»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.
00:00 reportable6 left 00:01 netrino left 00:03 reportable6 joined 00:09 lembark left 00:10 molaf left 00:55 Actualeyes left 00:56 leont left 00:59 Actualeyes joined 01:00 Actualeyes left 01:06 MasterDuke left 01:10 john_parr joined
discord6 <Aearnus> here's a thought w.r.t. the naming discussion that's been going around 01:12
<Aearnus> Raku for the language, Camelia for the subset that we used to refer to as "baby perl 6"
<Aearnus> c.f. github.com/perl6/problem-solving/i...-520650352
01:13 a3r0 joined 01:25 |oLa| left 01:37 uid9164 joined 01:52 uid9164 left 02:04 jaldhar_ left, jaldhar_ joined, Manifest0 left 02:05 Manifest0 joined 02:08 jaldhar_ left 02:09 jaldhar_ joined 02:16 jaldhar_ left 02:17 jaldhar_ joined 02:18 adu joined, jaldhar_ left 02:19 jaldhar_ joined 02:20 jaldhar_ left 02:21 jaldhar_ joined 02:22 jaldhar_ left, kerframil left 02:23 jaldhar_ joined, squashable6 left, Cabanossi left 02:24 squashable6 joined, ChanServ sets mode: +v squashable6 02:28 Cabanossi joined 02:33 lucasb left 02:36 jaldhar_ left 02:37 jaldhar_ joined 02:38 jaldhar_ left 02:39 jaldhar_ joined 02:40 jaldhar_ left 02:41 jaldhar_ joined 02:46 jaldhar_ left 02:47 jaldhar_ joined 02:48 jaldhar_ left 02:49 jaldhar_ joined 02:50 jaldhar_ left 02:51 jaldhar_ joined 02:54 jaldhar_ left 02:55 jaldhar_ joined 03:05 majensen joined 03:13 majensen left 03:38 jaldhar_ left 03:39 jaldhar_ joined 03:40 jaldhar_ left 03:41 jaldhar_ joined 03:42 jaldhar_ left 03:43 jaldhar_ joined 03:44 jaldhar_ left 03:45 jaldhar_ joined 03:46 jaldhar_ left 03:47 jaldhar_ joined 03:52 jaldhar_ left 03:53 jaldhar_ joined 03:54 jaldhar_ left 03:55 jaldhar_ joined, khisanth_ left 04:09 Sgeo_ joined, khisanth_ joined 04:13 Sgeo__ left 04:28 __jrjsmrtn__ left 04:37 avalenn joined 04:43 Sgeo__ joined 04:44 curan joined 04:45 veesh left 04:46 Sgeo_ left 04:54 xinming left 05:11 veesh joined 05:28 ponbiki joined 05:40 xinming joined 05:59 snarkyboojum_ joined 06:00 reportable6 left 06:03 snarkyboojum__ left 06:04 reportable6 joined 06:07 [particle] left 06:15 sauvin joined 06:17 rindolf joined 06:19 [particle] joined 06:28 veesh left, ravenousmoose joined 06:39 [particle]1 joined 06:40 [particle] left 06:43 veesh joined, [particle]1 left 06:44 [particle] joined 06:51 [particle] left 06:53 molaf joined 07:00 Geth left 07:01 [particle] joined 07:05 [particle]1 joined 07:07 [particle] left, __jrjsmrtn__ joined 07:12 molaf left 07:14 |Sno| left 07:19 [particle]1 left 07:20 kensanata joined 07:21 abraxxa joined, molaf joined 07:22 [particle] joined 07:25 chloekek joined 07:29 El_Che_ joined, El_Che_ left 07:30 zakharyas joined 07:32 ravenousmoose left, ravenousmoose joined 07:33 molaf left 07:35 dolmen joined 07:39 [particle] left 07:41 robertle left 07:46 [particle] joined 07:51 epony left 07:57 [particle] left 07:59 dolmen left 08:01 leont joined 08:02 dakkar joined 08:03 Manifest0 left 08:04 Manifest0 joined 08:05 lgtaube left 08:07 [particle] joined 08:09 leont left 08:12 dolmen joined
AlexDaniel Aearnus: just no 08:16
08:17 dolmen left
AlexDaniel I want the naming outcome to be easy to understand 08:17
08:17 dolmen joined
AlexDaniel so you'd have camelia, raku, rakudo, star… and which one is which again? 08:18
raku the language, camelia the distribution, rakudo the compiler, no star?
I mean… come on :)
08:21 [particle]1 joined, dolmen left 08:22 lgtaube joined
lizmat is awake now, but is first going to have breakfast & 08:23
08:23 [particle] left
discord6 <Aearnus> AlexDaniel: good point 08:24
08:29 dolmen joined 08:32 molaf joined 08:38 ravenousmoose left, ravenousmoose joined 08:41 [particle] joined 08:42 [particle]1 left
El_Che I find the nqp, moarvm, rakudo, perl6 distinction pretty confusing 08:49
08:49 sena_kun joined
El_Che 1 compiler reached maturity, let's be happy with that and don't act like people are queuing up to write new ones 08:49
let's call everything rakudo and move on ;) 08:50
it made sense to dream big when Perl was still popular 08:51
now? Less so.
