sena_kun AlexDaniel++ 00:00
sena_kun haz some sleep 00:01
AlexDaniel sena_kun: here's the final version: 00:02
00:03 patrickb left
AlexDaniel ah no, damn 00:03
text-align: justify; went wrong here
ok whatever, use the previous one without justify, I already spent too much time on this :D
00:04 rmmm joined
tbrowder .tell jmerelo wordpress is not showing my latest version!!!! wp sucks rocks 00:04
tellable6 tbrowder, I'll pass your message to jmerelo
00:11 stoned75 left, stoned75_ left
tbrowder .tell jmerelo it looks like it's finally taken my update-whew! 00:12
tellable6 tbrowder, I'll pass your message to jmerelo
00:13 colomon left 00:29 sena_kun left 00:38 finanalyst left 00:55 chloekek left 01:15 wildtrees left 01:32 Tirifto left 01:38 zacts joined, zacts left 01:49 colomon joined 01:51 colomon left 02:01 rbt left 02:10 wamba left 02:19 rmmm left 02:51 Kaiepi left 02:53 Kaiepi joined, Kaiepi left 02:54 Kaiepi joined 02:55 Kaiepi left 02:56 Kaiepi joined 03:04 Manifest0 left 03:06 Manifest0 joined 03:08 cognominal joined 03:10 Kaiepi left 03:11 Kaiepi joined, cognomin_ left 03:14 cognomin_ joined 03:18 cognominal left 04:08 stoned75 joined, stoned75_ joined 04:20 cpan-raku left 04:21 cpan-raku joined 04:38 mister_flash joined
mister_flash hi 04:38
What is happening to Perl ???
AlexDaniel mister_flash: seems to be going strong 04:43
mister_flash: perl5 keeps improving and perl6 (now Raku) is getting its own identity :) 04:44
mister_flash So perl6 does not exist yet ? 04:45
perl 5 is the latest perl version ?
AlexDaniel mister_flash: first release of raku (back when it was still called perl6) was in 2015 04:46
mister_flash So perl6 appellation should be ban 04:47
AlexDaniel mister_flash: sorry, what do you mean? 04:48
mister_flash The word "perl6" should be remove everywhere since it causes misunderstandings. 04:50
AlexDaniel mister_flash: yes, correct, that's what we've been doing :)
mister_flash nice. 04:51
AlexDaniel as for perl5, the latest version seems to be v5.30.1 released just a month ago
mister_flash I wonder if Perl will not be deprecated. 04:53
AlexDaniel I don't think perl5 should be deprecated. It is nice and it keeps getting better. 04:54
Raku is a separate language now, yes it is very inspired by perl5 but it doesn't attempt to deprecate perl5
mister_flash but It will shade i 04:56
AlexDaniel I hope it will shade other languages equally :) 04:57
mister_flash If it could shade Python it will be more than a dream 04:58
but I no longer dream 04:59
Xliff Yeah. It has a long road ahead if it is to do that.
AlexDaniel well, I personally already use Raku for writing things for which I would have used Python if Raku didn't exist
Xliff Getting the language bindings and attaining some percentage of module parity will be a big help. 05:00
AlexDaniel I understand it's subjective, but it's already the case for me
Xliff: just a single web framework can do wonders :) 05:03
Xliff: Red seems to be on its way so that's one missing piece done
05:05 stoned75 left 05:06 stoned75_ left
Kaiepi >v5.30.1 05:07
shit the book i got to learn it a while ago is way out of date then
AlexDaniel, what are some of the other missing pieces? are they all in the most wanted modules repo? 05:11
Xliff AlexDaniel: Red also has a long way to go.
But hey, Raku isn't even 5, yet.
I think we're doing quite well.
AlexDaniel Xliff: I'm not sure. I think if someone started using Red seriously it'll get ready very quickly 05:12
so IMO it's ready to be used by somebody :)
Xliff AlexDaniel: Still doesn't cover much of the use cases I'd absolutely need it to cover before I'd use it. 05:13
But it's getting there.
