»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_log/perl6 | UTF-8 is our friend! 🦋
Set by Zoffix on 25 July 2018.
discord6 <RaycatWhoDat> m: say 42; 00:23
00:24 pamplemousse joined
discord6 <RaycatWhoDat> Not sure if these messages are getting through. 00:24
<RaycatWhoDat> But it's a shame that the IRC bot doesn't trigger from the bridge
Elronnd I would probably add a separate evaluation bot to the discord, instead of having the irc bot respond 00:30
that way each network is self-contained
AlexDaniel and other bots? 00:35
we have 18 in whateverable project
no, the bots should understand messages from the discord bridge, that's for sure
tobs: ↑ something important to implement :) 00:36
RaycatWhoDat: we'll get it working eventually, sorry :)
tobs: somewhere in Whateverable.pm6 we need `irc-privmsg-channel` which redispatches messages from discord6 00:38
tobs: I think redispatching to this is going to work: github.com/perl6-community-modules...t.pm6#L233 00:39
it's a private method, but… so what?
.tell tobs colabti.org/irclogger/irclogger_lo...-08-26#l10 00:41
tellable6 AlexDaniel, I'll pass your message to tobs
00:45 aborazmeh joined, aborazmeh left, aborazmeh joined 01:04 aborazmeh left 01:32 zakharyas joined 01:33 molaf left 01:36 mahafyi left 01:40 MilkmanDan left 01:42 MilkmanDan joined 01:43 epony left 01:45 molaf joined 01:49 Kaiepi joined 01:58 cpan-p6 left, cpan-p6 joined, cpan-p6 left, cpan-p6 joined 01:59 pamplemousse left 02:02 mowcat left 02:04 Manifest0 left, Manifest0 joined 02:31 Xliff left 02:34 epony joined 02:58 Herby_ joined
Herby_ o/ 02:58
03:04 Kaiepi left 03:16 Herby_ left 03:29 astj joined 03:40 aborazmeh joined, aborazmeh left, aborazmeh joined 03:58 [Sno] left 04:09 irced joined
irced hey all i can't seem to make a simple grammer that parses a block of text surrounded by { } where the inside text may or may not be surrounded by { } without it matching beyond the closing } to a different closing curly brace. you dig? 04:11
like say for the case of illustration i define a struct within a struct in a c source code then outside of that initial struct i do the same. only i want to match the structs (outer structs) separately 04:14
seems the ~ surrounding operator will overreach 04:15
consider for those who know c++ a namespace. so i have like namespace MyDomain { blah blah namespace InnerDomain { blah blah struct { ... }; }; }; followed by maybe class MyClass { ... } 04:19
given grammar namespace { regex TOP { .* \n namespace <namespace-token> <block> .* } token namespace-token < <.ws> <ident>+ <.ws> } regex block { '{' ~ '}' .* } } I can match namespace-token and block but block includes the MyClass { ... } 04:21
surely such parsing is very common where one wants to capture a specific grouping. i imagine i can eventually find a regex book or something that gets into it. but is there some idiomatic perl6 way perhaps ? (using perl6 regex special sauce) 04:23
irced can't seem to spell grammar 04:27
04:41 camelia left
irced not camlelia! 04:42
moritz irced: using .* in a grammar is basically always wrong
irced moritz: k, thanks for the tip
moritz at the very least, ask yourself what *can't* be inside the block (for example), and make a negated character class out of it 04:44
irced moritz: sounds like a good way to go about things if only i can stay inside the outside { }'s 04:45
maybe this particular problem is not one a regex would most facilely address. i suppose i could just iterate and track opening and closing braces 04:53
as an action perhaps 04:54
to get the best of both worlds
track == count 04:55
count == +/-
irced still explores the regex way tho. 04:56
irced has got some ideas about the negated character classes, try 1002 04:58
05:05 molaf left
irced moritz: i'm making some progress here especially taking the focus off of too-encompassing metacharacters 05:10
moritz: thanks
05:22 Kaiepi joined
irced k 05:23
05:55 robertle left, jaldhar joined 06:15 lizmat joined, lizmat left 06:46 aborazmeh left 06:53 vike left 06:58 vike joined 07:00 k0Le left 07:01 k0Le joined 07:07 domidumont joined 07:31 Doc_Holliwood left 07:34 kensanata joined
El_Che irced: :q! to escape vim :) 07:35
07:36 cpan-p6 left 07:37 cpan-p6 joined, cpan-p6 left, cpan-p6 joined 07:43 Doc_Holliwood joined 07:54 patrickb joined
irced El_Che: hahah 07:55
07:55 robertle joined
irced El_Che: funny how you got that seeing that i didn't write k hkl as jl would make a funny insertion 07:57
07:59 irced left 08:00 astj left
El_Che :) 08:00
08:02 jjmerelo joined 08:04 [particle] joined 08:07 [particle]1 left
jjmerelo Hey 08:08
Just to warn you about some bumpy road ahead for the documentation github.com/perl6/doc/milestone/2 08:09
We are going to be using mainly rba's staging servers and maybe others for testing, but behond the milestone there's going to be a serious overhaul of the documentation repo.