AlexDaniel El_Che: well… not really 08:55
08:55 noisegul joined
AlexDaniel El_Che: people are still implementing perl6-like languages 08:56
El_Che perl6-like
that just a fork like people do in python and go 08:57
08:59 mowcat joined, mowcat left 09:00 mowcat joined, mowcat left, mowcat joined 09:10 satori__ left 09:13 smash joined
discord6 <Aearnus> El_Che: that's why I like the idea of "rakudo" implementing "raku" 09:18
<Aearnus> and even though it's not english proper, it's as if the compiler for raku would be called raku-do
09:20 molaf left
kensanata The way of the raku? 09:21
raku-jutsu!
El_Che discord6: yes, I like the vim example Damien gave 09:22
Aearnus: yes, I like the vim example Damien gave 09:23
Aearnus: I would prefer rakudo, but raku is nice as well
:rakudo
that would rock
09:23 epony joined
El_Che in vim 09:23
rakudo > raku > camelia for me 09:24
but even if it's camelia, it would be OK for me
09:24 dolmen left
El_Che unf*ck the situation once and for all 09:24
09:28 dolmen joined
discord6 <Aearnus> definitely. most anything is better than the current situation 09:28
AlexDaniel` That everyone seems to agree with 09:33
chloekek I like Perl 6. 👌 09:35
discord6 <Aearnus> why do you say? 09:36
El_Che is stmuk leaving perl 6? (the ticket about planet) 09:44
09:46 wamiks joined 09:47 satori__ joined 09:49 Black_Ribbon left 09:51 pilne left
lizmat El_Che: I think stmuk and virtualsue hold me responsible for zoffix leaving the community 09:52
github.com/perl6/problem-solving/i...-520773602 10:06
tadzik rakudo implementing rakudo is a pun, willingly or not, and the idea grows on me
discord6 <Aearnus> lizmat: 👍 10:07
tadzik (while standing by everything I wrote in the ticket)
but lizmat++, I like how this is being handled
AlexDaniel lizmat: what's the mistake we're talking about? 10:22
10:22 nwc10 joined
nwc10 Good *, #perl6 10:22
It landed: www.nntp.perl.org/group/perl.perl6...18326.html 10:23
lizmat Many, many years ago, Damian suggested to Larry to rename "Perl 6" to "Rakudo"
instead it was used to name the compiler
AlexDaniel: ^^
AlexDaniel nwc10: iirc one of the biggest issues is that we still have to return a match object 10:26
lizmat: seems like some people strongly dislike raku and some strongly dislike camelia. Meh, you can't win… 10:27
tadzik to be fair, some people also strongly dislike perl 6 ;) 10:29
AlexDaniel no that's not the question anymore 10:30
lizmat fwiw, 1nickt's response exactly describes the vibe I've been getting at events like the Perl Toolchain Summit
tadzik nwc10: that's one insightful writeup, thanks for this :) 10:31
yeah, I know that's not the question
lizmat or at perlmonks, for that matter 10:32
tadzik I meant something more like: you'll never please anybody, and pleasing everybody shouldn't be the primary concern
nwc10 AlexDaniel: is the "I failed to match" object effectively a singleton?
AlexDaniel nwc10: no
nwc10 ah OK.
lizmat tadzik: trying to please everybody, is what got us here
tadzik kind of, yes
nwc10 anyway, one can still generate the match objects even if one changes how the "matching" works
but that's beyond what I know. What I knew about, I tried to write there 10:33
discord6 <Aearnus> i always thought that the bright & colorful branding of camelia was kind of odd for a programming language. not bad necessarily, just odd.
10:33 nwc10 left
lizmat Aearnus: I think it shows diversity, in colours, in age, in mindsets 10:34
and with the eyes a little cross-eyed, not taking yourself too seriously 10:35
AlexDaniel how is raku pronounced, btw? 10:36
I've heard somebody say it during the conference and I was thinking like “waait, that's not how you say that”, and now I wonder if I'm wrong 10:37
sena_kun AlexDaniel, as in Japanese, I suspect
so "it reads as it is written", being "r" as in "risk", "a" as in "archer", "k" as "c" in "cat", "u" as in... eeeh... 10:38
discord6 <Aearnus> lizmat: all things that are really important for the language's branding, I agree. I just can't help but let my mind wonder to what a modernistic or minimalist sort of branding might look like 10:39
sena_kun translate.google.com/#view=home&am...;text=raku <- "u" as it pronounces it. :)
AlexDaniel and the stress is on what?
tadzik ra-koo I think? : 10:40
sena_kun AlexDaniel, no stress!
tadzik, ah, yes, English "oo" is suitable, as in "boom".
tadzik english is at fault for pronouncing "u" "you" instead of just "u" like we perfectly correctly do in Polish :>
AlexDaniel what do you mean no stress? 10:41
sena_kun Japanese has no stressing (as we understand it, at least), so...
tadzik ah: RAku not raKUU?
sena_kun you just pronounce both syllabiles without making a stress anywhere
tadzik nodnod 10:42
AlexDaniel :S
10:42 dolmen left 10:43 kensanata left
sena_kun it(absence of stressing) is not completely true for the language as there are various interesting cases, but it doesn't apply to case with "raku" and "rakudo". no long vowels or anything like that, just two-three syllabiles 10:45
10:46 TimToady joined
sena_kun >For example, 70% of native nouns are unaccented, while only 50% of kango and only 7% of loanwords are unaccented 10:46
of course, there are always those 30%. ;)
10:49 scimon joined 10:51 veesh1 joined
AlexDaniel actually, I think I'm wrong about failed matches 10:52
kawaii sena_kun: rakUwUdo
AlexDaniel in case of a simple regex, we just return Nil and set $/ to Nil, right?
so that's not extremely difficult, probably?
