6macros: discussing the finer points of Perl 6 macros, Qtrees, and how to stay sane | irclog: irclog.perlgeek.de/6macros/today
Set by moderator on 28 July 2015.
04:42 vendethiel joined 06:10 FROGGS joined 06:38 Ven joined 08:14 Ven joined
Ven www.worldcrunch.com/culture-society.../c3s17017/ 10:04
that's a bit insane, huh.
masak I know a guy who is at least partway towards that. 10:07
I felt a little bit better when he confessed he hadn't gotten very far with Mandarin, 'cus it was too different from the PIE-based languages he was used to :) 10:08
10:46 Ven joined 12:07 Ven joined 12:13 Ven joined 13:13 Ven joined
Ven noticed ↑'s monadic version doesn't work in GNU APL :( 13:43
13:51 Ven joined 15:16 Ven joined 15:24 Ven joined 16:04 Ven joined 17:42 FROGGS joined 18:25 vendethiel joined
vendethiel oh masak 19:02
btw
for "dep types" 19:03
types need to stop being "an optimization" ;-)
masak please elaborate 19:04
I *think* I see what you mean, but...
vendethiel disable the optimizer
they're not ran anymore, AFAIK
masak aaaah
right, now it's all mixed together
vendethiel and the synopsis don't actually mandate that they're checked too, right?
masak inextricably
I think the synopses require that they be checked
vendethiel do they? 19:05
masak yes, I think so
vendethiel at least
the fact that they get *statically* checked is definitely optional.
masak basically, if you opt in and use them... they're checked
vendethiel corrected myself
masak ok
yes
vendethiel also
I'd looove to get some static OO thing. I guess there's still "final", though 19:06
masak I'm doing a lot of useful thinking right now in 007. I guess you can tell by the notification fallout :) 19:15
vendethiel you think we should mix syntax and visitor macros? 19:19
how far are you along in 50 years of lisp? :-)
masak not far enough 19:20
is there something I should pay immediate attention to?
vendethiel because "visitor macros inside of a syntax macro" is basically macrolet
masak in this case, I'd make syntax macros for `amb` and `assert`, and a visitor macro for Q::Statement::Block 19:21
vendethiel well ok, say
transaction macro
masak so it'd actually be "syntax macro(s) inside of a visitor macro"
vendethiel you want "commit" and "rollback" only to be "macros" inside of this block
masak right, ok.
yes, that's still an open question for me. 19:22
vendethiel we talked about this many times
masak aye
during YAPC::Europe, I toyed with the idea of "the macros you define *inside* of a syntax macro are valid inside the thing it parses"
but it feels slightly too magical, even if it'd be easy-to-use 19:23
vendethiel yeaaah
well
again, macrolet could make more sense
masak ok, I have some reading to do, then
in general, I'm very pleased with how the three-types-of-macro gist came together.
but I'm still willing to be impressed and inspired by Let Over Lambda 19:24
vendethiel sometimes, perl6 tries to cram a few too many things in one place...
like :36(..). why'd you need to overload :() "again"...
masak :)
vendethiel that's what gives it this "line noise" name, methinks 19:25
say, rosettacode.org/wiki/IBAN#Perl_6 that I was reading with a friend
masak looks
vendethiel it's pretty readable, if you know "leading dot is $_." and regexps
(which isn't too much to ask)
but :36(~$_) is... heh..
masak I guess you'd prefer to have it as a simple function? 19:26
vendethiel *too cute*
yeah. $_.base(36)
amazing
masak nonono
vendethiel doesn't even need to convert anything to a string to then base it
masak you're making the classical mistake with these base converters 19:27
wrong direction :)
vendethiel yeah, well
masak :36(~$_) is converting *from* base 36
vendethiel see
that's the big issue
masak $_.base(36) already exists and converts *to* base 36
this was the big thing that the :35(~$_) syntax basically solved
vendethiel well
$_.as-base(36) 19:28
masak and if you don't appreciate that, I'm not sure I can respect any solution that you propose ;)
vendethiel amazing.
no, I don't
masak you still don't get it, I see
vendethiel I especially don't like that converting *to* a base is a method, but *from* isn't
masak meh
that actually helps, in my opinion
vendethiel and how is converting "from" a base something as common that it'd require that kind of huffmanization? 19:29
if I don't get it, please explain
masak some bases are more common than others
:2, :8, :16, specifically