Your input (and help) is welcome during the process 08:10
08:13 Black_Ribbon left
rba jjmerelo: As part of your tasks, may you please publish to docs-prod as well, as I like to take over docs.perl6.org soon, to offload the temp server. 08:14
jjmerelo rba: OK 08:15
08:23 sena_kun joined 08:31 ufobat joined
Geth doc: JJ assigned to antoniogamiz Issue Test documentable github.com/perl6/doc/issues/2972
8135a4442b | Antonio++ (committed using GitHub Web editor) | doc/Language/syntax.pod6
08:42 abraxxa joined 08:50 jaldhar left 08:51 jaldhar joined, rindolf joined 08:56 jaldhar_ joined, jaldhar left 08:57 aborazmeh joined, aborazmeh left, aborazmeh joined 09:04 aborazmeh left
synopsebot Link: doc.perl6.org/language/syntax
doc: a857488049 | Antonio++ | doc/Type/independent-routines.pod6
Change the value of kind to Language, fix #2973
synopsebot Link: doc.perl6.org/type/independent-routines
09:25 pecastro joined 09:37 migo joined 09:41 domidumont left 09:48 pat_js joined
migo Hi, I have problem with mysql connector, im returning iterator from method `return $!sth-aggregate.execute( $created-at, :finish ).hashes;` and then iterating over it using `for $message-stats-repository.aggregate($time-range) -> %row {...` and finally getting at "random" row error `Cannot look up attributes in a DB::MySQL::Native::ResultsBind type 09:49
object` any ideas what wrong?
tyil AlexDaniel: why fix it in IRC::Client, when the bot could check for / ^ [ "<" $nick ">" ]? "m:" / (or something like that, basically, check for an optional nick in front of m:) 09:50
09:54 pilne left 09:58 finanalyst joined
Geth doc: 735a673f87 | (JJ Merelo)++ | htmlify.p6
Adds deprecation notice to disambiguation files closes #2974
jjmerelo hi, finanalyst
finanalyst jjmerelo: hi. 10:02
10:14 jjmerelo left
cpan-p6 New module released to CPAN! WebService::Soundcloud (0.0.8) by 03JSTOWE 10:21
10:24 pat_js left 10:27 zakharyas left 10:34 rindolf left 10:51 rindolf joined, finanalyst left 11:01 pat_js joined 11:05 passingby joined, passingby left 11:06 jaldhar_ left 11:07 jaldhar_ joined 11:26 jaldhar_ left 11:27 jaldhar_ joined
Geth whateverable: taboege self-assigned Bots should respond to messages from the discord bridge github.com/perl6/whateverable/issues/356
4d49f92814 | Xliff++ | lib/Whateverable/Processing.pm6
11:53 dumarchie joined
whateverable: 889b1e8860 | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | lib/Whateverable/Processing.pm6
Merge pull request #347 from Xliff/master

  - Process code in between ```...``` markdown tags.