tadzik and the optimizer probably knows if $/ is used in its scope, in case it's not Nil 10:54
discord6 <Aearnus> kawaii: the discussion about the name makes me feel like rakudOwO
kawaii Both are acceptable corruptions of the name imo :)
10:54 satori__ left
tadzik m: do { 'foo' ~~ /foo/; do { 'bar ~~ /bar/ }; say ~$/ } 10:54
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in single quotes; couldn't find final "'" (corresponding starter was at line 1)
at <tmp>:1
------> 3~ /foo/; do { 'bar ~~ /bar/ }; say ~$/ }7⏏5<EOL>
expecting …
tadzik m: do { 'foo' ~~ /foo/; do { 'bar' ~~ /bar/ }; say ~$/ } 10:55
camelia bar
tadzik oof
jnthn The thing about $/ is that it has dynamic scope 10:56
So "is it used" is not easy :)
(This is why we changed $_ - because it being dynamic made *everything* hard to analyze. :))
11:02 Guest93 left 11:03 Guest93 joined 11:07 satori__ joined
tadzik right, I was hoping it's lexical 11:11
11:14 cygx joined
AlexDaniel jnthn: why should it have dynamic scope though? 11:16
I understand it'll break existing code, but that doesn't necessarily mean we have to keep it that way forever 11:17
jnthn AlexDaniel: Well, in the immediate so things like .match and .subst can set it :) 11:18
11:19 Geth joined
jnthn And I don't see how we get away from that 11:19
lunch, bbiab
11:24 molaf joined 11:27 leont joined 11:38 dolmen joined 11:42 ChoHag left 11:45 kensanata joined 11:47 zakharyas left
daxim redd.it/9tws89#e92djoj 11:48
unicodable6: who is your daddy and what does he do? 11:49
unicodable6 daxim, U+0077 LATIN SMALL LETTER W [Ll] (w)
daxim, U+0068 LATIN SMALL LETTER H [Ll] (h)
daxim, 38 characters in total (who is your daddy and what does he do?): gist.github.com/f703b63764a6df07a4...2685a9e157
daxim unicodable6: ������������� 11:51
unicodable6 daxim, U+FFFD REPLACEMENT CHARACTER [So] (�)
daxim, 13 characters in total (�������������): gist.github.com/bc90e4d7ef4e9d5a44...9c70f3f915
AlexDaniel unidump: z�｢ 11:59
12:00 reportable6 left
unicodable6 AlexDaniel, gist.github.com/7fea34135530432795...05119236da 12:00
AlexDaniel don't know what the daddy is doing, but unicodable6 can do some cool stuff like this ↑
12:04 reportable6 joined, ChanServ sets mode: +v reportable6
AlexDaniel so hold on a second, grammar.parse “Returns a Match object on success, and Nil on failure.”, and we're changing that in 6.e, right? 12:05
El_Che lizmat: oh I see, the comment could be read like that. I wondered what it meant, it was kind of out of context/place 12:06
AlexDaniel so now you'll be getting a Failure if it doesn't match
but we don't do the same for regexes… should we?
woolfy Op 2 augustus heeft Jolanda keurig netjes die 120 euro overgemaakt op jouw bankrekening. 12:08
oh shit wrong window
sorry... eek... that is so personal... :-(
El_Che I read it as rahkoo
with the emphasis on the oo
daxim use IPA 12:09
the approximations just trick you into thinking you have communicated your thought successfully, but >80% of the time it doesn't
12:12 leont left
AlexDaniel lizmat: IMO it's a bit more important now to establish a PR with the things asked for by jnthn, we may be jumping a bit too fast to choose a particular name 12:13
tbrowder i just watched jnthn's talk on p6 performance--very informative, as usual! 12:14
El_Che AlexDaniel: I feel the balance is moving towards raku
tbrowder: yeah, it was a nice talk!
AlexDaniel like it's first camelia, then raku, then who knows we might jump back to camelia… let's just focus on things that need doing
El_Che rakudo :)
lizmat I won't be jumping back to Camelia
El_Che rakudodo
lizmat or rakudo
I want to move ahead, as many other people do 12:15
El_Che: rakudodo == raku dodo
El_Che indeed
lizmat not a positive thing I would say ?
AlexDaniel haha
tbrowder question: is anyone working on precompilation for the script as well as its modules? i
El_Che I have tons of bad jokes
daxim just to confirm my understanding: with issue 81 that would be raku the language and rakudo the implementation?
El_Che call the language "straw"
The straw that broke the camel's back 12:16
lizmat daxim: that would be the idea, yes
AlexDaniel daxim: that's right, yeah
El_Che (you're not laughing now :) )
tbrowder i think madeline's gsoc work should also pay off in parts of the head-to-head speed contesr with other languages, no? 12:17
lizmat
.oO( the last straw to rule them all :-)
 12:18
12:18 AlexDani` joined
El_Che lizmat: I see you acknowledge the potential of the name 12:19
12:20 AlexDani` left, AlexDani` joined, AlexDaniel left, AlexDani` is now known as AlexDaniel
cygx tbrowder: being able to precompile scripts will help with startup time 12:27
That's an immediate benefit
There's a bit more to do to make module linking a reality 12:28
I'm currently looking into getting the missing pieces into place
El_Che nice 12:29
cygx I should probably talk to her about that :) 12:30
woolfy At PerlCon in Riga, I sold 31 Perl 6 books and 2 Perl 5 books. 12:34
tadzik nice! :) 12:35
timotimo it's already a nice win to have just the "perl6" command start up faster
daxim one of the green books was by recommendation from me 12:36
woolfy Including a set of 3 books to a small company of which the boss said that using Perl 6 was out of the question and after seeing several Perl 6 talks came and said "we should have started a lot sooner".
