»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.
00:08 atweiden-air joined 00:09 p6bannerbot sets mode: +v atweiden-air
atweiden-air rakudo has landed in homebrew github.com/Homebrew/homebrew-core/...1c9fce7b68 00:09
00:10 Jmlv joined, p6bannerbot sets mode: +v Jmlv 00:11 Jmlv left 00:14 scovit left, dct left
Geth doc: 2dcce0a0b0 | cfa++ | doc/Language/traps.pod6
Formatting.
 00:15
synopsebot Link: doc.perl6.org/language/traps
00:17 cpage_ joined 00:18 p6bannerbot sets mode: +v cpage_ 00:19 cpage left, cpage_ is now known as cpage, MilkmanDan left 00:20 MilkmanDan joined, p6bannerbot sets mode: +v MilkmanDan 00:23 rindolf left 00:29 cfa left 00:40 MorayJ left
lookatme_q morning #perl6 00:43
sena_kun o/ 00:46
00:46 atweiden-air left
lookatme_q sena_kun, o/ 00:47
sena_kun though it's 02:47 here. :)
lookatme_q here is 08:47 00:48
sena_kun all hail timezones! 00:49
lookatme_q haha :) 00:50
sena_kun anyway, I'll go to sleep soon, I think 00:51
lookatme_q sena_kun, good night, bye :) 00:52
sena_kun just give me 5 more minutes to stare into to code...
01:10 scovit joined 01:11 p6bannerbot sets mode: +v scovit 01:14 DarthGandalf joined, p6bannerbot sets mode: +v DarthGandalf 01:18 lizmat left 01:19 sena_kun left
zachk how can i compare types of variables? 01:29
timotimo you can get an object's type object with the .WHAT pseudo-method
if you are interested in the variable, actually, like the constraint it was declared with, you'll need to access the .VAR 01:30
01:34 Ven`` joined 01:35 p6bannerbot sets mode: +v Ven``
zachk but say i want to compare two variables and see if their types are equal? 01:47
01:51 kurahaupo_ joined 01:52 kurahaupo left, p6bannerbot sets mode: +v kurahaupo_
timotimo there are many different ways to interpret that 01:58
m: my Int $foo; my Str $bar; say "$foo.VAR.WHAT.^name() equal to $bar.VAR.WHAT.^name()?"
camelia Scalar equal to Scalar?
timotimo m: my Int $foo; my Str $bar; say "$foo.VAR.of.^name() equal to $bar.VAR.of.^name()?" 01:59
camelia Int equal to Str?
timotimo m: my Int $foo = <0xff>; my Int $bar = 255; say "$foo.VAR.of.^name() equal to $bar.VAR.of.^name()?"; say "$foo equal to $bar?"
camelia Int equal to Int?
0xff equal to 255?
timotimo m: my Int $foo = <0xff>; my Int $bar = 255; say "$foo.WHAT.^name() equal to $bar.WHAT.^name()?"; say "$foo equal to $bar?"
camelia IntStr equal to Int?
0xff equal to 255?
timotimo what semantics are you interested in? 02:00
the kind of variable, aka Scalar vs Proxy vs Native Reference? 02:01
the constraint the variable was declared with? i.e. my Int $foo vs my Str $bar?
the type of the value that's in the variable? that'd be $foo.WHAT vs $bar.WHAT
and do you want to accept subclasses of one in the other?
02:02 mowcat joined, p6bannerbot sets mode: +v mowcat
timotimo i think what you want will be $foo.WHAT === $bar.WHAT 02:03
not many prog languages have first-class scalar variables or introspectable lexicals or stuff like that 02:04
Ven`` certainly can't think of many off the top of their head 02:14
m: subset MyInt of Int; say MyInt ~~ Int; say Int ~~ MyInt; 02:15
camelia True
True
Ven`` interesting
02:22 jme` joined 02:23 p6bannerbot sets mode: +v jme`, Ven`` left
vrurg Can I have own control exceptions defined? I.e. to have them processed in a CONTROL block? Application of X::Control is not sufficient, seemingly. 02:26
zachk whats that triple equals? === ? 02:27
vrurg zachk: same object equality. 02:28
m: my $a = 1; my $b = 1; say $a === $b; $b := $a; say $b === $a
camelia True
True
vrurg Ah, constant. 02:30
m: class Foo {}; say Foo.new === Foo.new
camelia False
lookatme_q m: say 1 == True; say 1 === True; say True === True; 02:32
camelia True
False
True
lookatme_q zachk, you can using === check if they are same object
zachk does that compare instance variables too inside an object instance? 02:33
02:38 ryn1x joined 02:39 p6bannerbot sets mode: +v ryn1x
vrurg zachk: It would compare whatever is in the variables. 02:41
02:51 melezhik joined, p6bannerbot sets mode: +v melezhik
melezhik Hi! How do I declare dependency on version equal or greater in META.json file? 02:52
02:54 molaf left
lookatme_q melezhik, I don't think they support that things 02:54
melezhik yeah ... I guessed ... just wanted to ensure ... 02:55
02:58 Kaiepi left
lookatme_q melezhik, maybe you can submit a ticket for that, but IDK which repo to 02:59
zachk so my code was working, compiled and installed the latest perl6, now I get this This Seq has already been iterated, and its values consumed 03:02
lookatme_q zachk, refer this documents: docs.perl6.org/type/Seq 03:03
03:09 kurahaupo_ left 03:10 kurahaupo joined, p6bannerbot sets mode: +v kurahaupo
zachk do i still need to apply the execstack fix for ubuntu on windows to rakudo? 03:11
er moarvm or whichever
lookatme_q yes, I think so, for moarvm
03:23 ryn1x left 03:27 zachk left 03:32 w_richard_w left 03:42 melezhik left 03:52 leont left 04:34 Cabanossi joined, Cabanoss- left 04:35 p6bannerbot sets mode: +v Cabanossi 04:42 sauvin joined, p6bannerbot sets mode: +v sauvin 04:47 jast left 04:50 jast joined 04:51 p6bannerbot sets mode: +v jast 05:02 ferreira left 05:05 jme` left 05:16 molaf joined 05:17 p6bannerbot sets mode: +v molaf 05:23 w17t joined, p6bannerbot sets mode: +v w17t 05:28 reach_satori left 05:41 w_richard_w joined 05:42 p6bannerbot sets mode: +v w_richard_w 05:44 w_richard_w1 joined 05:45 p6bannerbot sets mode: +v w_richard_w1, w_richard_w left 05:46 zacts joined, p6bannerbot sets mode: +v zacts 05:48 MilkmanDan left 05:49 kurahaupo left, MilkmanDan joined, p6bannerbot sets mode: +v MilkmanDan, kurahaupo joined 05:50 p6bannerbot sets mode: +v kurahaupo 05:52 curan joined, p6bannerbot sets mode: +v curan 05:56 reach_satori joined 05:57 p6bannerbot sets mode: +v reach_satori 06:03 ZzZombo joined, p6bannerbot sets mode: +v ZzZombo 06:10 zacts left 06:18 vrurg left 06:20 jmerelo joined 06:21 p6bannerbot sets mode: +v jmerelo 06:26 domidumont joined, p6bannerbot sets mode: +v domidumont 06:27 molaf left
SmokeMachine Isn’t just: “MudileName:ver<2.0>”? 06:28
06:29 vrurg joined 06:30 p6bannerbot sets mode: +v vrurg
SmokeMachine .tell melezhik something like this: "Pod::To::Markdown:ver<0.1.4+>" 06:36
yoleaux SmokeMachine: I'll pass your message to melezhik.