He's working on joins, though. That's a big step.
AlexDaniel Kaiepi: I'm not sure, actually. We have Cro, we have Red, someone just needs to actually make a web framework 05:14
Kaiepi Net::Telnet's probably the closest thing i have to a module that could sway people to use raku since my goal is to make it even better than any existing telnet library 05:15
Xliff AlexDaniel: Um. I thought we had a Web framework. I know Cro could be used to build one, but wasn't there another one? 05:16
AlexDaniel Xliff: no
Xliff Bailador? 05:17
AlexDaniel it's dead
Xliff What? When?
AlexDaniel honestly I don't remember it ever not being dead… 05:19
Xliff Hrm. Better to pick that up than re-invent the wheel?
05:20 rindolf joined
Xliff Honestly, I'd need to see a spec for a decent web framework and try my hand at it. 05:20
AlexDaniel maybe I'm wrong about Bailador 05:22
thing is, I tried it in 2017 or maybe earlier
it did the job, a little bit, but otherwise was dead :)
but now I look at the commit history and there was a lot of stuff going on in 2017-2018 05:23
as for reworking Bailador instead of starting from scratch, I don't think it makes sense 05:24
but please feel free to :)
nowadays you'd probably start with cro and red
mister_flash Mojolicious is actually good, someone could port it on Raku and add some things 05:25
05:32 huf left 05:40 huf joined 06:30 Kaiepi left 06:37 Kaiepi joined, Kaiepi left 06:38 Kaiepi joined 06:39 Kaiepi left 06:40 discord61 joined, Kaiepi joined, discord6 left, discord61 is now known as discord6 06:55 discord6 left, discord6 joined
rindolf AlexDaniel: hi, sup? 06:57
07:13 Kaiepi left 08:06 wamba joined 08:07 Kaiepi joined 08:23 Xliff left 08:48 freezing joined 09:16 Kaiepi left 09:20 shinobicl joined 09:22 Kaiepi joined 09:27 stoned75 joined, stoned75_ joined 09:31 shinobicl left 09:37 stoned75_ left, stoned75 left 10:06 Tirifto joined 10:11 Doc_Holliwood left 10:16 Doc_Holliwood joined
Geth doc: c7ec7cd62e | (JJ Merelo)++ | doc/Language/unicode.pod6
Adds version of Unicode supported

Refs #3104
10:23 ufobat joined 10:38 stoned75 joined, stoned75_ joined 10:40 finanalyst joined
ufobat is Array.clone() supposed to clone multi dimensional arrays deeply? It seems it isn't. But I would assume it should, since you can do a @array[1;2;3] aka It does not feel like it is an Array of Arrays 10:48
10:50 sena_kun joined 11:02 sena_kun left 11:16 sena_kun joined 11:17 stoned75_ left, stoned75 left
Kaiepi m: class Foo { has $!foo is readonly; method set-foo($!foo) { } }; say 1 11:31
camelia Potential difficulties:
useless use of 'is readonly' on $!foo
at <tmp>:1
------> class Foo { has $!foo is readonly; method set-foo($!foo) { } }
Kaiepi m: class Foo { has $.foo is readonly; method set-foo($!foo) { } }; say 1
camelia Nil
Kaiepi m: class Foo { has $.foo is readonly; method set-foo($foo) { $!foo = $foo } }; say 1
camelia 1
Kaiepi what does readonly actually prevent you from doing?