12:02 robertle left 12:04 robertle joined 12:19 kensanata left 12:20 migo left
AlexDaniel tyil: I didn't say it should be fixed in IRC client, I just pointed out which IRC::Client method should be called from whateverable 12:22
tyil ah, okay 12:23
AlexDaniel and I think tobs we'll get it working soon :)
tyil awesome! 12:27
tobs no ETA yet. I wrote the code, now I have to write the tests :) 12:37
tellable6 2019-08-26T00:41:49Z #perl6 <AlexDaniel> tobs colabti.org/irclogger/irclogger_lo...-08-26#l10
AlexDaniel tobs: maybe somewhere here: github.com/perl6/whateverable/blob...e.pm6#L148 12:38
tobs: these are used for every bot
feel free to add tests to specific bots too :) 12:39
tobs oh great, I was looking in a completely different place
AlexDaniel tyil: which nick discord6 will use if discord6 is taken?
tyil: discord6` or discord6_ or … ? 12:40
12:41 [Coke] left
AlexDaniel tobs: and you can probably do $!irc-client.nick(‘discord6’) 12:41
12:50 mahafyi joined 12:54 domidumont joined
tyil discord6_ iirc, but we can test that 13:00
13:00 discord61 joined, discord61 left, discord61 joined
tyil AlexDaniel: also, it should always have the cloak "perl6/bot/discord6" 13:01
it seems to have become discord61
13:01 discord61 left
AlexDaniel hah 13:01
tobs the numbering seems fine. AFAICS Testable does the same thing
but I'm not sure if I should match all of discord6\d+… 13:02
AlexDaniel we can't check for cloaks yet…
tyil oh
that removes that option, then 13:03
AlexDaniel tobs: you can slap `1?` into the regex
might save us a little bit of frustation in the future :)
tobs: fwiw testable needed to have random nicknames (with numbers) when a single irc server was used for testing all the bots 13:04
tobs: now that it starts a new irc server on a random port each time, there's no need for that
13:04 jaldhar_ left 13:05 jaldhar_ joined
tobs ah, ok 13:05
AlexDaniel which is why you can now do .nick(‘discord6’) without worrying that it'll clash with some other process :) 13:06
tobs: or, you can start a new irc-client: github.com/perl6/whateverable/blob...ble.t#L230
13:07 pat_js left 13:09 molaf joined 13:19 robertle left
tobs e: use IRC::Client; IRC::Client.new.nick('x') 13:22
evalable6 (exit code 1) ===SORRY!===
Could not find IRC::Client at line 1 in:
tobs, Full output: gist.github.com/5ff0213b96467cf6ec...7e8668cb25
tobs ah
anyway, did anybody use the .nick method recently? It gives me a "wrong number or arguments error"
AlexDaniel tobs: yes 13:24
tobs: I mean, this test works: github.com/perl6/whateverable/blob...ble.t#L251
tobs touché 13:25
13:28 molaf left 13:33 lucasb joined 13:34 |oLa| joined 13:40 vike left
Geth whateverable: f1563a651f | (Aleks-Daniel Jakimenko-Aleksejev)++ | 4 files
Fix code block processing in md files

This time with tests to make sure it actually works.
13:46 grondilu joined, committable6 left, vike joined
grondilu m: say "1 2 3" ~~ m/<digit> ** 3/ 13:46
evalable6 False
tellable6 gist.github.com/6d3386ef88a4f99e3b...4df6d63432
13:46 evalable6 left, bisectable6 left
AlexDaniel grondilu: oh oops, that gist has a message for you 13:47
and tellable6 is acting weird :)
13:48 evalable6 joined 13:49 bisectable6 joined
AlexDaniel e: gist.github.com/AlexDaniel/c5c1aa0...099d0f9b19 13:50
evalable6 41
13:50 committable6 joined, ChanServ sets mode: +v committable6
grondilu m: say "1 2 3" ~~ m/[<digit>\s] ** 3/ 13:53
evalable6 False
grondilu m: say "1 2 3" ~~ m/[<digit><ws>] ** 3/
evalable6 「1 2 3」
digit => 「1」
ws => 「 」
digit => 「2」
ws => 「 」
digit => 「3」
ws => 「」
grondilu m: say "1 2 3" ~~ m/[<digit><.ws>] ** 3/
evalable6 「1 2 3」
digit => 「1」
digit => 「2」
digit => 「3」
grondilu I really thought this was simpler 13:54
m: grammar Foo { rule TOP { <digit>**3 } }; say Foo.parse: "1 2 3" 13:55
evalable6 Nil
grondilu m: grammar Foo { rule TOP { <digit>**3 } }; say Foo.parse: "123"
evalable6 「123」
digit => 「1」
digit => 「2」
digit => 「3」
grondilu m: grammar Foo { rule TOP { <digit> **3 } }; say Foo.parse: "1 2 3"
evalable6 「1 2 3」
digit => 「1」
digit => 「2」
digit => 「3」
grondilu oh my, the space is relevant here?
that seems LTA 13:56
AlexDaniel grondilu: but that's the point of using a rule, no? 13:57
grondilu I thought that on the contrary, spaces were relevant in tokens, not rules.
