6 BenGoldberg, Found nothing!
BenGoldberg u: { .uniname.ends-with: " CLOCK" } 19:51
unicodable6 BenGoldberg, U+23F0 ALARM CLOCK [So] (⏰)
BenGoldberg, U+1F570 MANTELPIECE CLOCK [So] (🕰)
BenGoldberg, U+23F2 TIMER CLOCK [So] (⏲)
ugexe the nice thing about using .zip is that windows has support to unzip archives built into powershell / .net
so if someone replaced all the .git links in the ecosystem with their .zip links on github zef on windows could fetch/extract/test/install without git out of the box 19:55
we should probably be linking to our .tar.gz/.zip anyway though. the only reason not to is because panda doesnt understand non-git sources 19:56
timotimo nadim: i don't remember what i answered you about? :o 19:57
nadim timotimo: thank you ... because moritz did answer :) 20:00
timotimo oh
nadim I have this code: { $m.sort(*.key)>>.kv.map: -> ($k, $v) {$k, ' => ', $v} }
now I want to grep out some of the elements out. Can I do that in a nifty way? 20:01
moritz m: say (:a, :b, :c)>>.kv.perl
camelia (("a", Bool::True), ("b", Bool::True), ("c", Bool::True))
moritz m: say (:a, :b, :c)>>.kv.map(-> ($k, $v) {$k, ' => ', $v }).perl 20:02
camelia (("a", " => ", Bool::True), ("b", " => ", Bool::True), ("c", " => ", Bool::True)).Seq
moritz m: say (:a, :b, :c)>>.kv.map(-> ($k, $v) {$k, ' => ', $v }).grep(*[0] ne 'a').perl
camelia (("b", " => ", Bool::True), ("c", " => ", Bool::True)).Seq
moritz nadim: ^^ like that? 20:03
nadim great! 20:05
20:06 risou_awy is now known as risou
nadim is argument parsing (there was another name) in the form of ->($, $v) passed as part of the block? 20:08
IOninja unpacking
It's a signature on the block, so yes, it's passed
m: -> ($, $v) { &?BLOCK.signature.say }($, $) 20:09
camelia Too few positionals passed; expected 2 arguments but got 0 in sub-signature
in block <unit> at <tmp> line 1
nadim right, unpacking.
IOninja m: -> ($, $v) { &?BLOCK.signature.say }([$, $])
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of $] variable; in Perl 6 please use $*PERL.version or $*PERL.compiler.version
at <tmp>:1
------> 3($, $v) { &?BLOCK.signature.say }([$, $]7⏏5)
IOninja *sigh*
m: -> ($, $v) { &?BLOCK.signature.say }([ $, $ ])
camelia ($ ($, $v))
moritz nine: fwiw I get the impression that Inline::Python introduces unnecessary scalar containers; for example if I return a json-decoded data structure, I get { foo => ($[list elements]) }
sounds like a pre-GLR relict
and it means I can't iterate over %result<foo>, but have to write %result<foo>.list 20:10
nadim Any onw knows where types Data and Test are defined? 20:17
I have this code: multi method get_elements (Stash $s) { $s.sort(*.key)>>.kv.map(-> ($k, $v) {$k, ' => ', $v}).grep{ *[2] !~~ Data && *[2] !~~ Test}}
but it won't compile as those types are not defined in the compilation context
moritz nadim: there can be arbitrary types in any scope you introspect 20:18
nadim OK, but I don't get the point. 20:19
20:20 domidumont left
lizmat nadim: perhaps grep on ^name eq 'Data' ? 20:20
moritz nadim: what are you actually trying to accomplish? 20:21
20:21 dct left
nadim moritz: the method gets called when a Stash is dumped. I want to remove the Data and Test entries as they crash the dumper 20:22
but the ^name is a good idea
20:24 risou is now known as risou_awy
moritz nadim: but that will happen for any lexically defined types 20:25
20:25 TEttinger joined
moritz do { my class A { }; dump A } 20:25
or something like that
maybe you want to instead whitelist names by whether they are available? 20:26
you can a dynamic lookup with ::($name)
m: say ::('nosuchname')
camelia No such symbol 'nosuchname'
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
moritz ... and use exception handle to cover the case where a name isn't available 20:27
blacklisting just two names that happen to occur in your test ignores the wider problem
nadim Indeed, but the method would be added by the user. in this case I am adding metlhods in the dumper to handle P6 types. user defined types handler are added elsewhere
nine moritz: could you file that as a bug please?