s/readonly/is readonly/ 11:32
the docs say it only applies to the container and not the value itself, but shouldn't $!foo = $foo throw if that's the case? 11:36
11:40 ufobat_ joined
lizmat ufobat: I think .clone is defined to be shallow, with optional args for alternate values for public attributes at the top level 11:40
11:43 ufobat left 11:50 ufobat joined 11:52 ufobat_ left
Kaiepi oh, it doesn't look like this trait is fully implemented for attributes 11:53
it sets its $!ro attribute when it gets applied, but nothing uses this, not even its own readonly method
apart from its default_to_rw method ig 11:56
11:58 MasterDuke left
lizmat shouldn't "is read-only" only apply to public attributes ? 11:58
Kaiepi aren't public attributes already readonly by default? 12:00
m: class Foo { has $.foo }; say Foo.^get_attribute_for_usage('$!foo').get_value(Foo).readonly 12:02
camelia Cannot look up attributes in a Foo type object
in block <unit> at <tmp> line 1
Kaiepi m: class Foo { has $.foo }; say^get_attribute_for_usage('$!foo').get_value(Foo).readonly
camelia Cannot look up attributes in a Foo type object
in block <unit> at <tmp> line 1
Kaiepi m: class Foo { has $.foo }; say^get_attribute_for_usage('$!foo').get_value(
camelia No such method 'readonly' for invocant of type 'Any'
in block <unit> at <tmp> line 1
Kaiepi m: class Foo { has $.foo }; say^get_attribute_for_usage('$!foo').readonly
camelia True
Kaiepi m: class Foo { has $.foo }; say (my $attr =^get_attribute_for_usage('$!foo')).^get_attribute_for_usage('$!ro').get_value($attr) 12:03
camelia No such method 'get_value' for invocant of type 'BOOTSTRAPATTR'
in block <unit> at <tmp> line 1
Kaiepi m: use nqp; class Foo { has $.foo }; say nqp::getattr((my $attr =^get_attribute_for_usage('$!foo')), $attr.WHAT, '$!ro')
camelia P6opaque: no such attribute '$!ro' on type Attribute in a Scalar when trying to get a value
in block <unit> at <tmp> line 1
Kaiepi m: use nqp; class Foo { has $.foo }; say nqp::getattr((my $attr :=^get_attribute_for_usage('$!foo')), $attr.WHAT, '$!ro') 12:04
camelia Cannot find method 'gist' on 'int': no method cache and no .^find_method
in block <unit> at <tmp> line 1
Kaiepi m: use nqp; class Foo { has $.foo }; say nqp::getattr_i((my $attr :=^get_attribute_for_usage('$!foo')), $attr.WHAT, '$!ro')
camelia 0
Kaiepi huh
m: use nqp; class Foo { has $.foo }; say nqp::getattr_i((my $attr :=^get_attribute_for_usage('$!foo')), $attr.WHAT, '$!rw') 12:06
camelia 0
Kaiepi ig they're "readonly" in that they're not rw by default
m: class Foo is rw { has $.foo is readonly; method set-foo($!foo) { } }; 1 12:09
camelia ( no output )
Kaiepi m: class Foo is rw { has $.foo is readonly; method set-foo($foo) { $!foo = $foo } }; 1
camelia ( no output )
Kaiepi m: class Foo is rw { has $.foo; method set-foo($foo) { $!foo = $foo } }; 1 12:10
camelia ( no output )
Kaiepi m: class Foo { has $.foo is readonly; method set-foo($foo) { $!foo = $foo } }; BEGIN Foo.^set_rw; 1
camelia ( no output )
Kaiepi m: class Foo { has $.foo is readonly; method set-foo($foo) { $!foo = $foo } }; BEGIN Foo.^set_rw; (my $foo =^get_attribute_for_usage('$!foo').set_value($foo, 1) 12:12
camelia ( no output )
Kaiepi even in the one case i can find where $!ro is used in any way, it doesn't seem to do anything
12:14 chloekek joined
Kaiepi wait i was missing `is rw` on the last couple tests 12:15
nvm i forgot about the BEGIN that sets what `is rw` sets 12:16