And I mean the space before the ** operator 13:58
basically I thought <foo>**number was the same as <foo> **number 13:59
apparently not :/
grammars are hard. Everytime I try to write one, I struglle with things like that :/ 14:00
14:02 |oLa| left
moritz grammars are a whole programming topic that you have to learn through practice, much like writing good OO code is 14:03
14:04 molaf joined 14:07 pat_js joined
tadzik oh, it took me ages too :/ 14:08
timotimo tbf our tooling around grammars could use some improvement 14:12
grondilu yeah, there is a debugger but it's in module space. Not sure it should not be in the core or something. 14:14
timotimo not necessary, i think
grondilu maybe not a debugger, but a failure to parse should be a bit more verbose than just returning "Nil". 14:15
timotimo that's just one thing, but yeah 14:16
there's been discussion on making failed parses return a Failure object with extra info
jnthn The problem is it broke back-compat, *but* it can be done for 6.e (and that was being worked on recently by vrurg++) 14:17
grondilu ok, good to know it's being looked at. 14:18
timotimo yep
14:20 robertle joined
tobs AlexDaniel: getting there. Now I'm getting the test error I always ancitipated :) 14:23
14:24 grondilu left 14:32 [particle] left, [particle] joined 14:38 jaldhar_ left 14:39 jaldhar_ joined 14:40 jaldhar_ left 14:41 jaldhar_ joined 14:44 mowcat joined 14:52 pat_js left 14:54 jaldhar_ left 14:55 jaldhar_ joined 14:56 jaldhar_ left 14:57 jaldhar_ joined
Geth doc: d29fe6990b | Antonio++ | 3 files
Delets trailing 's' from comment pod block, fix #2970
14:58 MilkmanDan left 14:59 MilkmanDan joined 15:03 antoniogamiz joined
antoniogamiz o/ 15:03
how can i know what roles implements a role?
15:04 pat_js joined, jaldhar_ left
moritz m: role A { }; role B does A { }; say B.^roles 15:05
evalable6 ((A))
moritz antoniogamiz: ^^
ah, you're asking the other way round
15:05 jaldhar_ joined
moritz no 15:05
in general, you can't, unless you have a list of all possible types
antoniogamiz moritz: oh thanks ! 15:07
I was asking the first one, actually haha
m: Telemetry.new 15:08
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/i3DdXtk5D1
Undeclared name:
Telemetry used at line 1
antoniogamiz how can I declare an instance of that class?
MasterDuke antoniogamiz: it's in Rakudo's lib directory, not part of the core, so you need to `use Telemetry` first 15:10
antoniogamiz ah ok :) 15:11
15:13 ufobat_ joined 15:16 dumarchie left, ufobat left 15:26 jaldhar_ left 15:27 jaldhar_ joined 15:28 pat_js left, jaldhar_ left 15:29 jaldhar_ joined 15:30 robertle left 15:35 Sgeo left 15:42 jaldhar_ left 15:43 jaldhar_ joined 15:44 jaldhar_ left 15:45 jaldhar_ joined 15:49 jaldhar_ left, jaldhar_ joined 15:52 antoniogamiz left 16:00 aborazmeh joined, aborazmeh left, aborazmeh joined 16:01 aborazmeh left 16:04 patrickb left, aborazmeh joined, aborazmeh left, aborazmeh joined 16:09 domidumont left 16:13 aborazmeh left 16:19 reach_satori joined 16:31 Kaiepi left 16:33 pat_js joined 16:41 lgtaube left 16:51 Black_Ribbon joined
Geth whateverable: taboege++ created pull request #357:
Support messages from the discord bridge
AlexDaniel oooh… 17:05
tobs: why not connect with the right nick right away? 17:08
instead of setting up `method irc-connected($m)`
`:nick` parameter is passed a few lines earlier :)
17:09 aborazmeh joined, aborazmeh left, aborazmeh joined
AlexDaniel tobs: another thing, we can't be testing just evalable… 17:13
17:13 robertle joined
AlexDaniel because it can affect other bots in unexpected ways 17:14
for example, tellable should also work
.seen and .tell and stuff
tellable6 AlexDaniel, I'll pass your message to and
tobs AlexDaniel: good catch, that simplifies a bunch of stuff. I think I wasn't sure about when exactly to set nick and when to get the faked one and from where when I wrote that.