daxim++
12:36 veesh1 left 12:37 lucasb joined
timotimo cool 12:37
daxim I (accidently) had breakfast with that boss, he was thoroughly impressed
woolfy daxim : heh, I hoped that bit would remain anonymous... :-)
AlexDaniel I'm not too surprised, a lot of great perl6 talks 12:38
woolfy: o/
woolfy AlexDaniel : it was nice finally meeting you! 12:39
AlexDaniel woolfy: :) 12:40
12:41 pmurias joined
pmurias SmokeMachine: hi 12:41
El_Che Hi woolfy
jnthn It was nice to meet up with lots of folks at PerlCon, though I'm sorry to any of you who met me and now have the cold I brought with me there. (All of my family here now have it. D'oh.)
AlexDaniel jnthn: it's fine, I feel better today! :D
daxim 🤧 12:42
woolfy Hi all\
jnthn Wish I did! :) 12:44
El_Che I am impressed by the civility of the discussion on github so far 12:45
AlexDaniel woolfy: but yes, it was truly great to meet all of you :)
12:46 pat_js joined
AlexDaniel El_Che: well, is there a chance for it to be any other way? 12:47
things are different now
daxim only a handful of outsiders in the thread, or rather fencesitters
El_Che AlexDaniel: most discussions need asbestos clothings 12:48
daxim imagine jjl contributing an opinion
12:49 mowcat left
AlexDaniel and we've finally hit the “Load more...” feature on github :) 12:51
12:52 zakharyas joined
AlexDaniel El_Che: but it's interesting that there are no downvotes 12:52
some on the “Camelia” proposal, but that's fine
also some confused emojis, it's ok to be confused :) 12:53
cygx Petsonally, I'd vote for the name 'Prince' - has a nice symmetry to 'The language formerly known as Perl6' :p 12:55
*Personally 12:56
12:56 wamiks left
El_Che cygx: I thought about the OD on utf8 12:56
13:02 lizmat left
pmurias SmokeMachine: I'm getting around to adding extra methods on top of the wrapped js objects in rakudo.js 13:09
SmokeMachine: which ones do you think we should add?
one questions is what should .gist return in rakudo.js 13:10
SmokeMachine pmurias: some... but now I’m on a train going back home... would you mind if I answer you in a few hours? 13:12
13:14 Possum left 13:17 Doc_Holliwood joined 13:20 Cabanossi left
Doc_Holliwood Can I express this any shorter? method roll { [+] ( 0 .. self.dice ).map({ (1 .. self.sides).pick }) } 13:20
timotimo it looks like you're using the 0 .. self.dice .map just to have a repetition 13:22
that is easier to do with xx self.dice at the end
Doc_Holliwood confused
13:23 mensvaga left
timotimo oh, you're right, i misread 13:23
13:23 sdu joined
Doc_Holliwood it's a simple die class. dice is the number if dice, sides the number of sides. so when dice is 2 and sides is 6, it should roll two d6 and add up 13:23
timotimo what you want is actually (1 .. self.sides).roll(self.dice) 13:24
.sum
evalable6 Use of uninitialized value of type Any in numeric context
in block <unit> at /tmp/vO4H3D3D18 line 1
AlexDaniel huh
13:24 Cabanossi joined
AlexDaniel .say 13:24
evalable6 (Any)
timotimo (1 .. 9).roll(4).sum.say 13:25
evalable6 12
Doc_Holliwood oh, roll is a builtin?
sena_kun m: my $dice-num = 5; my $sides = 6; say [+] ((1..$sides).roll xx $dice-num)
camelia 13
timotimo it is 13:26
jnthn $foo.roll xx $n is just $foo.roll($n), I thought?
timotimo roll takes the number of times to roll, so no need for xx either
pmurias SmokeMachine: there's no hurry, I'll be grateful for suggestions as estimating what other people want is hard
timotimo it also allows * or Inf and gives a lazy list of roll results
tyil on my Alpine perl6 docker image `zef install File::Which` fails with "Fetching [FAIL]: git://github.com/azawawi/perl6-file-which.git from git://github.com/azawawi/perl6-file-which.git" 13:27
trying to `git clone` that specific URL works just fine 13:28
wat do
13:29 zakharyas left 13:30 zakharyas joined, dakkar left, dakkar joined
tbrowder in jnthn' performance talk he mentions benefit of using contains and starts-with which i could use a lot more to be sure. 13:31
13:33 sdu left
tbrowder m: my $s = 'boo foo'; say $s.contains('foo'); 13:33
camelia True
13:34 OiannisJVM joined
SmokeMachine pmurias: but the ones I remember now were .Bool and .defined... 13:34
13:34 cpan-p6 left, OiannisJVM left 13:35 cpan-p6 joined, cpan-p6 left, cpan-p6 joined, antoniogamiz joined 13:37 sdu joined 13:41 jaldhar_ left 13:42 jaldhar_ joined, curan left
vrurg tyil: make a ticket for the module developer to change the URL in META6.json. Or create a PR for that. 13:48
tyil it's a valid URL, though
especially since `git clone` on that URL just works without any problem :/ 13:49
vrurg tyil: still, https works always.