SmokeMachine lookatme_q: something like this: "Pod::To::Markdown:ver<0.1.4+>"
06:44 vrurg left
jmerelo squashable6: status 06:45
squashable6 jmerelo, Next SQUASHathon in 24 days and ≈3 hours (2019-02-02 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
jmerelo releasable6: status
releasable6 jmerelo, Next release in ≈11 days and ≈12 hours. 3 blockers. 0 out of 173 commits logged
jmerelo, Details: gist.github.com/64dbb6ca855c7b2ebf...ee28ce5ce1
06:46 domidumont left
jmerelo .tell AlexDaniel I'll go for Kaiepi as a winner. Actually, you made the most contributions, but I guess you're not eligible 06:47
yoleaux jmerelo: I'll pass your message to AlexDaniel.
jmerelo .tell Kaiepi congratulations! You're the winner of the plush Camelia for your contributions to the last Squashathon!
yoleaux jmerelo: I'll pass your message to Kaiepi.
jmerelo .tell Kaiepi to contact lizmat to get your plush Camelia shipped to you. 06:53
yoleaux jmerelo: I'll pass your message to Kaiepi.
jmerelo .tell AlexDaniel I'm going to propose Squashathons for the next 6 months. Feedback is welcomed and appreciated. 06:59
yoleaux jmerelo: I'll pass your message to AlexDaniel.
jmerelo .tell AlexDaniel do you have some script to extract info from the log? I seem to remember you did... 07:00
yoleaux jmerelo: I'll pass your message to AlexDaniel.
07:02 domidumont joined, p6bannerbot sets mode: +v domidumont
jmerelo SmokeMachine: Why did you remove Red from the ecosystem? Any major problem? 07:08
squashable6: status 07:19
squashable6 jmerelo, Next SQUASHathon in 24 days and ≈2 hours (2019-02-02 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
jmerelo ^^^^ OK, I have inserted squashathon proposals for the next 6 months. Comments and suggestions welcome and appreciated
And we're still commenting on the issue of splitting the doc repository: github.com/perl6/doc/issues/2542 07:20
07:21 scovit_ joined, p6bannerbot sets mode: +v scovit_ 07:24 scovit left 07:26 jmerelo left 07:53 mowcat left 08:11 w_richard_w1 left 08:15 robertle_ joined 08:16 p6bannerbot sets mode: +v robertle_ 08:27 kurahaupo left 08:28 w_richard_w joined, p6bannerbot sets mode: +v w_richard_w 08:31 sftp left
Geth doc: d7841111e6 | (JJ Merelo)++ | doc/Language/modules.pod6
Restores the information on the `api` field

As requested by @jonathanstowe and @ugexe		 08:35
synopsebot Link: doc.perl6.org/language/modules
08:38 sftp joined, p6bannerbot sets mode: +v sftp 08:42 w_richard_w left 08:45 dakkar joined, p6bannerbot sets mode: +v dakkar 08:50 patrickb joined 08:51 p6bannerbot sets mode: +v patrickb 09:00 w17t left
lookatme_q SmokeMachine, don't know this usage, is that mention in the documents ? :) 09:02
09:08 w17t joined, p6bannerbot sets mode: +v w17t 09:18 rindolf joined, p6bannerbot sets mode: +v rindolf 09:26 ZzZombo left 09:41 reach_satori left
AlexDaniel . 09:42
yoleaux 06:47Z <jmerelo> AlexDaniel: I'll go for Kaiepi as a winner. Actually, you made the most contributions, but I guess you're not eligible
06:59Z <jmerelo> AlexDaniel: I'm going to propose Squashathons for the next 6 months. Feedback is welcomed and appreciated.
07:00Z <jmerelo> AlexDaniel: do you have some script to extract info from the log? I seem to remember you did...
09:56 kensanata joined, p6bannerbot sets mode: +v kensanata 10:09 scovit_ left 10:13 dpk joined 10:14 p6bannerbot sets mode: +v dpk 10:15 w17t left 10:16 Kaiepi joined 10:17 p6bannerbot sets mode: +v Kaiepi 10:21 lizmat joined, p6bannerbot sets mode: +v lizmat 10:22 w17t joined, p6bannerbot sets mode: +v w17t 10:34 Kaiepi left, zakharyas joined, reach_satori joined, w17t left 10:35 p6bannerbot sets mode: +v zakharyas, p6bannerbot sets mode: +v reach_satori 10:36 w17t joined, p6bannerbot sets mode: +v w17t 10:37 Kaiepi joined 10:38 p6bannerbot sets mode: +v Kaiepi 11:10 mbichon joined, p6bannerbot sets mode: +v mbichon 11:14 mbichon left 11:21 domidumont left 11:41 sena_kun joined 11:42 scovit joined, p6bannerbot sets mode: +v sena_kun 11:43 p6bannerbot sets mode: +v scovit 11:45 Xliff_ left 11:47 Xliff joined 11:48 p6bannerbot sets mode: +v Xliff 11:51 Xliff left, Xliff joined, mike_swie joined, p6bannerbot sets mode: +v mike_swie 11:52 p6bannerbot sets mode: +v Xliff
lizmat And yet another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2019/01/08/...-for-2019/ 11:57
11:59 rindolf left, Xliff left 12:01 rindolf joined
moritz lizmat++ 12:01
retweeted
12:02 p6bannerbot sets mode: +v rindolf, Xliff joined
mike_swie lizmat++ 12:02
12:03 p6bannerbot sets mode: +v Xliff
mike_swie I'm assuming the ++ is a thanks, or like, or complimentary. If I got that wrong, please forgive the accidental etiquette breach. 12:03
lizmat mike_swie: it's interpreted as such 12:04
12:04 scovit left
lizmat the usage is from the days that bots where roaming the channels assigning karma :-) 12:04
it indicates an increase in karma after things done 12:05
12:05 yoleaux joined, ChanServ sets mode: +v yoleaux
lizmat prefix ++ indicates karma to assigned for things to be done still 12:05
mike_swie Heh. I was using telnet chat in the '90s, I never got into IRC, Usenet, and similar. Funny how I missed them all at the time.
12:06 p6bannerbot sets mode: +v yoleaux, AlexDaniel left
masak it doesn't matter that bots are not actually accumulating karma at any given moment 12:07
somewhere, somehow, A Score is Being Kept 12:08
moritz Karma keeps score :)
masak exactly
anyway, lizmat++ for the post
Xliff masak: Ah! Spirituality! Surprising! :)
yoleaux 1 Jan 2019 12:34Z <jjmerelo> Xliff: it was a generic "Advent calendar" post, not specifically devoted to Perl 6. It was published Dec 2nd, I think.
masak and I had missed the thing with Zoffix leaving. that's sad. :/ 12:09
12:09 reach_satori left
Xliff Ah.... Slashdot. Those were the days. 12:09
masak almost wrote something about the Naming/Alias issue, but instead decides not to 12:10
SmokeMachine .tell jmerelo I removed Red from the ecosystem because I sent it to CPAN 12:17
yoleaux SmokeMachine: I'll pass your message to jmerelo.