moritz nine: will do, I'll try to find a minimal example first 20:28
20:28 llfourn joined
nadim moritz: hmm, I wasn 20:29
moritz: hmm, I wasnt very clear. the method I am adding now it to handle a specific type that is part of P6. For user defined types, the user can define handlers and those are either defined in the user type or in a method declared when the type is in scope 20:30
20:30 dct joined 20:31 Sound joined 20:33 lichtkind joined
moritz nadim: blacklisting still seems a rather fragile approach 20:33
nadim maybe but it is type specific, Stash, and element specific. 20:34
20:35 llfourn left
nadim otherwise i get a:too many positionals passed, 3 expecting 2. I have no idea where the 3rd argument comes from, the code passes one. 20:36
moritz I'm trying to understand this argument. You mean that since it's type specific, fragility is OK?
nadim you said it was fragile. I say it is not. this is applied randomly and on all types. It handles one specific case. 20:37
I'd prefer be able to dump Data and Test but the error is mystic at best.
moritz then I haven't seen the argument yet for it being not fragile 20:38
nadim the other way I have to handle it, I have multiple ways, is to handle Data and Test specificaly rather than Stash
moritz but well, it#s your code
nadim haven't seen an argument for not being fragile ... is a fragile argument. But I am open to other ideas of course. 20:39
Right now it is not working, so it is certainly fragile altogether for Stashes
BTW, it doesn't have to be black listed, the code can transform the non dumpable Data/Test into something else if necessary. Trying to have it run first, polishing comes later 20:41
moritz Data and Test come from code outside of your control, it seems 20:42
nadim Yes it does
moritz nobody gives you any gurantuees that there won't be more symbols like that in future versions 20:43
nadim P6 types or internals
moritz that makes blacklisting fragile
</argument>
nadim Indeed
moritz: : how something like $.my_method($e) generate error: too many positionals is what is makes me pull my hair 20:44
works for other types but not Data and Test
moritz have you tried it with --ll-exception ? 20:48
20:48 lichtkind_ joined
nadim don't know what that is. I have been away from all development for months now, particularely P6, don't remember all the vocabulary either. I guess it will take some time to get back to speed, whatever speed I gathered. 20:49
foound the doc, willl try 20:50
20:51 dct left
naxieAlDle m: say 5 ≤ 8 ≤ 10 # nobody is celebrating? :o 20:51
camelia True
20:53 lichtkind left 20:54 laouji left
Geth ecosystem: 7ojo++ created pull request #303:
Add Powerline::Prompt to ecosystem
20:54
ecosystem: e3612a70c1 | (Jarkko Haapalainen)++ | META.list
Add Powerline::Prompt to ecosystem

See github.com/7ojo/perl6-powerline-prompt
20:55
ecosystem: 2e74da6509 | Altai-man++ | META.list
Merge pull request #303 from 7ojo/patch-1

Add Powerline::Prompt to ecosystem
21:02 Gruber is now known as Grrrr 21:05 riatre left 21:06 bjz left 21:09 hankache joined
robertle I am a bit confused: I am starting to write an app, essentially a big networking monkey. so lots of outgoing and incoming sockets, timeouts etc 21:09
as I said lots at the same time, so I want to use IO::socket::async 21:10
now events on these come in on various threads, right? so to make this safe when modifying internal state as a reaction to an internal event I still need to use locks?