tyil: git@ requires authorization and zef cannot provide it. Perhaps somehow git:// does so too. 13:50
antoniogamiz one doubt, if I precompile a file in my local machine, can I use that precompiled code in another machine? 13:51
tyil `Fetching [FAIL]: github.com/azawawi/perl6-file-which.git from github.com/azawawi/perl6-file-which.git`
antoniogamiz without recompile again
tyil my main issue here is that zef doesn't really care to explain *why* its failing, tbh
so I can't really figure out if it's something I'm missing or if the URL is not supported or whatevers
timotimo antoniogamiz: precompiled modules for rakudo usually depend on the exact core setting, which means the exact contents of CORE.setting.moarvm 13:52
antoniogamiz tyil: I can install it using zef install File::Which
13:52 chloekek left
tyil yes, I can install it on my Debian image too 13:52
timotimo we do have stuff in place to ensure reproducible builds, i.e. that if you recompile your rakudo twice in a row with the same sources on the same machine you'll get the same file contents 13:53
tyil hence my problem remains: why is zef failing, and why doesn't it report anything about the failure :I
timotimo but i don't know what the guarantees are, if any, for different machines
vrurg tyil: Heh, https is confusing, yes... And there is definitely something on your side. I agree that more diagnostics would be useful.
antoniogamiz timotimo: mmm so if I use it always in the same environmet of circle ci it will work right?
tyil I already knew it was on my side
:/
antoniogamiz mmm zef nuke maybe fix it?
timotimo there is a chance
tyil antoniogamiz: it's a clean image as far as I can tell 13:54
antoniogamiz mmm then your case is weird :/
tyil I know
if it wasn't weird I wouldn't be going to IRC to ask about it :p 13:55
timotimo yeah, the point is to find out how it's weird
13:55 AlexDaniel left
timotimo personally, i'd probably try to strace -f the thing to see what it ends up doing, among other things the exact parameters to "git clone" it's providing and the output it's getting 13:55
tyil zef -v or --verbose neither seem to do much
timotimo that'll also want a "don't cut off strings early" flag for strace
tyil tbh I consider "no error reporting" a bug in zef 13:56
timotimo agreed
13:59 rainmanj_ joined 14:00 lizmat joined 14:01 rainmanjam left
antoniogamiz tyil: error is thrown here => github.com/ugexe/zef/blob/e7389737...t.pm6#L203 14:03
may you can add some says there to debug 14:04
14:05 satori__ left
antoniogamiz .fetch needs to create some .lock files, check if it has enough permissions 14:05
14:06 sdu left
timotimo the logging might want to go into the git plugin 14:06
14:07 eiro joined 14:08 eiro left 14:11 rainmanj_ left 14:12 cygx left
tbrowder grammar help please: where in the docs is the grammar construct "<?{" and kin discussed? 14:13
14:15 Ven`` joined
vrurg tbrowder: docs.perl6.org/language/regexes#Re...tion_check 14:17
14:18 Sgeo__ left 14:19 Sgeo__ joined
tbrowder vrug: thnx! tried a search for "<?{" before i asked but no hit 14:26
vrurg: ^^
14:28 Itaipu left 14:32 Itaipu joined 14:48 pmurias left 14:50 davido joined 14:54 Sgeo_ joined 14:57 Sgeo__ left 15:01 davido left, davido joined
tbrowder regex/grammar question: in p6 grammar there is a token <opener> defined which is a long alternative list of acceptable chars. in a later use is tested as <?opener>. i would like to temporarily stop the parse there and take an alternative, immediate action if it is false. so far i have had no success with attempts like "[ <?opener> || <.some-other-action> ]". any suggestions for another approach to try? 15:07
the actual file is rakudo/rakudo/src/Perl6/Grammar.nqp 15:08
15:09 eiro joined 15:16 lizmat left 15:17 pamplemousse joined 15:19 epony left 15:22 Ven`` left 15:23 kensanata left
tyil gitlab.com/tyil/docker-perl6/-/jobs nightly Docker images with GitLab, so my desktop being off doesn't ruin my nightly images anymore :D 15:25
antoniogamiz good :)
tyil now to use those for building other images on GitLab, to reach automatic CPAN uploads on GitLab eventually
15:25 [Coke] joined 15:26 [Coke] left 15:27 nepugia joined 15:29 epony joined 15:31 Doc_Holliwood left 15:33 lizmat joined
masak today's autopun: "5 out of 4 people struggle with math" 15:39
antoniogamiz xdd 15:41
15:44 jaldhar_ left 15:47 molaf left 15:56 pat_js left, harmil joined 16:10 mowcat joined, dolmen left 16:22 mensvaga joined 16:25 dolmen joined 16:29 scimon left 16:30 dolmen left 16:35 lizmat left 16:36 dakkar left 16:37 sena_kun left 16:41 rindolf left
tobs tbrowder: I think that's the approach. 16:43
m: my regex opener { 'a' }; sub f ($x) { $x ~~ / [ <?opener> || <?{ note "no opener"; False }> ] / }; for "abc", "zbc" { say "$_: " ~ f($_) }
camelia abc:
no opener
zbc:
no opener
no opener
no opener
Use of Nil in string context
in block at <tmp> line 1
Geth doc: f06c83cee7 | (Tom Browder)++ (committed using GitHub Web editor) | doc/Language/regexes.pod6
make example classier
 16:44
synopsebot Link: doc.perl6.org/language/regexes
tobs apart from the <?{}> assertion being executed more often than I would have expected, this works.
vrurg tobs: I think it's because of backtracking. 16:45
You have three letters – one call for each 16:46
tobs vrurg: ah, of course, because the match fails for each of them
vrurg Though it's not backtracking but normal forward search. 16:47
tbrowder tobs: thnx, i'll try that. the sneaky problem i think is the following code which nibbles or quibbles and shifts langs, but your syntax may work--fingers crossed!