SmokeMachine .tell lizmat Any reason to not mentioning Red on p6weekly? 12:25
yoleaux SmokeMachine: I'll pass your message to lizmat.
lizmat SmokeMachine: it's mentioned in the New Modules section 12:26
yoleaux 12:25Z <SmokeMachine> lizmat: Any reason to not mentioning Red on p6weekly?
SmokeMachine Sorry o just saw it!!!
lizmat and it's the module I consider the potential killer feature
SmokeMachine I think I’m becoming blind... 12:27
lizmat just write a nice blog post about it and you'll get much more exposure :-)
SmokeMachine lizmat: thanks!
12:28 reach_satori joined
SmokeMachine lizmat: I didn’t mean mentioning outside the new modules... I couldn’t see the on the new modules area... (I’ve just waked up) 12:29
12:29 Xliff left, p6bannerbot sets mode: +v reach_satori
SmokeMachine lizmat: sorry... 12:29
lizmat SmokeMachine: should I mention it as RED (in full capitals) in the future ? 12:32
then it will stand out more ? 12:33
SmokeMachine I call it Red... I have not decided why it’s called Red yet... :) 12:34
lizmat Really Excellent Databaseabstraction 12:35
timotimo 04Red
SmokeMachine :)
timotimo ReDorm
12:37 MorayJ joined, p6bannerbot sets mode: +v MorayJ 12:40 Guest87s joined, p6bannerbot sets mode: +v Guest87s
lizmat afk for a bit& 12:46
12:50 reach_satori left 12:55 scovit joined 12:56 p6bannerbot sets mode: +v scovit 12:58 domidumont joined 12:59 p6bannerbot sets mode: +v domidumont, tobs` joined 13:00 p6bannerbot sets mode: +v tobs` 13:03 scovit left 13:05 Guest87s left 13:08 lucasb joined, p6bannerbot sets mode: +v lucasb, AlexDaniel joined, p6bannerbot sets mode: +v AlexDaniel 13:29 salva joined 13:30 p6bannerbot sets mode: +v salva 13:32 w17t left
lucasb I just noticed the comments about the departure of our fellow developer. I'm sorry to hear that and hope he returns when he wants. 13:47
His modules should be preserved. Anyone did that already? 13:49
maybe a github.com/zoffixmirror org? :)
lizmat AlexDaniel has rights to Zoffix' repos as far as I knonw 13:51
*know
sena_kun isn't moving those under perl6-community-modules org enough?
lucasb yes, that should work 13:52
13:57 Xliff joined, sjoshi joined, sjoshi left, p6bannerbot sets mode: +v Xliff 14:05 AlexDaniel left 14:06 reach_satori joined, leont joined 14:07 p6bannerbot sets mode: +v reach_satori, p6bannerbot sets mode: +v leont 14:12 HaraldJoerg joined 14:13 p6bannerbot sets mode: +v HaraldJoerg
HaraldJoerg p6: say २०१९; 14:13
camelia 2019
tadzik :o 14:14
14:14 scovit joined
HaraldJoerg p6: say '२०१९' 14:14
camelia २०१९
14:15 p6bannerbot sets mode: +v scovit
HaraldJoerg p6: say २०१९ - ٢٠١٩ 14:16
camelia 0
HaraldJoerg Is there a way to have Perl 6 *write* integers as non-ASCII numerals? 14:17
timotimo well, there is a module for roman numerals 14:18
SmokeMachine there is a module for japanese too...
HaraldJoerg I know this one, but I'm more interested in extant systems 14:19
14:22 vrurg joined
timotimo it's relatively easy if the other numeral system is 10-based 14:22
14:23 p6bannerbot sets mode: +v vrurg
HaraldJoerg I've used two 10-based here: indian and devanagari 14:23
Geth DBIish: jnthn++ created pull request #137:
Performance improvements, with a focus on SQLite and re-use of statement handles
14:23 leont left
jnthn 5x-6x faster DBIish, anyone? :) 14:23
(OK, only for one benchmark, but still... :)) 14:24
timotimo um, yes please?!?
sena_kun >+16 −14
masak \o/
sena_kun line count is just wow. :)
14:25 tankf33der joined, p6bannerbot sets mode: +v tankf33der 14:27 andrzejku left
masak I read "TRUE_ZERO" entirely the wrong way :) 14:27
jnthn :P
masak "wait, but isn't that... hasn't it been... oooooooh"
14:28 curan left
lizmat commute to Amsterdam.pm& 14:28
14:29 lizmat left
AlexDaniel` Yes, I should move the modules eventually 14:36
[Coke] the memory panic I got for 32-bit moarvm on windows doesn't happen on the R* 64bit distro.
HaraldJoerg p6: my $year= २०१९; $year++ 14:39
camelia ( no output )
HaraldJoerg p6: my $year= २०१९; say ++$year
camelia 2020
HaraldJoerg p6: my $year= "२०१९"; say ++$year 14:40
camelia २०२०
14:53 fluca1978 joined 14:54 p6bannerbot sets mode: +v fluca1978 14:59 zakharyas left
HaraldJoerg my $year = 𝟐𝟎𝟏𝟗; say ++$year 15:05
evalable6 2020
HaraldJoerg p6: my $year = "𝟐𝟎𝟏𝟗"; say ++$year
camelia 𝟐𝟎𝟏𝟗
HaraldJoerg p6: my $year = 𝟐𝟎𝟏𝟗; say ++$year
camelia 2020
15:23 fluca1978 left 15:25 fluca1978 joined 15:26 p6bannerbot sets mode: +v fluca1978, zacts joined 15:27 p6bannerbot sets mode: +v zacts
Geth DBIish/master: 7 commits pushed by (Jonathan Worthington)++, (Salvador Ortiz)++ 15:28
15:28 aindilis left
Kaiepi aw man zoffix quit? 15:29
yoleaux 06:47Z <jmerelo> Kaiepi: congratulations! You're the winner of the plush Camelia for your contributions to the last Squashathon!
06:53Z <jmerelo> Kaiepi: to contact lizmat to get your plush Camelia shipped to you.
Kaiepi oh sweet 15:30
.tell lizmat msg me once you're on 15:31
yoleaux Kaiepi: I'll pass your message to lizmat.
15:31 aindilis joined 15:32 p6bannerbot sets mode: +v aindilis 15:33 glom joined, p6bannerbot sets mode: +v glom
tadzik the backstories of those commits are fun to read, like github.com/perl6/DBIish/commit/113...c237e6b656 15:38
github.com/perl6/DBIish/commit/972...b5392e715d I find the most surprising 15:40
Kaiepi i had no idea conditional method calls were possible
15:41 glom left
fluca1978 I read about conditional method calls in brian d foy's book 15:46
however I did not realized it was optimized with regard to .^can 15:47
jnthn It's optimized to the point that, if in type-specialized code, it can even throw out the conditional part, rendering it as a constant or a normal method call, and the latter is then an inline candidate. 15:48
Kaiepi brian d foy's book?
jnthn So in the best cases it's "free"
fluca1978 Kaiepi:Learning Perl 6: Keeping the Easy, Hard, and Impossible 15:49
jnthn Goodness, punctuation really saves that book title :P
fluca1978 anyone can help me understanding the relationship between nqp and rakudo? I mean, nqp is somehow used to implement rakudo (bootstrapping) or am I misunderstanding? 15:51
Kaiepi is it worth giving a read if you've already been using perl 6 for around a year?
jnthn fluca1978: It's a little confusing in that we tend to use "NQP" to refer to two things. One of them is NQP the language, which is a bootstrapped subset of Perl 6 that's easier to implement and optimize. 15:52
fluca1978 Kaiepi: I don't think, one year is enough to get all the concepts expressed in the book. But, on the other side, brian is a very proficent developer and writer, so...
jnthn However, also contained in the NQP repository is a bunch of compiler infrastructure (such as code generation) that Rakudo uses too 15:53
So you can see NQP as being a mini Perl 6 that includes a compiler toolkit in its standard library. :)
fluca1978 jnthn: but what compiles NQP then? 15:54
jnthn fluca1978: NQP. The repository includes a snapshot that is good enough to compile the current version. 15:55
fluca1978: Of course, if you go back enough years, there was an NQP written in something else.