there aren't any tests for using this trait with attributes in roast either, so ig it just isn't supported rn 12:27
12:31 ufobat left
Kaiepi ohh, i was testing the one case where it does get used wrong 12:37
m: class Foo is rw { has $.foo }; say (try = 1) // 0 12:38
camelia 1
Kaiepi m: class Foo is rw { has $.foo is readonly }; say (try = 1) // 0
camelia 0
Geth problem-solving/CoC: 48f4c51ebe | (Elizabeth Mattijsen)++ | solutions/meta/
Rework the "Responsibility" section
13:02 sena_kun left 13:04 stoned75 joined, stoned75_ joined
Kaiepi ok i was definitely misunderstanding what the trait's for after reading through the design docs 13:08
i was expecting it to work like ::= would if it were implemented
13:15 normanrockwell joined, normanrockwell left 13:17 sena_kun joined, sena_kun left 13:30 MasterDuke joined
AlexDaniel rindolf: o/ 13:37
rindolf AlexDaniel: sup? 13:42
AlexDaniel I woke up :) 13:45
chloekek was thinking about implementing symlinkfs; a FUSE file system that stores all data in symlinks, to trick du 13:47
du reports 0 for symlinks, but they can store like a kilobyte of data. :)
AlexDaniel m: say "a\x[E01E8]b\x[E01E9]" 14:01
camelia a󠇨b󠇩
AlexDaniel chloekek: similarly, use variation selectors to mix binary data into text? :) 14:02
14:09 sena_kun joined
cpan-raku New module released to CPAN! Operator::dB (0.0.95) by 03ALLSOPP 14:20
New module released to CPAN! HTML::BoreDOM (0.0.6) by 03ALLSOPP
14:30 curiouscreamkrac joined 14:48 curiouscreamkrac left 15:02 sena_kun left 15:13 normanrockwell joined 15:18 sena_kun joined
Geth whateverable/master: 5 commits pushed by Altai-man++, (Aleks-Daniel Jakimenko-Aleksejev)++ 15:23
whateverable: f5dcb9b653 | (Aleks-Daniel Jakimenko-Aleksejev)++ | services/whateverable-all.service
Add Sourceable service
whateverable: daf36f299b | (Aleks-Daniel Jakimenko-Aleksejev)++ | xbin/Sourceable.p6
Make Sourceable executable
sena_kun \o/ 15:31
tellable6 2019-12-01T11:03:29Z #raku-dev <jmerelo> sena_kun I've done a bit of editing here and there. Other than that, perfect!
15:41 normanro_ joined 15:44 normanrockwell left
stoned75 I was looking for WHAT method (or whatever) definition. I found a method definition in rakudo/src/core.c/operators.pm6, which leads me to MoarVM/src/6model/bootstrap.c:bootstrap_KnowHOW(). Am I on the right track ? 15:47
15:52 patrickb joined
stoned75 then I wonder why operators.pm6 has a comment saying "XXX Wants to be macros when we have them." when rakudo/docs/ChangeLog says for 'New in 2011.09': "+ .WHAT and friends as macros" 15:52
15:52 normanrockwell joined 15:56 normanr__ joined, normanro_ left
sena_kun stoned75 what do you want to do with WHAT? It is Rakudo-specific thing, maybe you want something else? What's the task? 15:58
stoned75 eh. only curiosity :) 15:59
16:00 ufobat joined, normanrockwell left
stoned75 also, doc/Language/objects.pod6 says "... by calling the WHAT method, which is actually a macro in method form" so I admit I am a bit confused 16:05
AlexDaniel m: Int.^name 16:08
camelia ( no output )
AlexDaniel m: say Int.^name
camelia Int
AlexDaniel m: say 42.^name
camelia Int
ufobat shallow clones suck :-/ 16:09
Geth problem-solving/CoC: e2b18bf640 | (Elizabeth Mattijsen)++ | solutions/meta/
Make discrimination clause more specific
16:28 ufobat left
chloekek p6: constant 無 = Mu; sub f(無 $x) { say $x }; f(1 & 2) 16:28
camelia all(1, 2)
lizmat m: say "無".uninames 16:29
lizmat hmmm.. :-)
ah, nothing! 16:30
chloekek It’s the Japanese spelling of the term where the name Mu comes from according to the docs.