I thought the whole plan was to not use locks anymore? 21:11
jnthn Locks are usually a last resort 21:12
Well, direct use of them, anyway
Depending on the scope of the state, use supply/react and whenever blocks
hankache hello #perl6
robertle ok, looking into react/whenever some more, don't really understand that yet
but also, if async events come in on threads from a pool? why is HTTP::Server::Async doing "start" and then handling stuff there? wouldn' it be more natural to simply call the handler on the IO thread when you have the request? 21:13
jnthn It's a concurrency control and subscription management construct primarily for working with supplies. 21:14
In a web server you'd typically want to just do the concurrency control per connection, so maybe it's something to do with that 21:15
Though there's ways of handling that without start too, but it depends on the overall code design
If you have something like: react { whenever IO::Socket::Async.listen(...) -> $conn { whenever $conn { say "got a packet" } } } 21:16
Then essentially you serialize everything for all connections 21:17
Depending on what you're doing, that may be a good option for you
Since you can't get into any kind of concurrency trouble then
At the same time, if there's any significant computation to do, then you might bottleneck on it 21:18
robertle but the supply itself doesn't do anything with trheads, right? the tap just gets run on whoever calls emit?
jnthn Indeed
A react block only allows a single thread to be inside of a given instance of it *at a time* 21:19
But it doesn't care if the first message is delivered on thread 2, the second on thread 3, etc.
robertle ok, that is what i wanted to do wiuth the lock :) 21:20
sweet
jnthn Also worth noting that if you do something like 21:21
IOninja grumbles
naxieAlDle: some do, apparently: twitter.com/mschatweet/status/8384...7179833344
jnthn whenever $conn { ..blah...; my $promise = start { some-long-running-operation }; await $promise; } }
Then it won't let stuff happen inside the react block untl the long-running operation is done 21:22
But if you wrote `whenever $promise -> $result { }` or so, it means that you don't mind if other stuff happens inbetween
21:23 cpage_ left, cibs left 21:24 zengargoyle joined
naxieAlDle well, 3½ is not going to fly any time soon, and √∛∜ are definitely not the highest priority. But you can always wish :P 21:25
21:25 cibs joined
hankache m: say 3½ 21:26
camelia 5===SORRY!5=== Error while compiling <tmp>
Bogus postfix
at <tmp>:1
------> 3say 37⏏5½
expecting any of:
infix
infix stopper
postfix
statement end
statement modifier
state…
21:26 bjz joined
hankache m: say 2 + ⅒ 21:27
camelia 2.1
naxieAlDle yes, just stick a + in between, as simple as that :P
hankache :)
naxieAlDle just like we don't allow 2$x
or 2x even 21:28
jnthn bbl
hankache m: say 3+½ == 3.5
camelia True
zengargoyle it is '3 and ½'
IOninja wonders how long it will take before people will be whining about √-1² producing "unexpected" results.
zengargoyle not 'three one-half'
21:28 pmurias left
awwaiid Y'all should come to TPC-2017-DC (YAPC::NA 2017)! CFP is still open for a bit, www.perlconference.us/tpc-2017-dc/cfp/ 21:28
IOninja Pfft. Yeah, come and get stuck in the country. No thanks :) 21:29
awwaiid If you can get in I'm pretty sure you can get back out
naxieAlDle IOninja: o, that's a good one :P
IOninja Not worth the risk with current administration. 21:30
naxieAlDle m: say False nand False 21:31
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3say False7⏏5 nand False
expecting any of:
infix
infix stopper
postfix
statement end
statement modifi…
21:32 llfourn joined
awwaiid I'm sad that that is such a huge factor on people's minds. I had a co-worker in Canada who didn't want to come to our office around the election time for that overall reason. On the one hand I don't think it's a realistic risk for most people. On the other ... I'm not the one taking even that miniscule risk so don't feel like I'm in a good position to evaluate. 21:32
21:33 sufrostico left
IOninja I've seen a cancelled conf fly by on Twitter for the exact same reason... 21:33
awwaiid Well for anyone willing to take the risk (even domestic!), I very much look forward to an IRL interaction!