Just like there was a first C compiler not written in C, for example. 15:56
fluca1978 jnthn: therefore, there is a binary snapshot of NQP to bootstrap the compiler that in turn compiles NQP the language, so in this sense it is meta-circular, right? 15:57
sena_kun it is 15:58
jnthn fluca1978: Yes
At some point it stops hurting to think about. :) 15:59
fluca1978 and what does it mean that NQP is lighter than Perl6? I mean, NQP has classes itself, so what makes the runtime smaller?
[Coke] supports less language syntax, for example 16:00
fewer builtin classes.
jnthn It has simplified/cheating versions of a lot of Perl 6 language features, and then excludes quite a number of other features.
[Coke] only binding, no assignment. 16:01
jnthn That also, which is a big simplification.
Also no laziness
No high-level concurrency stuff
No operator overloading or custom operators, etc. 16:02
fluca1978 got it
vrurg The above mentioning of .^can reminded me: what is the best way of getting method object of a class? So far, I'm stuck to find_method. 16:03
jnthn vrurg: What does "best" mean? .^find_method is best if you're just after something to invoke; .^lookup is best if you're after it for introspection.
Xliff jnthn: At which point does it stop hurting, and when it does, should I worry? 16:04
16:04 simple joined, p6bannerbot sets mode: +v simple
Xliff (and you are right about that title! LOL) 16:04
vrurg jnthn: For invoking. Thanks! I just thought that I might be missing something like &sub but for methods. 16:05
jnthn Xliff: I guess the point I really got settled with it was when I was working on the JVM porting, and realized that one can write a bytecode generator, then run the compiler on the VM that already exists but have it "cross-compile", and then use that cross-compiler to compile NQP itself. 16:06
So basically, you get it to spit itself out but for another virtual machine
I've done that twice over by now. Whether that's worrying, well... :-)
Xliff So kinda like Java and Android?!? 16:07
16:07 jmerelo joined
jmerelo O/ 16:07
yoleaux 09:51Z <AlexDaniel> jmerelo: no, I just looked at the numbers and wrote the summary :)
12:17Z <SmokeMachine> jmerelo: I removed Red from the ecosystem because I sent it to CPAN
fluca1978 Xliff: java and jikesrvm
16:08 p6bannerbot sets mode: +v jmerelo
Xliff fluca1978: So similar situation. jikesrvm uses the Java language spec, but does not compile to the JVM 16:08
Android apps work in the same way.
fluca1978 Xliff: jikesrvm is written in java and does compile to java bytecode (JIT), it bootstraps out of an image obtained by another virtual machine 16:09
jnthn Yeah, it's all the same kind of stuff.
16:11 ferreira1 joined, p6bannerbot sets mode: +v ferreira1
Xliff OK. So meta-circular. Got it. 16:12
Would that mean that NQP should be on this list: en.wikipedia.org/wiki/Meta-circular_evaluator
16:12 fluca1978 left
sena_kun is NQP interpreting though? 16:13
as in, a C compiler written in C doesn't really belong to the list. 16:14
jnthn Yeah, I'd tend to refer to it as a "bootstrapped compiler"
sena_kun yup
Kaiepi i only just noticed the MONKEY-SEE-NO-EVAL pragma's a pun 16:17
16:17 cfa joined, p6bannerbot sets mode: +v cfa
Kaiepi smh putting puns in my code 16:18
16:19 kensanata left 16:28 zacts left
SmokeMachine how can I create a function that `is native` and returns this (code.woboq.org/gcc/gcc/jit/libgccj..._context)? 16:32
I mean this function: code.woboq.org/gcc/gcc/jit/libgccj...xt_acquire 16:34
Xliff SmokeMachine: Any reason you can't use a CPointer repr for that?
class gcc_jit_context is repr('CPointer') { }; 16:35
SmokeMachine Xliff: Im trying this... but doesnt feel right... www.irccloud.com/pastebin/EMO9XDFb/
Xliff sub my_jit_func(...) returns gcc_hit_context is native (...)
Why doesn't it feel right? 16:36
SmokeMachine Xliff: this? www.irccloud.com/pastebin/nB4tDswi/ 16:37
16:37 qqq joined, p6bannerbot sets mode: +v qqq
Xliff No. Your first one was better. 16:37
16:38 qqq left
Xliff 2nd one isn't wrong. You just wouldn't need GccJitContext.new. 16:38
GccJitCont3ext.new doesn't really do anything.
my $jitc = gcc_jit_context_aquire() would return a GccJitContext, though. 16:39
16:39 zacts joined, p6bannerbot sets mode: +v zacts
SmokeMachine Xliff: you mean this? www.irccloud.com/pastebin/YwxZ69N9/ 16:45
16:47 domidumont left
Xliff Looks fine. You won't know until to try to do something with that pointer. 16:47
sjn o/ 16:48
Xliff You used gcc_jit_context_get_type() and got something back. That's a good sign.
\o sjn
16:51 robertle_ left
sjn .tell lizmat link in 2019.01 weekly to Brad Gilbert's tweet seems wrong 16:52
yoleaux sjn: I'll pass your message to lizmat.
17:00 mux__ joined, p6bannerbot sets mode: +v mux__
mux__ I'm experiencing a strange behavior from perl6 + tmux. I want to record the cursor position and restore it using tput. 17:01
This works as expected in a bare Xterm or VT: perl6 -e'print "save"; run <tput sc>; say ""; sleep 1; print "restore"; run <tput rc>; sleep 1; print 42'
But running it in tmux seems to only restore the column, not the row.
Using tput in the shell works just fine everywhere though: echo -n save; tput sc; sleep 1; echo; sleep 1; echo -n restore; sleep 1; tput rc; echo 42
Maybe someone here has a clue whats going on? 17:02
17:04 domidumont joined
timotimo huh, interesting. can you try outputting %*ENV<TERM> inside perl6 and echo $TERM outside perl6? 17:05
i see no reason for them to differ, but if they do it'd be a sure sign for trouble
17:05 p6bannerbot sets mode: +v domidumont
mux__ timotimo: No, it's the same in both 17:08
Kaiepi how do i get the \c name for a character?