sena_kun - what is 無? - ah, nevermind, it's nothing. 16:32
lizmat is there an ideograph for nothing + 1 ? 16:33
chloekek constant 任意 = Any; sub f(任意 $x) { say $x }; f(1 & 2) 16:36
evalable6 1
chloekek Bad name. 16:38
lizmat I was thinking of adding another easter egg for it :-) 16:41
sena_kun m: constant 何なり = Any; sub f(何なり $x) { say $x }; f(1 & 2) 16:43
camelia 1
lizmat m: my $beer = "🍺"; $beer++; dd $beer 16:44
camelia Str $beer = "🍻"
chloekek does use 「」 very frequently.
Also qq「」 for interpolation. 16:45
Currently programming in Perl 5 though. 16:51
No such goodies :(
lizmat m: my $perl = "🐪"; $perl++; say $perl # so glad we didn't turn that into a butterfly 16:54
camelia 🐫
chloekek lmao
m: my $perl = "🐫"; $perl++; say $perl 16:56
camelia 🐫🐪
chloekek m: my $perl = "🐫"; $perl++; $perl++; say $perl 16:57
camelia 🐫🐫
chloekek m: my $perl = "🐫"; $perl++; $perl++; $perl++; say $perl
camelia 🐫🐪🐪
chloekek Not quite base 2 it seems.
17:00 sena_kun left
lizmat it is if you consider "🐪" and "🐫" one ? 17:04
m: my $perl = "🐪"; say :2(($perl++).trans("🐫🐪" => "10")) for ^10 17:05
camelia 0
chloekek p6: my $perl = „🐫“; for 1 .. 10 { say $perl++ } 17:06
camelia 🐫
chloekek It never flips the initial digit.
kurahaupo looks like counting in binary with 🐫 == 1, 🐪 == 0 ? 17:07
lizmat it does from 0 to 1
and in all other cases, the first is the most significant, so always a 1
chloekek p6: my $perl = „🐫“; for 1 .. 10 { ($perl++).flip.say } 17:08
camelia 🐫
chloekek p6: my $perl = „🐫“; for 1 .. 10 { ($perl++).substr(1).flip.say } 17:09
chloekek p6: my $perl = „🐫“; for 1 .. 10 { ($perl++).substr(1).say }
chloekek p6: my $perl = „🐫“; for 1 .. 10 { ($perl++).say } 17:10
camelia 🐫
chloekek Oh yeah of course the MSb is always 1 lol otherwise it would not show up.
17:11 sergot left, freezing left, matiaslina left, CIAvash left, stoned75 left, discord6 left, AlexDaniel left, vrurg left, jjatria left, libertas left, moon-child left, pochi left, skyl4rk left, guifa left, marcusr left, markmarkmark left, tejr left, jnthn left, KotH left, renormalist left, elcaro left, patrickb left, stoned75_ left, finanalyst left, Doc_Holliwood left, Geth left, Ekho left, robertle left, daxim left, surrealpie left, ilbelkyr left, huf left, cognomin_ left, Manifest0 left, silug left, dmc00 left, veesh left, mid_home left, lgtaube left, nebuchadnezzar left, TreyHarris left, eythian left, rba[m] left, unicodable6 left, committable6 left, Demos[m] left, Nasrudin42 left, kawaii left, domm left, normanr__ left, Kaiepi left, rindolf left, johnjay left, maettu left, xinming left, [Coke] left, japhb left, wbn left, Grrrr left, lucs left, moritz left, jdv79 left, El_Che left, leah2 left, b2gills left, sjn left, zeylos left, pat_js left, broquaint left, Ulti left, uzl[m] left, timotimo[m] left, quotable6 left, bloatable6 left, benchable6 left, coverable6 left, reportable6 left, greppable6 left, nativecallable6 left, statisfiable6 left, shareable6 left, squashable6 left, tellable6 left, chsanch left, kubrat left, rba left, epony left, dpk left, zostay left, Grinnz left, MasterDuke left, mister_flash left, Tirifto left, wamba left, cpan-raku left, lizmat left, releasable6 left, notable6 left, bisectable6 left, maggotbrain left, MitarashiDango[m left, tyil[m] left, kybr left, evalable6 left, ZzZombo left, AlexDaniel` left, sarna left, BinGOs left