IOninja :)
naxieAlDle wtf is going on there? 21:34
IOninja Where?
hankache across the pond 21:35
IOninja Which pond?
naxieAlDle Atlantic Pond 21:36
hankache atlantic ocean
IOninja So... the where is UK? :)
21:36 llfourn left 21:38 girafe joined
zengargoyle a fucked up electoral system left over from 200 years ago that allows a winner who doesn't hold the majority of the popular vote. 21:39
21:39 bjz left
hankache what's an election? 21:39
:) 21:40
zengargoyle an attempt to balance densely populated areas vs sparsely populated areas from a time when news and communication traveled slowly 21:43
each area gets votes based both on population and votes based on just being a distinct area. in the edge case the total vote can go to the person with the least popular support. 21:45
IOninja But that's not the real reason for electors. 21:46
zengargoyle for the rest, wikipedia is probably better explainer.
IOninja There's nothing "fucked up" about ensuring your country doesn't always get run by how California and New York what it run.
naxieAlDle zengargoyle: well, 2% difference doesn't sound like a good reason to complain 21:47
what “popular vote” is it if roughly the same amount of people voted for another candidate… 21:48
21:49 lizmat left
zengargoyle when more voted for the looser but the winner still won. 21:50
21:50 lizmat joined
zengargoyle anyhow, politics is probably better somewhere else. 21:51
21:57 bpmedley left
zengargoyle see also: www.washingtonpost.com/news/wonk/w...-ever-see/ this is how the system is gemed on the state level. pretty easy to have a blue majority but have red win the election. 21:58
21:58 bpmedley joined
tbrowder ugexe: ok, i'll just use the zip. also, is naming releases "v1.2.3" ok, or should it be "1.2.3"? 21:59
21:59 cdg left 22:01 dct joined
ugexe i dont think it matters what you name them on github 22:02
tbrowder i was thinkink about how zef interprets the info to get a release verssion 22:05
22:10 robertle left 22:13 RabidGravy left 22:14 hankache left
tbrowder ugexe: is the definitive version inside the META6.json or not? and is it critical to have the github release version number the same as that in the META6.json? excuse my perhaps dumb questions but i want to get it entirely correct so i can put a recipe for adding github modules to the current ecosystem somewhere. 22:15
zengargoyle tbrowder: i think zef/panda get version info from the META6.json so look at others for guidance. 22:16
nm :)
tbrowder well at the moment it isn't so standard for entries in the modules list, hence the questions 22:17
ugexe it always comes from inside the META6.json, because thats what CUR uses 22:18
22:18 Ven left, wamba joined
ugexe so you should name your things reflecting the version inside META6.json file 22:19
22:19 Ven joined 22:20 rindolf left
tbrowder makes sense--thanks 22:22
zengargoyle ugexe: how does that match with the :ver<???> inside any modules? 22:30
22:31 dct left, geekosaur left 22:33 geekosaur joined, Ven left, llfourn joined
ugexe CUR has no concept of ver outside of whats in META6.json 22:36
22:38 llfourn left 22:39 Ven joined, Voldenet left
zengargoyle so, inside my module the 'unit class Algorithm::Trie::libdatrie:ver<v0.2>:auth<github:zengargoyle>;' doesn't really do anything and only the META6.json "version" : "0.2" and "auth" : "github:zengargoyle" actually count for anything at the moment at least. 22:42
or maybe later for automated tools that create META6.json from files ala dzil... 22:44
22:44 Voldenet joined, Voldenet left, Voldenet joined 22:49 ZzZombo_ left, ZzZombo joined 22:50 ZzZombo is now known as Guest59914 22:51 ZzZombo_ joined 22:52 llfourn joined 22:54 Guest59914 left 22:55 kurahaupo__ left, kurahaupo__ joined, kurahaupo__ left, ZzZombo_ left 22:56 kurahaupo__ joined 22:58 ZzZombo_ joined
lichtkind_ yay finally panda works 23:06
23:07 risou_awy is now known as risou 23:08 wamba left 23:18 sena_kun left 23:23 dct joined 23:24 risou is now known as risou_awy 23:26 espadrine left 23:41 kurahaupo__ left 23:44 kurahaupo__ joined 23:49 spebern joined