17:08 AlexDaniel joined, p6bannerbot sets mode: +v AlexDaniel
SmokeMachine Xliff: until the CArray it was working... www.irccloud.com/pastebin/FdNjcm7R/ 17:15
jnthn Kaiepi: If you have the codepoint? uniname, I think
SmokeMachine Xliff: wrong parameter value... 17:18
but with error yet...
xLIFF: www.irccloud.com/pastebin/eoP5pRIQ/
Xliff: ^^
Kaiepi m: '😂'.uniname 17:19
camelia ( no output )
Xliff I'm reading. Be aware, I am also trying to eat lunch. :)
Kaiepi m: say '😂'.uniname
camelia FACE WITH TEARS OF JOY
Kaiepi thanks
m: sub term:<<\c[FACE WITH TEARS OF JOY]>> is rw { state $ = 0 }; 😂 = e ** (i * pi); say 😂 17:20
camelia -1+1.2246467991473532e-16i
Kaiepi shouldn't that be -1?
SmokeMachine Fixed!!! wrong parameters! 17:21
Xliff Ah! OK. 17:23
17:23 zacts left
Xliff m: i².say 17:27
camelia -1+1.2246467991473532e-16i
AlexDaniel Kaiepi: isn't it close enough?
e-16 that's really small
Xliff That's a really small imaginary portion.
AlexDaniel yeah 17:28
17:28 lizmat joined, p6bannerbot sets mode: +v lizmat
Xliff Not smallest representable floating point value, though. 17:28
That's probably a residue of the imaginary math, though. 17:29
17:30 dakkar left 17:31 dakkar joined, zacts joined, p6bannerbot sets mode: +v zacts, p6bannerbot sets mode: +v dakkar
[Coke] e and pi are approximate constants here. 17:31
moritz just standard floating point arithmetics 17:32
SmokeMachine how can I receive a function pointer?
Xliff SmokeMahcine: That's one thing NativeCall does NOT make easy. 17:33
s/SmokeMahcine/SmokeMachine/
SmokeMachine :(
Xliff Besides... what good is a C function pointer going to do for you in Perl6?
SmokeMachine I think that's the last think I need to make my sample work...
Kaiepi how do i generate a range like 1,2,3,3,4,5,5,6,7,7,8,9? 17:34
Xliff (there is utility, but it's limited)
17:34 rindolf left
SmokeMachine Xliff: please, show me 17:34
Xliff SmokeMachine: I can't.
17:34 rindolf joined
Xliff I mean, about the best you can do is say "returns Pointer" 17:35
sena_kun Kaiepi, what's the rule of this progression? are 1,2,9 hard-coded?
17:35 p6bannerbot sets mode: +v rindolf
Kaiepi it starts with 1,2,3, then the next three digits are the last digit of the previous triple and its consecutive digits 17:36
so 1,2,3, then 3,4,5, then 5,6,7, etc.
SmokeMachine Xliff: but how do I call it?
Xliff SmokeMachine: You can't. 17:37
Not from Perl6.
sena_kun hmm... that seems quite specific to me, I'd go with a simple iterative approach to make later understanding easier. though it sounds like a golfing task where cryptic == better, yes?
Xliff SmokeMachine: About the only thing you could do is write a wrapper that would take the pointer and do the invocation from C. Then you have the whole hassle of parameters.
C calling C is best left to C. 17:38
sena_kun, Kaiepi: That looks like something better for Iterator.pull-one 17:39
s/better/better-suited/
Since some numbers in the sequence are duped.
Kaiepi m: my @nums = lazy gather { my $x = 1; loop { take ($x, ++$x, ++$x) } }; say @nums[0..^3]
camelia ((7 2 3) (7 4 5) (7 6 7))
17:40 molaf joined
Kaiepi 7? 17:40
m: my @nums = lazy gather { my $x = 1; loop { take ($x, $x + 1, $x + 2); $x += 2 } }; say @nums[0..^3]
camelia ((5 2 3) (5 4 5) (5 6 7))
17:40 p6bannerbot sets mode: +v molaf
Kaiepi m: my @nums = lazy gather { my $x = 1; loop { take $x..$x+2; $x += 2 } }; say @nums[0..^3] 17:41
camelia (1..3 3..5 5..7)
Kaiepi m: my @nums = lazy gather { my $x = 1; loop { take $x..$x+2.flat; $x += 2 } }; say @nums[0..^3]
camelia (1..2 3..4 5..6)
Kaiepi m: my @nums = lazy gather { my $x = 1; loop { take $x..$x+2.List; $x += 2 } }; say @nums[0..^3]
camelia (1..2 3..4 5..6)
Kaiepi m: my @nums = lazy gather { my $x = 1; loop { take $x..$x+2.list; $x += 2 } }; say @nums[0..^3]
camelia (1..2 3..4 5..6)
Kaiepi m: my @nums = lazy gather { my $x = 1; loop { take ($x..$x+2).list; $x += 2 } }; say @nums[0..^3] 17:42
camelia ((1 2 3) (3 4 5) (5 6 7))
[Coke] wonders if he can get unicode working this cmd.bat/ssh/tmux combo
Kaiepi there we go
17:43 dakkar left
Geth doc: 1075133cef | (JJ Merelo)++ | doc/Language/modules.pod6
Moves `source-type` to a footnote refs #2543
 17:45
doc: 72c02cbfe0 | (JJ Merelo)++ | 2 files
Extends the description of the *-depends fields.

Including the *hash* form, with examples. Closes #2543.