AlexDaniel what the hell was that, tellable6 ?
17:14 pecastro left
AlexDaniel .seen MasterDuke bot test 17:15
17:15 patrickb joined
tellable6 AlexDaniel, I'll pass your message to MasterDuke 17:15
AlexDaniel oh
tadzik :D
AlexDaniel well it was kinda meant to work that way…
tobs AlexDaniel: but doing it right in :nick works fine.
AlexDaniel: we could duplicate the entire test suite 17:18
17:20 ravenous_ joined 17:21 pecastro joined 17:33 pilne joined 17:35 aborazmeh left 17:37 aborazmeh joined, aborazmeh left, aborazmeh joined 17:38 aborazmeh left
AlexDaniel tobs: hmm 17:38
tobs: well, every test goes through this: github.com/perl6/whateverable/blob...le.pm6#L92
so in theory it should be doable to just run everything twice, once through the bridge and once directly
buuuut it won't work for all bots
definitely not for all tests. e.g. tellable will then start passing messages two times, and stuff like this 17:39
tobs right, maybe the role is not the ideal solution.
AlexDaniel the role is fine it's just the way we test things…
sooo what about adding a test here: github.com/perl6/whateverable/blob...e.pm6#L149 17:40
if the bot can reply to the help command through the bridge, then it's not too broken
17:41 sena_kun left
tobs wouldn't that test less than what is tested now? 17:42
AlexDaniel tobs: but I mean in addition to that
17:44 aborazmeh joined, aborazmeh left, aborazmeh joined 17:46 khisanth_ left 17:49 holyghost joined 17:54 aborazmeh left
tobs AlexDaniel: the $!discord flag still needs to be set in the Testable instance so that it uses the right nick. So should there be an environment variable that every test inspects to see if :discord should be passed to Testable.new? 17:55
17:56 pat_js left 17:57 dumarchie joined
AlexDaniel tobs: I think it will be easier if you connect another irc-client for the bridge 17:58
tobs: then the `test` routine can have a `:bridge` flag
17:58 khisanth_ joined
AlexDaniel tobs: if True, it can then send a message through the bridge client with `<$.our-nick> $command` 17:59
and the rest is processed as normal 18:00
18:00 aborazmeh joined, aborazmeh left
AlexDaniel tobs: so self.test(:bridge, ‘help message’, “$.bot-nick, helP”, “$.our-nick, $help # blah blah”) 18:01
tobs AlexDaniel: and remove the :discord flag for the entire instance? 18:02
AlexDaniel tobs: yeah, don't need it
18:02 jaldhar_ left
tobs Ok, let me see. 18:02
AlexDaniel tobs: and then, after that, we can perhaps add `:both` named parameter, which will run tests with both :bridge and without 18:03
18:03 jaldhar_ joined
AlexDaniel so after sprinkling some “:both” on the whole test suite we can get some nice coverage, maybe? 18:03
18:04 aborazmeh joined, aborazmeh left, aborazmeh joined, aborazmeh left 18:05 jaldhar_ left, jaldhar_ joined 18:06 aborazmeh joined, aborazmeh left, aborazmeh joined, jaldhar_ left 18:07 jaldhar_ joined 18:18 mowcat left 18:22 sauvin left 18:28 Kaiepi joined 18:29 chloekek joined 18:34 chloekek left, jaldhar_ left 18:35 jaldhar_ joined 18:36 jaldhar_ left 18:37 jaldhar_ joined 18:41 aborazmeh left, wildtrees joined 18:47 aborazmeh joined, aborazmeh left, aborazmeh joined, aborazmeh left 18:48 veesh left 18:50 aborazmeh joined, aborazmeh left, aborazmeh joined 18:54 [Sno] joined 19:02 jaldhar_ left 19:03 jaldhar_ joined 19:06 molaf left 19:08 aborazmeh left 19:12 aborazmeh joined, aborazmeh left, aborazmeh joined, aborazmeh left 19:13 _jrjsmrtn left, kensanata joined, __jrjsmrtn__ joined 19:14 aborazmeh joined, aborazmeh left, aborazmeh joined 19:15 aborazmeh left 19:17 aborazmeh joined, aborazmeh left, aborazmeh joined 19:19 aborazmeh left 19:22 aborazmeh joined, aborazmeh left, aborazmeh joined, aborazmeh left
tobs e: sub f { once { say "hi" } }; f; f; f; f 19:23
evalable6 hi
tobs e: sub f { if True { once { say "hi" } } }; f; f; f; f
evalable6 hi
tobs this is unfortunate ^
19:24 mahafyi left
tobs can someone explain to me how once works? 19:24
19:29 aborazmeh joined, aborazmeh left, aborazmeh joined
AlexDaniel tobs: docs.perl6.org/language/traps#Usin...once_block 19:30
tobs ah, it's the same with a state variable inside an if block. 19:31
AlexDaniel yup 19:32
tobs now _that_ is a trap for an occasional perl5 programmer like me 19:33
19:34 aborazmeh left 19:35 Ven`` joined
tobs AlexDaniel: I wonder if :both should be the default and opt-out for tests known to fail? 19:36
AlexDaniel tobs: we can try that!