vendethiel but converting between bases isn't really common 19:30
I know I havn't done it in several *years
*
masak the problem with Perl 5's bin(), oct(), hex() was that people could never remember if they were *from* or *to* functions
vendethiel just add to- or from- function.
masak I would say that from a language design perspective, it's really not as simple as that
vendethiel I'm not saying all languages "fixed" that. ruby's Integer.to_s(base) is also ugh
masak and that's what this whole :36() thing is about, in the end 19:31
vendethiel maybe I just never stumbled in fields that needed this kind of "power" :)
I disagree *strongly* with you
because I can't remember the way it's going with :36() either
masak though it's not essential to my main point, I'll note that I have no such problem remembering what direction :36() goes 19:33
to me it pretty clearly reads as "what comes here is in base 36 (and hence, it's a string)"
vendethiel well, you can't justify "it's easy to remember" with "I remember it" :P 19:34
masak which is why I didn't add it as support for my main point ;)
vendethiel fair enough
masak by the way, www.cs.cmu.edu/~fp/courses/15317-f0...edule.html is pretty great 19:35
I'm on 04*.pdf now
vendethiel I'm probably never going to find time for that
masak ok, I will endeavour to send less of these :) 19:36
vendethiel :( 19:37
they're a bit too big
I have many of 'em piled up...
masak fwiw, you're not the only one struggling to find time for reading nice things... :)
vendethiel *g*
I appreciate every other link though 19:38
masak .oO( so, about 50% of them ) :P
idioms.thefreedictionary.com/every+other 19:39
English is weird.
vendethiel I know, I know
that's somewhat what I meant, too
masak ok
vendethiel well, not the 50% part, but the "it's a long list of youtube videos of 4 hrs
" or tens of PDFs
usually PDFs are fine though 19:40
it's mostly "class videos"
I really can't
masak well, so far, www.cs.cmu.edu/~fp/courses/15317-f0...natded.pdf has been the best one
vendethiel I already have a playlist with ~100 programming-related videos :/
masak so if you have time to read one, choose that one
vendethiel thanks! 19:41
masak it's 12 pages of easily-digestible material
also might be good to know that I was recommended this course by that list of links that you sent me on type theory
vendethiel ...I admit I havn't gotten around to that link yet 19:42
masak ;)
vendethiel but since GNU APL isn't that feature-complete, I guess I'll go back to something else
maybe Ada.
(I do have a dyalog APL license, but it doesn't seem to be usable for "scripts")
then I'll go back to functional shenanigans 19:44
and I have a very big backlog
masak relax. you're young. you've got plenty of time. 19:45
just focus on things you enjoy. :)
vendethiel do I look like I don't thoroughly enjoy myself? :-)
.oO( don't try to make me slow down so that other younglings can catch up! *g* )
masak :P 19:46
masak attempts to introduce Q::Compunit in 007 19:47
vendethiel yay!
I might take a stab at object literals
(famous last words...) 19:48
masak cool
patches welcome
vendethiel ..famous first words :) 19:58
masak I'd be honored to get a patch on that 20:01
expect some comments, though
even if I accept it, I might have some suggestions
vendethiel oh, a Martin-Löf quote :-) 20:31
masak yeah, he seems to be mentioned a lot in dep-typed contexts 20:33
vendethiel I wonder why.. *g*
masak can't think of a good reason :P
vendethiel soo many references though... 20:34
sigh. my backlog is too big :(
masak I skimmed the references, and I don't even feel bad about it
vendethiel I never check them 20:37
masak well, that's all right then
vendethiel huh,I just realized that in some alternative perluniverse, maybe Perl6 got abandoned along the way 20:45
"the jnthn never came" or something
masak aye 20:46
quite likely, actually
vendethiel yeah. bus factor of one isn't that high :-)
even if it's almost 2 *g*
masak it is? 20:47
vendethiel considering the tons of amazing work larry, nine, timo, liz and froggs have been putting out, I'd certainly consider it so
no, it's definitely higher than 2 no
now*
masak yeah
though jnthn is still pretty indispensible in various ways 20:48
vendethiel definitely.
also, brrt++ :-)
masak aye 20:49

Powered by Cro and the Raku® Programming Language.

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

Thank you!