chloekek Welp. 17:12
17:12 normanr__ joined, patrickb joined, MasterDuke joined, stoned75_ joined, stoned75 joined, finanalyst joined, Doc_Holliwood joined, Tirifto joined, Kaiepi joined, freezing joined, wamba joined, discord6 joined, huf joined, rindolf joined, mister_flash joined, cognomin_ joined, Manifest0 joined, silug joined, AlexDaniel joined, ZzZombo joined, Geth joined, sergot joined, lizmat joined, Ekho joined, vrurg joined, dmc00 joined, robertle joined, veesh joined, mid_home joined, quotable6 joined, releasable6 joined, notable6 joined, unicodable6 joined, bloatable6 joined, committable6 joined, benchable6 joined, coverable6 joined, bisectable6 joined, reportable6 joined, nativecallable6 joined, greppable6 joined, statisfiable6 joined, shareable6 joined, squashable6 joined, skyl4rk joined, jjatria joined, johnjay joined, maggotbrain joined, lgtaube joined, libertas joined, uzl[m] joined, CIAvash joined, matiaslina joined, Demos[m] joined, rba[m] joined, AlexDaniel` joined, tyil[m] joined, MitarashiDango[m joined, timotimo[m] joined, maettu joined, daxim joined, xinming joined, surrealpie joined, moon-child joined, dpk joined, pochi joined, Nasrudin42 joined, kybr joined, nebuchadnezzar joined, [Coke] joined, sarna joined, tellable6 joined, japhb joined, TreyHarris joined, El_Che joined, evalable6 joined, guifa joined, leah2 joined, wbn joined, Grrrr joined, marcusr joined, b2gills joined, epony joined, markmarkmark joined, sjn joined, eythian joined, tejr joined, zeylos joined, jnthn joined, kubrat joined, lucs joined, KotH joined, ilbelkyr joined, camelia joined, dogbert17 joined, cxreg joined, tony-o joined, bartolin joined, mspo joined, eiro joined, SyrupThinker joined, nine joined, brass joined, tadzik joined, skaji_ joined, dotdotdot joined, masak joined, tyil joined, Altreus joined, tobs joined, Hotbees joined, kurahaupo joined, xkr47 joined, klapperl joined, go|dfish joined, a6502 joined, k-man joined, xi joined, timotimo joined, samebchase joined, pat_js joined, Ulti joined, elcaro joined, renormalist joined, moritz joined, broquaint joined, kawaii joined, chsanch joined, domm joined, jdv79 joined, BinGOs joined, rba joined, zostay joined, Grinnz joined 17:15 sena_kun joined 17:17 cpan-raku joined, cpan-raku left, cpan-raku joined
.oO( chloekek made me do it: )
chloekek XD
17:40 AlexDaniel left 17:41 AlexDaniel joined, AlexDaniel left, AlexDaniel joined, a6502 left
chloekek Now make a rosettacode exercise and show off how consise it is in Raku 17:42
17:42 skyl4rk left 17:48 skyl4rk joined 17:51 mister_flash left 17:53 skyl4rk left 17:56 skyl4rk joined 18:00 chloekek left 18:01 a6502 joined 18:26 ufobat joined
ufobat m: my @a[3;3]; say @a.clone.shape 18:52
camelia (3 3)
ufobat ^ lizmat
18:57 jmerelo joined, lucasb joined 19:00 sena_kun left
jmerelo Thanks, lizmat, for posting the advent calendar entry in Reddit 19:05
19:05 chloekek joined
jmerelo Everyone else, comment, check, spread the word, check out the Advent Calendar 19:06
In ~ 4 hours, a fresh new entry by tmtvl
El_Che jmerelo++ 19:10
tbrowder El_Che: thnx! 19:12
19:17 sena_kun joined
jmerelo tbrowder: it's getting a few visits, the problem is the old calendar is getting even more right now... 19:18
So we really need to generate some buzz about it: spread it over email groups, social media, and all over...