evalable.t will be fine, committable.t will be fine too 19:37
bisectable should be alright
unicodable too
notable and tellable definitely not
19:37 Ven`` is now known as Ven_de_Thiel
AlexDaniel actually, maybe all files except notable and tellable 19:38
and bisectable.t will become reeeeeeaally slow :)
but I don't mind
19:40 cpan-p6 left
AlexDaniel tobs: also, see this: github.com/perl6/whateverable/issues/340 19:40
19:40 cpan-p6 joined, cpan-p6 left, cpan-p6 joined
AlexDaniel tobs: you can simply ignore anything that is known to fail, I'll take a look later 19:41
tobs hah, running the tests is pretty annoying now :) 19:44
AlexDaniel I'm not sure if there's any other way 19:45
as we just had a `what:` issue uncovered by a bridge test
so it's useful for sure, just takes time :) 19:46
19:47 chloekek joined
tobs AlexDaniel: regarding the list of non-functional tests: benchable.t also doesn't work for me 19:47
instead of the expected data, it returns "No new data" for me. Maybe my repo is not fully initialized?
AlexDaniel no… I think benchable changed and the tests were not updated 19:48
but hey, at least we have all these tests, they just need some love
19:49 aborazmeh joined, aborazmeh left, aborazmeh joined
AlexDaniel tobs: I updated the list: github.com/perl6/whateverable/issues/340 19:50
tobs I'm hitting timeouts on the segfault test in evalable now... 19:51
AlexDaniel tobs: possibly means that the bot doesn't respond? 19:52
tobs AlexDaniel: no, doubling the timeout makes it pass. The whole thing seems to slow down a bit with time. 19:54
AlexDaniel hhhhhhhhuuuuuuh 19:55
tobs: what's your rakudo version?
tobs: can you try monitoring memory usage while it runs?
tobs the same but bridged test afterwards passes really quickly though and I just nuked /tmp/whateverable, so maybe it has to do some extra work
AlexDaniel it can also be a bug in rakudo
tobs AlexDaniel: This is Rakudo version 2019.07 built on MoarVM version 2019.07 19:56
AlexDaniel this rings some bells because bots used to slow down quite a bit when memory leaks were present
it shouldn't be the case anymore, but please check… it can be a bug in rakudo…
tobs AlexDaniel: the memory consumption grows steadily, but not obscenely, from 190M to 290M over the entire test 19:59
AlexDaniel OK that's normal
tobs with my 2019.07 rakudo.
AlexDaniel it doesn't exactly grow, it just tends to settle on an amount that's a bit too high. It's ok
20:00 Xliff joined
AlexDaniel tobs: could it be the IRC server doing some throttling? 20:00
Xliff \o
tobs good idea.