synopsebot Link: doc.perl6.org/language/modules
17:52 mux__ left
Xliff m: my @nums = lazy gather { my $x = 1; loop { take ($x..$x+2).list; $x += 2 } }; @nums[0..2].map(|*).say 17:53
camelia (1 2 3 3 4 5 5 6 7)
sena_kun m: my @nums = lazy gather { my $x = 1; loop { take ($x..$x+2).list; $x += 2 } }; @nums[0..2].join.say 17:54
camelia 1 2 33 4 55 6 7
sena_kun ouch
Xliff :)
sena_kun shame on me. :P
Kaiepi m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my Rat $total = 3.Rat; for @denominators -> $a, $b { $total = $total + 4/$a - 4/$b; }; $total }; say pi(100) 17:56
camelia Type check failed in assignment to $total; expected Rat but got Num (3.1415766854350315e0)
in sub pi at <tmp> line 1
in block <unit> at <tmp> line 1
Kaiepi m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my Num $total = 3.Num; for @denominators -> $a, $b { $total = $total + 4/$a - 4/$b; }; $total }; say pi(100)
m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my Num $total = 3.Num; for @denominators -> $a, $b { $total = $total + 4/$a - 4/$b; }; $total }; say pi(1)
m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my $total = 3; for @denominators -> $a, $b { $total = $total + 4/$a - 4/$b; }; $total }; say pi(100)
m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my $total = 3; for @denominators[0..$prec] -> $a, $b { $total = $total + 4/$a - 4/$b; }; $total }; say pi(100) 17:57
camelia (timeout)
Too few positionals passed; expected 2 arguments but got 1
in sub pi at <tmp> line 1
in block <unit> at <tmp> line 1
Kaiepi m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my $total = 3; for @denominators[0..^$prec].rotor(2) -> ($a, $b) { $total = $total + 4/$a - 4/$b; }; $total }; say pi(100)
camelia 3.141592410971983
Kaiepi m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my $total = 3; for @denominators[0..^$prec].rotor(2) -> ($a, $b) { $total = $total + 4/$a - 4/$b; }; $total }; say pi(1000)
camelia 3.1415926533405445
[Coke] m: say (1..*).rotor(3=>-1).flat[^10] 17:58
camelia (1 2 3 3 4 5 5 6 7 7)
Kaiepi m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my Rat $total = 3.Rat; for @denominators[0..^$prec].rotor(2) -> ($a, $b) { $total = $total + Rat(4/$a) - Rat(4/$b); }; $total }; say pi(10000) 17:59
camelia Type check failed in assignment to $total; expected Rat but got Num (3.1415766854350315e0)
in sub pi at <tmp> line 1
in block <unit> at <tmp> line 1
Kaiepi m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my Rat $total = 3.Rat; for @denominators[0..^$prec].rotor(2) -> ($a, $b) { $total = Rat($total + 4/$a - 4/$b); }; $total }; say pi(10000)
camelia 3.141592920353982451
Kaiepi m: sub pi(Int $prec) { my @denominators = lazy gather { my Int $x = 2; loop { take [*] ($x..$x+2).list; $x += 2; } }; my Rat $total = 3.Rat; for @denominators[0..^$prec].rotor(2) -> ($a, $b) { $total = Rat($total + 4/$a - 4/$b); }; $total }; say e ** (i * pi(10000)) 18:00
camelia -0.9999999999999645-2.6676418928249877e-07i
Kaiepi m: say (1..*).rotor(3 => -1).rotor(3).first(10)
m: say (1..*).list.rotor(3 => -1).rotor(3).first(10) 18:01
m: say (1..*).list.rotor(3 => -1).rotor(3).first
evalable6, say 1
evalable6 1
camelia (timeout)
((1 2 3) (3 4 5) (5 6 7))
Kaiepi evalable6, say (2..*).rotor(3 => -1).first(10) 18:02
evalable6, say (2..*).list.rotor(3 => -1).first(10)
evalable6, [1,2,3].first(2)
evalable6 (signal SIGHUP) «timed out after 25 seconds»
(signal SIGHUP) «timed out after 25 seconds» 18:03
Kaiepi evalable6, say [1,2,3].first(3)
evalable6 3
Kaiepi evalable6, say [1,2,3].first(0..^3)
evalable6 1
Kaiepi evalable6, say [1,2,3].head(3)
evalable6 (1 2 3)
[Coke] wonders why rotor takes a pair to indicate the special arg, and not a named parameter 18:04
Kaiepi what does the pair in rotor do?
the way you used it
[Coke] If the element of @cycle is a Pair instead, the key of the pair specifies the length of the return sublist, and the value the gap between sublists; negative gaps produce overlap: 18:06
(from the docs)
Kaiepi ah 18:07
[Coke] ah, because you can have multiple pair rules also, duh. 18:09
m: say (^100).rotor: 3=>-1, 2=>2
camelia ((0 1 2) (2 3) (6 7 8) (8 9) (12 13 14) (14 15) (18 19 20) (20 21) (24 25 26) (26 27) (30 31 32) (32 33) (36 37 38) (38 39) (42 43 44) (44 45) (48 49 50) (50 51) (54 55 56) (56 57) (60 61 62) (62 63) (66 67 68) (68 69) (72 73 74) (74 75) (78 79 80) (8…
18:10 tobs` left
Xliff m: say (1..*).rotor(3=>-1).flat[^12] 18:11
camelia (1 2 3 3 4 5 5 6 7 7 8 9)
Xliff Now why couldn't I come up with that?!
m: say (1..*).rotor(3=>-2).flat[^12]
camelia (1 2 3 2 3 4 3 4 5 4 5 6)
[Coke] beacus rotor is weird. 18:12
Xliff LOL!
18:13 ExtraCrispy left 18:16 Kaiepi left
[Coke] *because 18:18
Xliff I got it.
18:19 zakharyas joined, vrurg left 18:20 p6bannerbot sets mode: +v zakharyas
simple what's the perl6 equivalent of perl5's " $var =~ /(.begin .* .end)/smg " ? 18:20
i tried m:smg
but that adverb inot allowed on m
18:22 Kaiepi joined, p6bannerbot sets mode: +v Kaiepi 18:25 rindolf left, sauvin left
[Coke] . already matches \n, so you don't need s 18:25
simple ah
[Coke] you're not using ^ or $, so you don't need m even on the p5 side 18:26
simple well, same error, just one less letter now
oh, true
that's all better then =) thanks 18:27
[Coke] m: say "i have a hat" ~~ m:g/ 'ha' /
camelia (｢ha｣ ｢ha｣)
[Coke] :)
18:37 lucasb left
Xliff Is this Rakudo's way of telling me I have an endless loop? -- "Failed to register: Timeout was reached" 18:42
If there's an endless look, it's in compile then. Weird! 18:45
18:47 molaf left
[Coke] failed to register doesn't sound rakudo specific. 18:55
Xliff Actually, no it was GTK, weirdly enough. 19:02
What's the best way to get the types passed into a signature capture?
Ala "method something(|c)
m: sub cap(|c) { 19:03
camelia 5===SORRY!5=== Error while compiling <tmp>
Missing block
at <tmp>:1
------> 3sub cap(|c) {7⏏5<EOL>
Xliff m: sub cap(|c) { say c.^name }
camelia ( no output )
Xliff m: sub cap(|c) { say c.^name }; cap(1, 2, 'A')
camelia Capture
Xliff m: sub cap(|c) { say c.signature.^name }; cap(1, 2, 'A')
camelia No such method 'signature' for invocant of type 'Capture'
in sub cap at <tmp> line 1
in block <unit> at <tmp> line 1
Xliff m: sub cap(|c) { .^name say for c }; cap(1, 2, 'A') 19:04
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3sub cap(|c) { .^name7⏏5 say for c }; cap(1, 2, 'A')
expecting any of:
infix
infix stopper
statement end
sta…
Xliff m: sub cap(|c) { .^name.say for c }; cap(1, 2, 'A') 19:05
camelia Int
Int
Str
SmokeMachine Xliff: its working!!! \o/ www.irccloud.com/pastebin/FdZRxIbF/ 19:11
Xliff: getting home, Ill make a nice oo frontend to it! 19:12
19:15 zacts left
Xliff SmokeMachine: :-O \o/ 19:20
NICE!
SmokeMachine needs to go back to Red... 19:24
SmokeMachine is loosing to much time playing with bernalang... 19:25
19:26 guifa joined, p6bannerbot sets mode: +v guifa
guifa When calling a grammar with an action, the grammar can refer directly to its action class with .actions but is there any built-in way to do the reverse (reference the grammar in the action) 19:28
masak no 19:29
what is it you want to do, more broadly?