moritz I'll post a quick note on the old one that there is a new site 19:20
jmerelo great, moritz. Thanks. 19:21
Please also accept the invitation for the new site, and connect it when you do with raku_news, please. 19:22
patrickb jmerelo: Is there a reason we don't try to get a stackoverflow alias perl6 -> raku up? I seem to recall you were on this some time i the past, not sure though... 19:24
jmerelo patrickb: I requested write permission to do it by hand, and got burned, first
patrickb: second, we can create an alias, but there's some numbers involved, I think there must be a balance between old and new name, somethign like 30% difference 19:25
patrickb: and then we need 10k karma, but after the last recomputation of karma, I can do it.
patrickb: let me see if I can do it now, but I think I will not be able 19:26
now I remember 19:27
We can't create a synonym for perl6 because it's a "version specific synonim" 19:28
I guess it's like creating a "debian7" synonyn for whatever, like "sid"
Let me try in the other direction
moritz jmerelo: do you know where I can do the twitter thing you mentioned? 19:29
jmerelo patrickb: no, I can't: "Can't create synonym; the 'perl6' tag appears more than 1.25 times the 'raku' tag"
patrickb jmerelo: Ugh... that sounds like it shouldn't apply to our case. The 6 in Perl6 is no version number. Should I try to bug some SO staff about it?
jmerelo So we need to delete many perl6 tags, and add many raku tags
patrickb: that's apparently the only way to do it. 19:30
Grinnz you should be able to get a moderator to do it
moritz ah, 19:31
jmerelo moritz: great. Thanks!
patrickb I think a question is the way to go. I'll create a question. 19:34
jmerelo patrickb: good luck 19:35
patrickb: and thanks.
patrickb There already is a respective question: 19:36
Please up vote!
El_Che favorited and upvoted 19:39
jmerelo patrickb: the thing is, "synonyms" does not quite reflect what StackOverflow does. It's not symmetrical, for instance. 19:40
If we suggest perl6 as a synonym for raku, every time someone tags something with raku, the system will tell them if they are sure about that. 19:41
What we want is raku as a synonym for perl6. Which we can't do due to their "no synonyms for versions" policy.
patrickb jmerelo: I think it's the other way round. 19:42
jmerelo patrickb: hm
El_Che If someone has time, please try the tar.gz relocable install of rakudo. It's a new feature and I think it's a killer one. So please test it:
(there is also a 32-bit version if someone is running 32-bit linux: 19:43
patrickb El_Che: You are doing relocatable builds? - Great! I was just about starting to do those my self. (literally now)
Grinnz jmerelo: a moderator can make it the correct direction regardless, afaik
jmerelo patrickb: here's the list of synonyms and "targets" 19:44
El_Che patrickb: yes, I started from your notes!
jmerelo patrickb: so what we want is a target raku for the synonym perl6... So right, we need to create perl6 as a synonym for raku. 19:45
In that case, what we need is 1. Delete perl6 tags, add raku tags 2. Suggest perl6 as a synonym 3. Vote. Moderator might be faster, but it's still doable.
So go back to your old questions and retag them, and if you've got the privs, do that to other questions too. 19:46
patrickb jmerelo: I thinkg the meta question I linked above is the way to go. We just need a moderator to have a look at it. It already has the 'support' tag added to it, so I guess upvoting and waiting is all we can do.
jmerelo: Just as you say. :-)
jmerelo patrickb: not all... we can still re-tag questions and then we won't need moderators.
patrickb El_Che: I am rebuilding the website to also offer such precompiled packages as download. Would you be willing to to the builds for that? 19:47
El_Che how many point do we need to be allowed to remove tags? 19:48
patrickb: sure, they are already automated on travis
patrickb Woot!