AlexDaniel o/
Xliff: ↓
e: gist.github.com/AlexDaniel/c5c1aa0...099d0f9b19
evalable6 41
AlexDaniel thanks for the PR! Had to fix it a little bit and write some tests, but the approach was exactly right :) 20:01
Xliff ? 20:02
Oh! LOL! 20:03
That was from back in Feb! Glad you like it :)
20:18 lgtaube joined
tobs AlexDaniel: another failing test here github.com/perl6/whateverable/issu...-525015600 20:20
20:23 dumarchie left 20:24 molaf joined
AlexDaniel commit: releases say $*PERL 20:26
committable6 AlexDaniel, gist.github.com/6ced877769f319279e...993fa77bcf 20:27
AlexDaniel tobs: meh and a bunch of tests after that are probably broken too
20:27 holyghost left 20:28 holyghost joined
AlexDaniel tobs: just change the query to `say 42` and that's it 20:28
well, and the output accordingly
these tests were working before we release v6.d :D 20:29
20:31 aborazmeh left, lizmat joined
lizmat and yet another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2019/08/26/...-atlantic/ 20:31
tellable6 2019-08-23T19:51:51Z #perl6-dev <AlexDaniel> lizmat: that GB18030 roast merge was a bit too early, moarvm pull request is not merged yet
2019-08-25T11:43:30Z #perl6-dev <dogbert17> lizmat did you by any chance fix R#3069 ?
synopsebot R#3069 [open]: github.com/rakudo/rakudo/issues/3069 Date formatter invalid after the twenty-eight day
tellable6 gist.github.com/d4c1ca169f83e5a04a...8cefbb7ff9
20:32 lizmat left, ravenous_ left
AlexDaniel lizmat: re “FWIW, only these people are allowed to vote, after which Jonathan Worthington will decide.”, it's other way around 20:37
tellable6 AlexDaniel, I'll pass your message to lizmat
AlexDaniel lizmat: so jnthn will be giving feedback so that we get a PR that is aligned with his view, then everyone votes (technically jnthn too) 20:38
tellable6 AlexDaniel, I'll pass your message to lizmat
20:46 MasterDuke left 20:49 scotticles joined 20:50 scotticles left 20:53 irced joined
irced hello, i am having difficulty using subparse. is there a site to which i can paste my code and discuss the issues is? 20:54
AlexDaniel irced: what about gist.github.com/ 20:58
irced that'll work, i thought maybe there was one where it can compile and run with output 20:59
Grinnz f.perl.bot/#perl6 can, not sure if simcop2387 is keeping up with the rakudo updates 21:00
AlexDaniel glot.io/ and a bunch of other websites…
evalable6 can run a gist just fine
21:00 lizmat joined, kensanata left
irced hmm, ok, f.perl.bot seems ok. what's this about evalable6 ? 21:01
AlexDaniel irced: evalable6 can run gists, so you upload your code on gist.github.com/ and we'll give a link to the bot 21:02
irced that sounds ok too. f.perl.bot is more readable than an irc bot, so let me see if i can get help with that first 21:04
f.perl.bot/raw/evq89c why won't this grammar match on the subparse ? 21:05
AlexDaniel e: f.perl.bot/raw/evq89c 21:06
evalable6 AlexDaniel, Successfully fetched the code from the provided URL
namespace Details {
some namespace details
some more namespace details

#<failed match>
irced kewl
a quick inspection will show ratcheting is not the issue 21:08
not the central one anyway
tobs commit: v6c say $*PERL 21:15
irced brb
committable6 tobs, gist.github.com/1a8a239299752c4a00...15a1f5516f
tobs AlexDaniel: should this include v6.d also? 21:16
21:18 Ven_de_Thiel left
irced burps. 21:19
AlexDaniel irced: sry I'm in the middle of a bunch of tasks at the same time, can't look right now. Maybe someone else will :) 21:20
tobs: yyyyyeeea… github.com/perl6/whateverable/issues/345
irced no problem
tobs AlexDaniel: so much low-hanging(?) fruit :) 21:21
AlexDaniel tobs: most whateverable tickets are relatively easy. Discord thing is probably one of the hardest ones 21:22
tobs I only have the very last test of unicodable failing (and bisectable not run yet) before I can post the next version of the PR 21:23
irced i think the central issue with my code was that the rule was the only thing subparsed and it may not have been complete enough 21:27
21:41 patrickb left 21:48 lizmat left 21:49 Cabanossi left 21:51 Cabanossi joined
irced yeah it seemed to be a lack of understanding of the mechanics of subparse so disregard the question as stated thanks all 21:55
21:56 irced left
guifa jnthn: is there a place for bug reports for Comma? 21:57
It doesn’t like my perfectly acceptable indirect colon syntax lol 21:58
22:01 chloekek left
guifa irced: Take a look at the name token. Two things: 22:03
tellable6 guifa, I'll pass your message to irced
22:07 pecastro left 22:11 pecastro joined
guifa (1) When you do the subparse, it assumes an initial ^ 22:13
That means if a token/rule begins with an <?after>, it can never match on subparse.