16:47 cpan-p6 left, cpan-p6 joined, cpan-p6 left, cpan-p6 joined
tbrowder and i'm only interested in failing at the first letter encountered, maybe that constraint would help...reporting back later... 16:49
El_Che AlexDaniel`: kudos on all the project management work you're doing, by the way 16:50
16:51 Actualeyes joined, harmil left 16:56 davido left
tbrowder tobs: 17:03
vrurg: bummer, same behavior: skipping the action when <?opener> fails. 17:04
17:06 wildtrees joined
vrurg tbrowder: I would print $/ in a code block following the <?opener> and see what you have in it. 17:11
[ || ] is the construct you're looking for, as tobs example shows. 17:13
17:14 mtj_ is now known as mtj 17:17 chloekek joined, mtj is now known as mtj_ 17:18 mtj_ is now known as mtj
antoniogamiz tyil: this link is broken: git.tyil.nl/docker/perl6 17:28
(it's here hub.docker.com/r/tyil/perl6)
17:37 MilkmanDan left 17:38 MilkmanDan joined 17:39 zakharyas left
tyil yes, I'm already working on making something better as well 17:39
17:39 antoniogamiz left
tyil using automated builds on gitlab 17:40
(still inconvenient though, I'll try to update that description later today) 17:43
17:47 lizmat joined 17:50 sjm_uk joined 17:53 mowcat left 17:54 mowcat joined 17:56 antoniogamiz joined 18:00 reportable6 left, snarkyboojum__ joined, reportable6 joined, ChanServ sets mode: +v reportable6
antoniogamiz tyil: ok, thanks :) 18:01
18:03 snarkyboojum_ left 18:05 |oLa| joined, |oLa| left
tyil antoniogamiz: are you using those images right now? 18:05
or is it just something you noticed :p 18:06
18:06 rindolf joined
antoniogamiz No, but I plan to use some docker image providing perl6 18:06
because I want to avoid build it each time I push I commit
tyil ah, alright
antoniogamiz s/I/1
tyil I know the pain :p
anyway, those particular images are going to be replaced with gitlab.com/tyil/docker-perl6/tree/fresh 18:07
mostly because this way, automated builds happen every day on gitlab's hw, since mine is less trustworthy, so to speak 18:08
antoniogamiz hugely thankful
tyil++
tyil it's hopefully also easier for other people to contribute fixes like this, if they find any issues
El_Che maybe this from travis is applicable for gitlab? github.com/nxadm/rakudo-pkg#using-...-on-travis 18:09
tyil on GitLab you can just use a different image
Travis is limited by only allowing their own images
El_Che tyil: no they don't
github.com/nxadm/rakudo-pkg/blob/m...travis.yml 18:10
I get them from the docker hub or build the locally
tyil looks like you're using their image, and then run another docker inside of that
El_Che tyil: yes, not good enough? 18:11
antoniogamiz zef cache takes a while to update agh
tyil El_Che: no
as I said, you're limited by their images, you can't just use another image as your build env
18:11 pilne joined
tyil gitlab.com/tyil/perl6-app-cpan/blo...-ci.yml#L7 here I'm telling GitLab CI to use my image as the environment to work in 18:12
(or well, rakudo-star, but I can put in whatever)
El_Che for me giving access to docker means using whatever dev environment I want. But needs vary of course
tyil it's not a need, it's just a very inconvenient way of doing it with Travis
tbh, the tools available on GitHub have been tragically outdated compared to GitLab for years 18:13
El_Che we pretty much copy github at work with gitea and jenkins 18:14
I don't know why the linux admins picked that over gitlab (they tested that as well)à
tyil Im very happy we're using GitLab where I work 18:15
18:15 sena_kun joined
El_Che I am not complaining though 18:15
it does what it needs to do, and complicated stuff has been automated
tyil I do enjoy not the complicated stuff being made less complicated, though 18:16
without the first "not"
github/travis just makes it more complicated over time, whereas gitlab seems to actually try to improve things
I'm working towards automated CPAN uploads for my modules on gitlab rn 18:18
and if I get it working (which seems very likely), I'll make an advent blogpost on it, probably
18:18 Xliff joined
Xliff \o 18:18
tyil o/
18:19 Kaiepi joined
Xliff m: use NativeCall; my num32 $d = 0.1; my $a = (0.01..0.5).pick; say "Hi" if $a ~~ 0..$d 18:19
camelia 5===SORRY!5=== Error while compiling <tmp>
Cannot assign a literal of type Rat (0.1) to a native variable of type num. You can declare the variable to be of type Real, or try to coerce the value with 0.1.Num or Num(0.1)
at <tmp>:1
------> 3…
18:19 molaf joined
Xliff m: use NativeCall; my num32 $d = 0.1e0; my $a = (0.01..0.5).pick; say "Hi" if $a ~~ 0..$d 18:20
camelia Hi
Xliff m: use NativeCall; my num32 $d = 0.1e0; my $a = (0.01..0.5).pick; say "Hi ({$a})" if $a ~~ 0..$d
camelia Hi (0.01)
Xliff m: use NativeCall; my num32 $d = 0.1e0; my $a = (0.01..0.5).pick; say "Hi ({$a})" if $a ~~ 0..$d
camelia Hi (0.01)
Xliff m: use NativeCall; my num32 $d = 0.1e0; my $a = (0.01..0.5).rand; say "Hi ({$a})" if $a ~~ 0..$d