WRT naming of the files: What I currently have in my changes is rakudo-[version]-[platform]-[architecture].[zip|tar.gz] where platform is [win|macos|linux] and architecture is [x86|x86_64]. 19:50
19:51 rindolf left
patrickb I'd try not to do "32bit" or similar, as that could equally well be arm64 vs arm32 or so. Better be literal about the architecture. 19:51
El_Che: Do you think it's worth it to have the "-nojit" stuff in the name? I'm leaning to the side of "not worth it". 19:53
El_Che from a coherence point of view: no
from a pragmatic yes: it's a warning as "don't be discouraged if it's slow, we know" 19:54
patrickb: the name of the file can change if needed
(I am in the processing of moving the create-pkgs-with-docker part to a different repo (now private) and use that in rakudo-pkg) 19:55
patrickb El_Che: You have a point. Can we decide on a defined format? (The website relies solely on the name to determine what's in it). What about rakudo-[version]-[platform]-[architecture]-[comment].[zip|tar.gz] Where the -[comment] part is optional? 19:56
El_Che works for me
we can do even do without the comment
and point it out in the download page 19:57
I have worked somewhat on the windows variant, but it has been painful
relocatable --> split out pkg creation -> windows -> macox 19:58
1 is done, 2 is written but need finetuning
patrickb El_Che: You even fixed up the raku and raku-debug sym links. :-)
El_Che 3: the situation in windows is bad, so it needs to be done
19:58 jmerelo left
El_Che 4: homebrew has regular rakudo next to rakudo-star, so for homebrew users it's ok for now 19:59
patrickb: yeah, my travis builds keep failing :)
patrickb El_Che: Did the windows instructions in work when done manually? 20:00
El_Che well, no because I was doing them on travis 20:01
the idea was that the builds happen there 20:02
but it's a painful process
I used your notes, though
but things that are easy on linux/mac are *hard* on windows 20:03
finding the right make/nmake
from all the soft installed
patrickb I tried doing it in a windows docker container (Microsoft offers those!). But failed because the small containers miss a lot of plumbing...
El_Che running find / -name "*make.exe" on travis runs to find it :)
do those work on travis or other free-for-foss providers? 20:04
20:05 Tirifto left
El_Che patrickb: This is as far as I got: 20:05
patrickb El_Che: It requires "Docker for Windows" which requires putting Windows itself in a special mode. In that mode most other virtualization technology doesn't work. So I doubt it it will work on travis.
El_Che patrickb: there is a problem with the right perl to use
patrickb: modules mixed and so on
patrickb: so I need to get some paths right
patrickb: hyperv? 20:06
patrickb hyperv rings a bell. That might be it
El_Che that's a no-no then
travis (beta) and appveyor offer windows support 20:07
I am kind of invested on travis so that's my target nr 1, but others are possible
since I discovered the parallel builds I am happy :)
patrickb I'm all for doing what works. 20:08
El_Che rakudo-pkg does 24 builds for every commit 20:09
so serial building is not an option = after 50min you have an automatic fail
patrickb El_Che: The rakudo builds you linked work on my machine. You saved me the trouble of doing these builds myself (I need them for some work on rakudobrew I'm doing) 20:11
El_Che patrickb++ 20:12
about the format: rakudo-[version]-[platform]-[architecture].[zip|tar.gz]
there is also a revision needed
if we change something besides the release; new version of zef, e.g. 20:13
patrickb What's a revision?
El_Che like packages do:
patrickb like a packaging point release?
El_Che yes
in this case, zef is included 20:14
what if we need to bump zef but rakudo stays on the same version
e.g. a bug or change in the ecosystem
patrickb So that's what the -01 in your filenames was about. 20:15