(2) I can appreciate you wanting to capture just the name. Rather than use the <?after> just use 22:14
rule name-declarator { ‘namespace’ <name> ‘{‘ } 22:15
token name { \S+ }
simcop2387 Grinnz: i am sadly not, but with my current efforts to begin re-doing all that infrastructure I'll probably try to make that easier to manage. 22:29
22:39 rindolf left
jnthn guifa: Yes, make an account (in the buy process); you don't have to actually buy a copy to be able to reach the feedback page, just be logged in :) 22:41
Though I think there's actually a ticket for indirect colon syntax, which is probably waiting for somebody to actually want it :P 22:42
22:43 Sgeo joined
guifa raises hand 22:53
I’m probably the only person who use push array COLON item but it lines up so much nicer when I do that with my own objects in the same block 22:54
jnthn :) 22:55
Found it in our internal bug tracker, shuffled it into the To Do Soon column. Apparently it's assigned to me. :)
AlexDaniel tobs: oh, I know
tobs: unicodable does some tricks in order to get the full message
tobs: otherwise IRC::Client strips unicodey stuff like whitespace and whatever
discord6: help 22:56
well… what I like about whateverable bots is that at least it's easy to get help… 22:57
tobs: anyway, if discord6 is using IRC::Client, which I assume it does, then it's probably hard to get unicode through the bridge anyway
tobs AlexDaniel: and yes, bisectable.t takes a while and makes for a nice heating 23:00
AlexDaniel hah 23:01
tobs it was the only test that ran through without problems the first time though
AlexDaniel u: raw.githubusercontent.com/perl6/wh...stable.pm6 23:05
unicodable6_ AlexDaniel, Successfully fetched the code from the provided URL
AlexDaniel, U+0075 LATIN SMALL LETTER U [Ll] (u)
AlexDaniel, U+0073 LATIN SMALL LETTER S [Ll] (s)
AlexDaniel, Cowardly refusing to gist more than 5000 lines
AlexDaniel tobs: that test is bullshit 23:06
tobs: $t.last-test; is missing from that file
tobs: it is needed so that can make sure that there are no messages in the queue from the last test 23:07
tobs: because of that, the test file was succeeding
so add $t.last-test; before done-testing, aaand…
and we can change the test to something like 23:08
u: gist.githubusercontent.com/AlexDan...b9/foo.txt
unicodable6_ AlexDaniel, Successfully fetched the code from the provided URL
AlexDaniel, U+221E INFINITY [Sm] (∞)
AlexDaniel I wonder if this works too
u: gist.github.com/AlexDaniel/b392c71...6b4aeab511
unicodable6_ AlexDaniel, U+221E INFINITY [Sm] (∞)
AlexDaniel lol, cool, we can test this too :) 23:09
tobs heh, and my guess about `u:` turned out to be completely wrong, it was `use` :D 23:13
AlexDaniel xD 23:17
23:17 Kaiepi left
tobs AlexDaniel: I fixed the expectations of that specific test instead of replacing it by the ∞-gist. 23:23
probably the file being .pm6 was important, looking at the test description?
AlexDaniel ah, last-test was misplaced, doh 23:24
tobs: not really, the idea is just that it accepts something reasonable
unicodable should probably accept *any* mime type, but it shares the code path with other bots so maybe it doesn't 23:25
ok we're ready to merge I think
tobs \o/
thanks for the mentoring 23:26
AlexDaniel that private_method_table trick is pretty cool
Geth whateverable/master: 6 commits pushed by (Tobias Boege)++, (Aleks-Daniel Jakimenko-Aleksejev)++ 23:27
AlexDaniel I'm happy to help :)
23:28 evalable6 left, bisectable6 left
timotimo cool 23:28
23:28 committable6 left
AlexDaniel is anybody on discord? 23:28
discord6 <timotimo> m: say "you are"
AlexDaniel OK good, too fast though!
timotimo :)
AlexDaniel also, that is not going to work!! 23:29