camelia ( no output ) 18:21
Xliff m: use NativeCall; my num32 $d = 0.1e0; my $a = (0.01..0.5).rand; say "Hi ({$a})" if $a ~~ 0..$d
camelia ( no output )
Xliff m: use NativeCall; my num32 $d = 0.1e0; for ^20 { my $a = (0.01..0.5).rand; say "Hi ({$a})" if $a ~~ 0..$d }
camelia Hi (0.0872051877353525)
Hi (0.09193340758775725)
Hi (0.0766649772865303)
Hi (0.08196589619195753)
Xliff m: use NativeCall; my num32 $d = 0.1e0; my $a = (0.01..0.5).rand; say "Hi ({$a})" unless $a ~~ 0..$d 18:22
camelia Hi (0.2977662991346576)
Xliff m: use NativeCall; my num32 $d = 0.1e0; my $a = (0.01..0.5).rand; say "Hi ({$a})" unless $a < 0 || $a > $d
camelia ( no output )
Xliff m: use NativeCall; my num32 $d = 0.1e0; for ^20 { my $a = (0.01..0.5).rand; say "Hi ({$a})" unless $a < 0 || $a > $d }
camelia Hi (0.05569946138556649)
Hi (0.0334992762992786)
Hi (0.01663914715469208)		 18:23
18:25 antoniogamiz left 18:27 sauvin left
Xliff m: use NativeCall; constant TIME_PER_SPARK = 0.1; my num32 $d = 0.1e0; for ^20 { my $diff-time = (0.01..0.5).rand; unless $diff-time < 0 || $diff-time > TIME_PER_SPARK { say "Hi ({$diff-time})" } } 18:30
camelia Hi (0.08352882332718155)
Hi (0.075134915851173)
Hi (0.07344216336335872)
18:32 antoniogamiz joined 18:34 robertle joined
Xliff M: say Bool.pick for ^5 18:45
m: say Bool.pick for ^5
camelia True
True
True
False
True
19:01 MasterDuke joined 19:03 antoniogamiz left 19:16 [particle]1 joined 19:17 rindolf left, [particle] left 19:20 rindolf joined 19:22 discord6 left, [particle] joined, discord6 joined 19:23 [particle]1 left 19:24 pamplemousse left 19:27 [particle] left, [Sno] joined 19:31 [particle]1 joined
masak "4 out of 5 Bools agree" 19:33
Geth doc/traps: f6039af952 | threadless-screw++ | doc/Language/traps.pod6
Rewrite of interpolation trap section
 19:35
19:39 zakharyas joined 19:53 netrino joined
Xliff masak: "1 out of 5 Bools disagree" 19:53
19:55 Doc_Holliwood joined
tbrowder vrurg: ref $/ good idea... 19:56
Doc_Holliwood So, I am trying to merge some files and skip the first line for each. This is my attempt: pastebin.com/ZuqCVeJb
vrurg tbrowder: what what the cause after all? 19:57
Doc_Holliwood Somehow, in the resulting file, there is a space character inserted at the beginning of each line, except for the very first one
vrurg is glad he was helpful.
Xliff vrurg: Thanks for your hard work on 3075. Did you raise a bug with MoarVM? 19:58
vrurg Xliff: no. I'm still have doubts. Actually, I spend all my time now trying to find a good gdb fronted. 19:59
Xliff Oh! OK. 20:00
vrurg I'm shocked how horribly bad the current status of gdb support is...
Xliff I always thought DDD was a good front end.
vrurg Xliff: somehow they don't recommend it. And it doesn't work on macos. 20:01
tbrowder Doc_Holliwood: try removing the chomp. nice handle: great movie!
vrurg is afk
Doc_Holliwood When I remove the chomp the output will be chomped and end up all in one line
Geth problem-solving/path-to-raku: 07c423e825 | (Elizabeth Mattijsen)++ | PATH-TO-RAKU.md
Initial version of the name change PR

Please note, this is only the initial version of the PR, intended for internal discussion / augmentation before being turned into a real
  ("draft") PR, for wider discussion / voting.		 20:02
Xliff vrurg: stackoverflow.com/questions/264058...s-x-10-7-5
20:05 lizmat left, [particle] joined, [particle]1 left 20:07 nepugia left 20:10 netrino_ joined 20:11 netrino left 20:18 sena_kun left
timotimo vrurg: gdbgui isn't bad, but it's browser-based which i've no clue if you're fine with that 20:21
20:25 pmurias joined 20:47 sjm_uk left 20:51 Voldenet left 20:53 mst left, mst joined 20:56 Voldenet joined, Voldenet left, Voldenet joined 20:57 Doc_Holliwood left 21:02 zakharyas left
Geth doc/regexes: 7b5d7d3eac | threadless-screw++ | doc/Language/regexes.pod6
Upgrade of Literals section
 21:04
doc/regexes: b1df019ede | threadless-screw++ | doc/Language/regexes.pod6
Some additional amendments to updated Literals section
doc/regexes: d0d57cc7c9 | threadless-screw++ | doc/Language/regexes.pod6
Fix table layout in regexes/regex interpolation section
doc: threadless-screw++ created pull request #2950:
Fix table layout in regexes/regex interpolation section		 21:06
21:09 mst left, mst joined
jdv79 i'm more on the raku side than the camelia side so yay. 21:12
hopefully it'll mean less pointless arguing and more useful spending of resources. 21:13
21:14 chloekek left
Geth doc/master: 4 commits pushed by threadless-screw++ 21:16
Xliff m: class A { has $.x; method new ($x) { self.bless(:$x); }; }; A.new(42).gist.say 21:25
camelia A.new(x => 42)
21:27 pmurias left 21:29 pmurias joined 21:46 Black_Ribbon joined 21:48 AlexDaniel joined, AlexDaniel left, AlexDaniel joined 21:53 pamplemousse joined
Geth doc/raku: 7bef410553 | Coke++ | 30 files
First pass at Perl 6 -> Raku
 22:04
22:05 rindolf left
Geth doc: coke++ created pull request #2951:
First pass at Perl 6 -> Raku		 22:07
22:12 pamplemousse left 22:19 noisegul left 22:20 Cabanossi left
Xliff say .gist for ^10.rotor(2).kv 22:21
m: say .gist for ^10.rotor(2).kv
camelia Potential difficulties:
Precedence of ^ is looser than method call; please parenthesize
at <tmp>:1
------> 3say .gist for ^107⏏5.rotor(2).kv
Xliff m: say .gist for (^10).rotor(2).kv
camelia 0
(0 1)
1
(2 3)
2
(4 5)
3
(6 7)
4
(8 9)
Xliff m: say "L: { .gist }" for (^10).rotor(2).kv 22:22
camelia L: 0
L: (0 1)
L: 1
L: (2 3)
L: 2
L: (4 5)
L: 3
L: (6 7)
L: 4
L: (8 9)
vrurg timotimo: I would be happy to get any frontend which would display source alongside with data structures. That's all I need. 22:28
22:28 lizmat joined
vrurg BTW, CLion failed ingloriously. I didn't find a way to expand pointers into arrays. :( 22:28
timotimo does it offer an "evaluate expression" thing? 22:29
i wish there were more programs out there that can display objects from gdb as graphs :\
22:30 epony left
timotimo more pretty printers and little commands for moar objects in particular would be good 22:30
like, given just a pointer to an MVMObject it should be able to just give you the body pointer cast to the right struct, since it can look what REPR it has via the STable 22:31
vrurg timotimo: I saw a warning about possible problems with pretty printers if asked to use gdb from macports. But nothing related in the context menu. 22:32
timotimo CLion told you that?
vrurg It did. I guess it needed python support, but I'm currently having problems with this too. 22:33
BTW, for R#3075 all I would need is to be able to see compunit symbol table as a tree – just to locate the place where one cu gets substituted with another one. 22:34
synopsebot R#3075 [open]: github.com/rakudo/rakudo/issues/3075 [precompilation] Conflicting use statement does not produce an error, but fails dynamic compilation.
22:34 Cabanossi joined 22:35 pmurias left
timotimo you can always write the prettyprinter in C and just "call dump_my_thingie(0xabcdef)" in gdb 22:36
that can then go off and do what-the-hell-ever, or at least i hope so
vrurg I probably must take all my courage and switch under linux vm for debugging. Even though it has many downsides.
timotimo that is how we dump P6opaque, there's a debughelper written in C 22:37
with a prettyprinter written in python, you get the ability to catch errors for when a pointer doesn't point at something valid 22:38
but writing them in python is, like, super annoying
vrurg Hm, I thought it's only available through python. Unfortunately, I didn't need to debug any compiled code for like almost 20 years, not counting for occasional minor things. Pretty much left behind.
timotimo well, "automatic" pretty printing is only available through python
though tbh you can actually write the printing logic in C, then register a prettyprinter python function that literally calls that c function via gdb 22:39
if you need an example for python-written prettyprinter stuff, the moarvm repo has one under tools/
there's also code in there to scan the heap for stuff and collect some random statistics 22:40
it's rather slow
vrurg timotimo: I saw the script, yes. Hoped it would help me with MVMObject, but then read the source. ;) 22:41
timotimo ;( ;( ;( 22:42
it's also really, really tricky to debug python scripts inside of gdb
or at least i haven't found a good way
vrurg I won't even try. Though I'm generally indifferent to other languages, by pascal and python both irritate me. :) 22:44
22:45 pmurias joined 22:46 mowcat left
Geth doc: 65f0e414d2 | (Aleks-Daniel Jakimenko-Aleksejev)++ | CREDITS
Remove obsolete pumpking mention

See github.com/rakudo/rakudo/commit/63...889c1e3e53		 22:47
22:47 epony joined 22:50 lucasb left 22:51 mowcat joined, mowcat left 22:52 mowcat joined 22:55 netrino_ left 22:56 netrino joined 23:01 satori__ joined 23:09 netrino left 23:11 netrino joined 23:19 lembark joined 23:20 pmurias left 23:22 netrino left 23:24 lembark left 23:38 lembark joined
kent\n AlexDaniel`: I've noticed at least on PR's there's a mechanism to mark comments as "hidden", with a "reason: " field ( some are 'offtopic' and 'outdated' , etc ), but not sure if that also exists on issues yet. If it does exist on issues, feel free to use it to keep the assholery to a manageable level. 23:45
23:46 [particle]1 joined
kent\n AlexDaniel: ^ in case the ` doesn't trigger you 23:46
23:46 [particle] left
AlexDaniel` Thanks 23:47
kent\n ( people who want to see them still can, its just omitted from the default view )
AlexDaniel` I'm going to bed now, will check tomorrow
kent\n that way nobody complains of censorship :)
AlexDaniel` It already hides most of the comments cuz there's just so many xd 23:48
kent\n is already mentally imagining a future where people sell pre-boxed "Camelia-pi" to schools 23:50
23:57 irced joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!