SmokeMachine can you, from the Grammar, do something like `self.actions.grammar = self`? 19:30
guifa Just trying to better separate grammar from actions. I'm writing a pretty extensive/customizable brainfuck interpreter using grammar/actions, the main stumbling block I've had is handling the loops. Their inner code has to parsed separate from regular code to avoid early execution, but then sent back to get parsed as runnable code, which is easily done with the actions code is embedded in the grammar, but no really elegant way to do 19:34
masak hm 19:36
what you do sounds like a bit of a grammar code smell, if you don't mind me saying
notably, I'm pretty sure it should be possible to parse everything with one pass, and then not again 19:37
that should be your goal, too ;)
guifa Single pass would be optimal haha, but I'm a bit stumped at doing that
I mean, obviously there are more efficient ways to process BF, but I wanted to try to learn Grammar/Actions and it's actually be pretty useful as a exercise in learning 19:38
SmokeMachine guifa: I've been doing something like it... not with brainfuck, but with bernalang... github.com/FCO/bernalang 19:41
guifa: here is my grammar: github.com/FCO/bernalang/blob/mast...rammar.pm6 19:42
Xliff: Im thinking of doing a machine-code bernalang^^ compiler using gccjit... :) 19:43
19:44 zakharyas left
guifa I'm glancing it over really fast but it looks like you're parsing/compiling and then doing a separate execution phase yeah? 19:46
masak oh, you're not? then I understand your predicament better 19:47
so your actions are basically an interpreter?
guifa Indeed. BF is sufficiently simple enough that you can do that. When it parses a +, it increments the value at the point as the action right then and there.
19:49 cfa left
masak I think you'll find the order the actions fire is a bad fit for your interpreter idea 19:49
with the example of a +, at the point the action sees it, you don't know if that + is inside a loop (and should be deferred) or not inside one (and shouldn't) 19:50
because actions fire bottom-up, and loops (kinda) are more top-down :)
guifa Yeah. That was something that took me a bit to work around, but I did get it all working.
masak heh :)
don't really know what to say except it seems you're working against the grain 19:51
guifa Probably :-)
masak I think you'll find it's a lot easier to write a brainfuck *compiler* than a brainfuck *interpreter* using grammar/actions
guifa Possibly. I might do a later modification for that. My real goal wasn't so much learning to write compilers, but just getting a good feel for grammar rules that adequately catch stuff, and then attaching some actions. I had tortured myself way back when making an Applescript Brainfuck interpreter so figured what the hell ;-) 19:53
but as is typical for me, once I start, I like to see how far I can take it
masak moritz++ might amend what I'm about to say here, but... 19:54
...I find actions work best when you exploit the fact that they're bottom-up, rather than try to fight it
19:54 sena_kun left
masak it's one of those things, I realize that nowadays I've started to *think* that way, and I try to do stuff bottom-up even outside of actions sometimes 19:54
19:56 xinming_ joined, p6bannerbot sets mode: +v xinming_ 19:58 abraxxa1 joined
moritz I agree 19:58
19:59 xinming left, p6bannerbot sets mode: +v abraxxa1
masak second more advanced rule (for free, even though nobody asked): 20:00
20:00 El_Che joined
masak never *ever* fire a side effect before you know you're in the right rule 20:00
guifa Yeah, the only way to handle (AFAICT) a loop is to do a separate inner code rule from the main rule that doesn't have actions attached, and then inside the action call .parse(:rule<maincode>, :actions(self.actions) repeatedly. For anything more complex than BF I don't think an interpreter inside of the grammar itself is really feasible just because you'd have to include both executing and not executing rules for all code
20:00 p6bannerbot sets mode: +v El_Che
masak also, you're multiple-parsing, which is maybe fine for an interpreter, but just generally problematic :) 20:01
how do you handle missing `]` ?
guifa At the moment, I don't. The joys of BF is there's no single specification (which did let me have fun with some adverbs) 20:02
:8bit :16bit :32bit for instance is a cool way to take advantage of the adverbs 20:03
20:05 rindolf joined 20:06 p6bannerbot sets mode: +v rindolf
masak yes, that reads quite well :) 20:06
guifa I originally included ? as a built in dump command, but then I figured it'd be cooler as a grammar extension. I get that the not-best-practices in handling the loop but… I think I can make a pretty cool tutorial with it to show off some of the nicer features of p6 (while including a warning that it's probably best not to make an interpreter out of grammars in the real world ;-) ) 20:09
20:09 El_Che left 20:11 domidumont left, El_Che joined 20:12 p6bannerbot sets mode: +v El_Che 20:16 El_Che left 20:18 El_Che joined 20:19 p6bannerbot sets mode: +v El_Che 20:23 jmerelo left 20:32 guifa left 20:34 MorayJ left
scovit Hello, can anybody here help me to make an object of type Signature? Not a literal 20:38
moritz I can try :-) 20:39
what's your question? 20:40
masak m: say Signature.new
camelia ()
masak ...do I win an internet? :P
moritz m: say :().^name 20:41
camelia Signature
21:12 jast left 21:18 rindolf left, patrickb left 21:22 sena_kun joined
Xliff m: use MONKEY-SEE-NO-EVAL; my $m = 42; EVAL 'say $m'; 21:22
camelia 42
21:22 p6bannerbot sets mode: +v sena_kun 21:26 molaf joined 21:27 p6bannerbot sets mode: +v molaf 21:28 abraxxa1 left 21:29 Kaiepi left 21:32 lizmat left
sena_kun can someone point me to how names declared as enum values are promoted to types? I am reading github.com/rakudo/rakudo/blob/mast....nqp#L4851 but no much luck. When I am constructing an enum using MOP with base_type == Int and with a simple `add_enum_value($str-int-pair))`, is type created? If yes, how can I get it to set it a name and export appropriately? If not, how can I create one? 21:40
As in...
m: enum Normal <One Two>; say Two;
camelia Two 21:41
sena_kun but...
m: my $new-enum = Metamodel::EnumHOW.new_type(name => 'custom', base_type => Int); $new-enum.^add_enum_value((Two => 2)); $new-enum.^add_role(Enumeration); $new-enum.^add_role(NumericEnumeration); $new-enum.^compose; say $new-enum.^enum_values;
camelia {Two => 2}
sena_kun `Two` here is just a Str key, not a type object that can be assigned to e.g. `my Custom $aa`. 21:43
looking at how `add_enum_value` implementation simply binds pair value to an internal storage, it seems that I have to create each "type object" manually. 21:45
21:48 zachk joined
simple m: (5802..5830).map( {.chr} )[*-10..*-5] 21:48
camelia ( no output )
21:48 p6bannerbot sets mode: +v zachk
simple =( 21:49
scovit sorry moritz, I explain 21:50
I have a slurpy argument 21:51
I want to use the types passed in this slurpy
to setup a signature that corresponds to it
21:51 zachk left 21:52 zachk joined, cherryh.freenode.net sets mode: +v zachk, p6bannerbot sets mode: +v zachk
scovit Signature.new(params => (do for @args { $_.WHAT }), returns => Str) 21:52
something like this
but probably needs nqp
i need this signature to pass it to NativeCall 21:53
21:53 Kaiepi joined
scovit I do not care about performances now (I plan to add caching) 21:53
21:54 p6bannerbot sets mode: +v Kaiepi, Cabanossi left 21:56 Cabanossi joined, p6bannerbot sets mode: +v Cabanossi
Kaiepi is it possible to hotpatch a module? 21:56
21:58 tadzik joined 21:59 p6bannerbot sets mode: +v tadzik 22:01 rindolf joined 22:02 p6bannerbot sets mode: +v rindolf 22:09 Kaypie joined, Kaiepi left 22:10 p6bannerbot sets mode: +v Kaypie 22:11 melezhik joined, p6bannerbot sets mode: +v melezhik
Xliff Does $a infix:<eqv> $b do the equivalent of $a.eqv($b) ?? 22:12
sena_kun m: 1.eqv(5) 22:15
camelia No such method 'eqv' for invocant of type 'Int'
in block <unit> at <tmp> line 1
Xliff m: 1 eqv 5
camelia WARNINGS for <tmp>:
Useless use of "eqv" in expression "1 eqv 5" in sink context (line 1)
sena_kun Xliff, if $a type has `eqv` method implemented using infix op, maybe?
Xliff m: role Comp { method eqv(Cool $b) { $b.eqv(self) } }; my $a = 1 but Comp; my $b = 2 but comp; say $a eqv $b 22:16
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
comp used at line 1. Did you mean 'chomp', 'comb'?
Xliff m: role Comp { method eqv(Cool $b) { $b.eqv(self) } }; my $a = 1 but Comp; my $b = 2 but Comp; say $a eqv $b
camelia False
Xliff m: role Comp { method eqv(Cool $b) { $b.eqv(self) } }; my $a = 1 but Comp; my $b = 1 but Comp; say $a eqv $b 22:17
camelia True
Xliff m: my ($a, $b) = (1, 1); $a eqv $b; 22:18
camelia WARNINGS for <tmp>:
Useless use of "eqv" in expression "$a eqv $b" in sink context (line 1)
Xliff m: my ($a, $b) = (1, 1); say $a eqv $b;
camelia True
Xliff m: say 1 eqv 5
camelia False
Xliff OK. So not as elegant as I was expecting, but...
m: my $gtext = 'grammar Phil'; my $name = ($gtext ~~ /^^ \s* 'grammar' \s+ (\w+) // [])[0]; say $name 22:22
camelia Cannot resolve caller Numeric(Regex:D: ); none of these signatures match:
(Mu:U \v: *%_)
in block <unit> at <tmp> line 1
Xliff m: my $gtext = 'grammar Phil'; my $name = ($gtext ~~ /^ \s* 'grammar' \s+ (\w+) // [])[0]; say $name
camelia Cannot resolve caller Numeric(Regex:D: ); none of these signatures match:
(Mu:U \v: *%_)
in block <unit> at <tmp> line 1
Xliff m: my $gtext = 'grammar Phil'; my $name = ($gtext ~~ /^ \s* 'grammar' \s+ (\w+)/ // [])[0]; say $name
camelia ｢Phil｣
Xliff m: my $gtext = 'grammar Phil'; my $name = ($gtext ~~ /^ \s* 'grammar' \s+ (\w+)/ // [])[0].Str; say $name
camelia Phil
Xliff m: my $gtext = 'rammar Phil'; my $name = ($gtext ~~ /^ \s* 'grammar' \s+ (\w+)/ // [])[0].Str; say $name 22:23
camelia Use of uninitialized value of type Any in string context.
Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful.

in block <unit> at <tmp> line 1
Xliff m: my $gtext = 'grammar Phil'; my $name = ($gtext ~~ /^^ \s* 'grammar' \s+ (\w+)/ // [])[0].Str; say $name 22:25
camelia Phil
22:33 sena_kun left, sena_kun joined 22:34 p6bannerbot sets mode: +v sena_kun 22:36 rindolf left, rindolf joined 22:37 p6bannerbot sets mode: +v rindolf
rindolf hi all 22:38
sena_kun o/
rindolf how was the squashathon?
sena_kun testneeded tickets... Some were closed, I think. :) 22:40
AlexDaniel not very active actually 22:41
there were many easy tickets but people didn't feel like resolving any of them
maybe every month is too often, I dunno :)
sena_kun any ideas on how to imitate nqp::rebless op from high level code? I really don't want to do `use nqp` and copy-paste what rakudo does. 22:42
Kaypie i'm surprised i was considered to have made the most/best prs during the squashathon in order to win the camelia plushie 22:56
i only made one pullreq and it was like 5 lines
22:56 molaf left
sena_kun next one may be more popular. :) 23:02
23:03 Ven`` joined 23:04 p6bannerbot sets mode: +v Ven``
scovit I do not care about performances now (I plan to add caching) 23:06
sorry
jnthn sena_kun: nqp::rebless is used by .^mixin, so that's probably what you want 23:15
(This is in turn what's used to implement does and but)
(Which is also what would normally be used) 23:16
sena_kun jnthn, so if I want to turn `1` into my created enum type, it'd be like... hmm. `but` gives me `Int+{foo-enum}`, which fails a check against plain `my custom $a` type. 23:21
I have found github.com/rakudo/rakudo/blob/mast....nqp#L3900 but not yet sure how to recreate it without nqp. :) 23:22
jnthn sena_kun: You don't want to do that :) 23:23
m: enum A <B C D>; say A(1)
camelia C
jnthn Just use the coercion form like that
sena_kun jnthn, well, I have to create enum without using explicit `enum Foo <>`. And I hope to do it without EVAL. This is what I "want" to do. :) 23:25
jnthn Ah 23:26
sena_kun or, rather, I can create it, but values.
that's the issue.
as in, well...
I can create a type, but `add_enum_value` MOP call just does a binding, so I cannot just bind some Int as a value, it'll break typing. 23:27
jnthn Can you do $the-enum-type-object.new($the-integer) ?
sena_kun so I looked at how rakudo creates stuff and found out that e.g. Int value can be reblessed to newly created enum type(not even composed yed as I get it), but... Nope, I can't.
my $new-enum = Metamodel::EnumHOW.new_type(name => 'custom', base_type => Int); $new-enum.new(1).say; 23:28
m: my $new-enum = Metamodel::EnumHOW.new_type(name => 'custom', base_type => Int); $new-enum.new(1).say;
camelia P6opaque: must compose custom before allocating
in block <unit> at <tmp> line 1
sena_kun not how it would work.
jnthn Try $new-enum.^compose before the .new call
Or at least .^compose_repr
sena_kun jnthn, will it be possible to still `^add_enum_value` after compose? That's why I avoided it... Let me try... 23:29
jnthn I think yes, and if there's issues you can perhaps always re-compose
I wonder if we should provide some nice MOP method to save people figuring this all out...
sena_kun woooah 23:30
^compose_repr
ugh
sorry. *compose_repr has helped. \o/
jnthn yay ;)
*:)
sena_kun m: my $new-enum = Metamodel::EnumHOW.new_type(name => 'custom', base_type => Int); $new-enum.^compose_repr; my \Two = $new-enum.new(2); say Two;
camelia 2
sena_kun m: my $new-enum = Metamodel::EnumHOW.new_type(name => 'custom', base_type => Int); $new-enum.^compose_repr; my \Two = $new-enum.new(2); say Two.^name;
camelia custom
sena_kun one less issue towards the goal. \o/ 23:31
jnthn, thanks a lot!
jnthn Welcome :) 23:32
scovit m: gist.github.com/scovit/d328be9817e...98f99ee3a3 23:33
camelia Signature $sign = :(Int $roof, Str $banter --> Str)
scovit no idea what I did here! 23:34
but it might wojust some cargo cult code, it might work
23:40 rindolf left 23:46 melezhik left 23:53 vrurg joined 23:54 p6bannerbot sets mode: +v vrurg
scovit m: gist.github.com/scovit/d328be9817e...98f99ee3a3 23:55
camelia (signal SEGV)Signature $sign = :(Str $text --> int32)
scovit is working on my mac 23:56
:( 23:58
ok, works on mac 23:59
dies on linux

Powered by Cro and the Raku® Programming Language.

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

Thank you!